Show More
@@ -1,338 +1,338 b'' | |||||
1 | """Implementations for various useful completers. |
|
1 | """Implementations for various useful completers. | |
2 |
|
2 | |||
3 | These are all loaded by default by IPython. |
|
3 | These are all loaded by default by IPython. | |
4 | """ |
|
4 | """ | |
5 | #----------------------------------------------------------------------------- |
|
5 | #----------------------------------------------------------------------------- | |
6 | # Copyright (C) 2010-2011 The IPython Development Team. |
|
6 | # Copyright (C) 2010-2011 The IPython Development Team. | |
7 | # |
|
7 | # | |
8 | # Distributed under the terms of the BSD License. |
|
8 | # Distributed under the terms of the BSD License. | |
9 | # |
|
9 | # | |
10 | # The full license is in the file COPYING.txt, distributed with this software. |
|
10 | # The full license is in the file COPYING.txt, distributed with this software. | |
11 | #----------------------------------------------------------------------------- |
|
11 | #----------------------------------------------------------------------------- | |
12 |
|
12 | |||
13 | #----------------------------------------------------------------------------- |
|
13 | #----------------------------------------------------------------------------- | |
14 | # Imports |
|
14 | # Imports | |
15 | #----------------------------------------------------------------------------- |
|
15 | #----------------------------------------------------------------------------- | |
16 | from __future__ import print_function |
|
16 | from __future__ import print_function | |
17 |
|
17 | |||
18 | # Stdlib imports |
|
18 | # Stdlib imports | |
19 | import glob |
|
19 | import glob | |
20 | import imp |
|
20 | import imp | |
21 | import inspect |
|
21 | import inspect | |
22 | import os |
|
22 | import os | |
23 | import re |
|
23 | import re | |
24 | import sys |
|
24 | import sys | |
25 |
|
25 | |||
26 | # Third-party imports |
|
26 | # Third-party imports | |
27 | from time import time |
|
27 | from time import time | |
28 | from zipimport import zipimporter |
|
28 | from zipimport import zipimporter | |
29 |
|
29 | |||
30 | # Our own imports |
|
30 | # Our own imports | |
31 | from IPython.core.completer import expand_user, compress_user |
|
31 | from IPython.core.completer import expand_user, compress_user | |
32 | from IPython.core.error import TryNext |
|
32 | from IPython.core.error import TryNext | |
33 | from IPython.utils._process_common import arg_split |
|
33 | from IPython.utils._process_common import arg_split | |
34 | from IPython.utils.py3compat import string_types |
|
34 | from IPython.utils.py3compat import string_types | |
35 |
|
35 | |||
36 | # FIXME: this should be pulled in with the right call via the component system |
|
36 | # FIXME: this should be pulled in with the right call via the component system | |
37 | from IPython import get_ipython |
|
37 | from IPython import get_ipython | |
38 |
|
38 | |||
39 | #----------------------------------------------------------------------------- |
|
39 | #----------------------------------------------------------------------------- | |
40 | # Globals and constants |
|
40 | # Globals and constants | |
41 | #----------------------------------------------------------------------------- |
|
41 | #----------------------------------------------------------------------------- | |
42 |
|
42 | |||
43 | # Time in seconds after which the rootmodules will be stored permanently in the |
|
43 | # Time in seconds after which the rootmodules will be stored permanently in the | |
44 | # ipython ip.db database (kept in the user's .ipython dir). |
|
44 | # ipython ip.db database (kept in the user's .ipython dir). | |
45 | TIMEOUT_STORAGE = 2 |
|
45 | TIMEOUT_STORAGE = 2 | |
46 |
|
46 | |||
47 | # Time in seconds after which we give up |
|
47 | # Time in seconds after which we give up | |
48 | TIMEOUT_GIVEUP = 20 |
|
48 | TIMEOUT_GIVEUP = 20 | |
49 |
|
49 | |||
50 | # Regular expression for the python import statement |
|
50 | # Regular expression for the python import statement | |
51 | import_re = re.compile(r'(?P<name>[a-zA-Z_][a-zA-Z0-9_]*?)' |
|
51 | import_re = re.compile(r'(?P<name>[a-zA-Z_][a-zA-Z0-9_]*?)' | |
52 | r'(?P<package>[/\\]__init__)?' |
|
52 | r'(?P<package>[/\\]__init__)?' | |
53 | r'(?P<suffix>%s)$' % |
|
53 | r'(?P<suffix>%s)$' % | |
54 | r'|'.join(re.escape(s[0]) for s in imp.get_suffixes())) |
|
54 | r'|'.join(re.escape(s[0]) for s in imp.get_suffixes())) | |
55 |
|
55 | |||
56 | # RE for the ipython %run command (python + ipython scripts) |
|
56 | # RE for the ipython %run command (python + ipython scripts) | |
57 | magic_run_re = re.compile(r'.*(\.ipy|\.py[w]?)$') |
|
57 | magic_run_re = re.compile(r'.*(\.ipy|\.py[w]?)$') | |
58 |
|
58 | |||
59 | #----------------------------------------------------------------------------- |
|
59 | #----------------------------------------------------------------------------- | |
60 | # Local utilities |
|
60 | # Local utilities | |
61 | #----------------------------------------------------------------------------- |
|
61 | #----------------------------------------------------------------------------- | |
62 |
|
62 | |||
63 | def module_list(path): |
|
63 | def module_list(path): | |
64 | """ |
|
64 | """ | |
65 | Return the list containing the names of the modules available in the given |
|
65 | Return the list containing the names of the modules available in the given | |
66 | folder. |
|
66 | folder. | |
67 | """ |
|
67 | """ | |
68 | # sys.path has the cwd as an empty string, but isdir/listdir need it as '.' |
|
68 | # sys.path has the cwd as an empty string, but isdir/listdir need it as '.' | |
69 | if path == '': |
|
69 | if path == '': | |
70 | path = '.' |
|
70 | path = '.' | |
71 |
|
71 | |||
72 | # A few local constants to be used in loops below |
|
72 | # A few local constants to be used in loops below | |
73 | pjoin = os.path.join |
|
73 | pjoin = os.path.join | |
74 |
|
74 | |||
75 | if os.path.isdir(path): |
|
75 | if os.path.isdir(path): | |
76 | # Build a list of all files in the directory and all files |
|
76 | # Build a list of all files in the directory and all files | |
77 | # in its subdirectories. For performance reasons, do not |
|
77 | # in its subdirectories. For performance reasons, do not | |
78 | # recurse more than one level into subdirectories. |
|
78 | # recurse more than one level into subdirectories. | |
79 | files = [] |
|
79 | files = [] | |
80 | for root, dirs, nondirs in os.walk(path): |
|
80 | for root, dirs, nondirs in os.walk(path): | |
81 | subdir = root[len(path)+1:] |
|
81 | subdir = root[len(path)+1:] | |
82 | if subdir: |
|
82 | if subdir: | |
83 | files.extend(pjoin(subdir, f) for f in nondirs) |
|
83 | files.extend(pjoin(subdir, f) for f in nondirs) | |
84 | dirs[:] = [] # Do not recurse into additional subdirectories. |
|
84 | dirs[:] = [] # Do not recurse into additional subdirectories. | |
85 | else: |
|
85 | else: | |
86 | files.extend(nondirs) |
|
86 | files.extend(nondirs) | |
87 |
|
87 | |||
88 | else: |
|
88 | else: | |
89 | try: |
|
89 | try: | |
90 | files = list(zipimporter(path)._files.keys()) |
|
90 | files = list(zipimporter(path)._files.keys()) | |
91 | except: |
|
91 | except: | |
92 | files = [] |
|
92 | files = [] | |
93 |
|
93 | |||
94 | # Build a list of modules which match the import_re regex. |
|
94 | # Build a list of modules which match the import_re regex. | |
95 | modules = [] |
|
95 | modules = [] | |
96 | for f in files: |
|
96 | for f in files: | |
97 | m = import_re.match(f) |
|
97 | m = import_re.match(f) | |
98 | if m: |
|
98 | if m: | |
99 | modules.append(m.group('name')) |
|
99 | modules.append(m.group('name')) | |
100 | return list(set(modules)) |
|
100 | return list(set(modules)) | |
101 |
|
101 | |||
102 |
|
102 | |||
103 | def get_root_modules(): |
|
103 | def get_root_modules(): | |
104 | """ |
|
104 | """ | |
105 | Returns a list containing the names of all the modules available in the |
|
105 | Returns a list containing the names of all the modules available in the | |
106 | folders of the pythonpath. |
|
106 | folders of the pythonpath. | |
107 |
|
107 | |||
108 | ip.db['rootmodules_cache'] maps sys.path entries to list of modules. |
|
108 | ip.db['rootmodules_cache'] maps sys.path entries to list of modules. | |
109 | """ |
|
109 | """ | |
110 | ip = get_ipython() |
|
110 | ip = get_ipython() | |
111 | rootmodules_cache = ip.db.get('rootmodules_cache', {}) |
|
111 | rootmodules_cache = ip.db.get('rootmodules_cache', {}) | |
112 | rootmodules = list(sys.builtin_module_names) |
|
112 | rootmodules = list(sys.builtin_module_names) | |
113 | start_time = time() |
|
113 | start_time = time() | |
114 | store = False |
|
114 | store = False | |
115 | for path in sys.path: |
|
115 | for path in sys.path: | |
116 | try: |
|
116 | try: | |
117 | modules = rootmodules_cache[path] |
|
117 | modules = rootmodules_cache[path] | |
118 | except KeyError: |
|
118 | except KeyError: | |
119 | modules = module_list(path) |
|
119 | modules = module_list(path) | |
120 | try: |
|
120 | try: | |
121 | modules.remove('__init__') |
|
121 | modules.remove('__init__') | |
122 | except ValueError: |
|
122 | except ValueError: | |
123 | pass |
|
123 | pass | |
124 | if path not in ('', '.'): # cwd modules should not be cached |
|
124 | if path not in ('', '.'): # cwd modules should not be cached | |
125 | rootmodules_cache[path] = modules |
|
125 | rootmodules_cache[path] = modules | |
126 | if time() - start_time > TIMEOUT_STORAGE and not store: |
|
126 | if time() - start_time > TIMEOUT_STORAGE and not store: | |
127 | store = True |
|
127 | store = True | |
128 | print("\nCaching the list of root modules, please wait!") |
|
128 | print("\nCaching the list of root modules, please wait!") | |
129 | print("(This will only be done once - type '%rehashx' to " |
|
129 | print("(This will only be done once - type '%rehashx' to " | |
130 | "reset cache!)\n") |
|
130 | "reset cache!)\n") | |
131 | sys.stdout.flush() |
|
131 | sys.stdout.flush() | |
132 | if time() - start_time > TIMEOUT_GIVEUP: |
|
132 | if time() - start_time > TIMEOUT_GIVEUP: | |
133 | print("This is taking too long, we give up.\n") |
|
133 | print("This is taking too long, we give up.\n") | |
134 | return [] |
|
134 | return [] | |
135 | rootmodules.extend(modules) |
|
135 | rootmodules.extend(modules) | |
136 | if store: |
|
136 | if store: | |
137 | ip.db['rootmodules_cache'] = rootmodules_cache |
|
137 | ip.db['rootmodules_cache'] = rootmodules_cache | |
138 | rootmodules = list(set(rootmodules)) |
|
138 | rootmodules = list(set(rootmodules)) | |
139 | return rootmodules |
|
139 | return rootmodules | |
140 |
|
140 | |||
141 |
|
141 | |||
142 | def is_importable(module, attr, only_modules): |
|
142 | def is_importable(module, attr, only_modules): | |
143 | if only_modules: |
|
143 | if only_modules: | |
144 | return inspect.ismodule(getattr(module, attr)) |
|
144 | return inspect.ismodule(getattr(module, attr)) | |
145 | else: |
|
145 | else: | |
146 | return not(attr[:2] == '__' and attr[-2:] == '__') |
|
146 | return not(attr[:2] == '__' and attr[-2:] == '__') | |
147 |
|
147 | |||
148 |
|
148 | |||
149 | def try_import(mod, only_modules=False): |
|
149 | def try_import(mod, only_modules=False): | |
150 | try: |
|
150 | try: | |
151 | m = __import__(mod) |
|
151 | m = __import__(mod) | |
152 | except: |
|
152 | except: | |
153 | return [] |
|
153 | return [] | |
154 | mods = mod.split('.') |
|
154 | mods = mod.split('.') | |
155 | for module in mods[1:]: |
|
155 | for module in mods[1:]: | |
156 | m = getattr(m, module) |
|
156 | m = getattr(m, module) | |
157 |
|
157 | |||
158 | m_is_init = hasattr(m, '__file__') and '__init__' in m.__file__ |
|
158 | m_is_init = hasattr(m, '__file__') and '__init__' in m.__file__ | |
159 |
|
159 | |||
160 | completions = [] |
|
160 | completions = [] | |
161 | if (not hasattr(m, '__file__')) or (not only_modules) or m_is_init: |
|
161 | if (not hasattr(m, '__file__')) or (not only_modules) or m_is_init: | |
162 | completions.extend( [attr for attr in dir(m) if |
|
162 | completions.extend( [attr for attr in dir(m) if | |
163 | is_importable(m, attr, only_modules)]) |
|
163 | is_importable(m, attr, only_modules)]) | |
164 |
|
164 | |||
165 | completions.extend(getattr(m, '__all__', [])) |
|
165 | completions.extend(getattr(m, '__all__', [])) | |
166 | if m_is_init: |
|
166 | if m_is_init: | |
167 | completions.extend(module_list(os.path.dirname(m.__file__))) |
|
167 | completions.extend(module_list(os.path.dirname(m.__file__))) | |
168 | completions = set(completions) |
|
168 | completions = set(completions) | |
169 | if '__init__' in completions: |
|
169 | if '__init__' in completions: | |
170 | completions.remove('__init__') |
|
170 | completions.remove('__init__') | |
171 | return list(completions) |
|
171 | return list(completions) | |
172 |
|
172 | |||
173 |
|
173 | |||
174 | #----------------------------------------------------------------------------- |
|
174 | #----------------------------------------------------------------------------- | |
175 | # Completion-related functions. |
|
175 | # Completion-related functions. | |
176 | #----------------------------------------------------------------------------- |
|
176 | #----------------------------------------------------------------------------- | |
177 |
|
177 | |||
178 | def quick_completer(cmd, completions): |
|
178 | def quick_completer(cmd, completions): | |
179 | """ Easily create a trivial completer for a command. |
|
179 | """ Easily create a trivial completer for a command. | |
180 |
|
180 | |||
181 | Takes either a list of completions, or all completions in string (that will |
|
181 | Takes either a list of completions, or all completions in string (that will | |
182 | be split on whitespace). |
|
182 | be split on whitespace). | |
183 |
|
183 | |||
184 | Example:: |
|
184 | Example:: | |
185 |
|
185 | |||
186 | [d:\ipython]|1> import ipy_completers |
|
186 | [d:\ipython]|1> import ipy_completers | |
187 | [d:\ipython]|2> ipy_completers.quick_completer('foo', ['bar','baz']) |
|
187 | [d:\ipython]|2> ipy_completers.quick_completer('foo', ['bar','baz']) | |
188 | [d:\ipython]|3> foo b<TAB> |
|
188 | [d:\ipython]|3> foo b<TAB> | |
189 | bar baz |
|
189 | bar baz | |
190 | [d:\ipython]|3> foo ba |
|
190 | [d:\ipython]|3> foo ba | |
191 | """ |
|
191 | """ | |
192 |
|
192 | |||
193 | if isinstance(completions, string_types): |
|
193 | if isinstance(completions, string_types): | |
194 | completions = completions.split() |
|
194 | completions = completions.split() | |
195 |
|
195 | |||
196 | def do_complete(self, event): |
|
196 | def do_complete(self, event): | |
197 | return completions |
|
197 | return completions | |
198 |
|
198 | |||
199 | get_ipython().set_hook('complete_command',do_complete, str_key = cmd) |
|
199 | get_ipython().set_hook('complete_command',do_complete, str_key = cmd) | |
200 |
|
200 | |||
201 | def module_completion(line): |
|
201 | def module_completion(line): | |
202 | """ |
|
202 | """ | |
203 | Returns a list containing the completion possibilities for an import line. |
|
203 | Returns a list containing the completion possibilities for an import line. | |
204 |
|
204 | |||
205 | The line looks like this : |
|
205 | The line looks like this : | |
206 | 'import xml.d' |
|
206 | 'import xml.d' | |
207 | 'from xml.dom import' |
|
207 | 'from xml.dom import' | |
208 | """ |
|
208 | """ | |
209 |
|
209 | |||
210 | words = line.split(' ') |
|
210 | words = line.split(' ') | |
211 | nwords = len(words) |
|
211 | nwords = len(words) | |
212 |
|
212 | |||
213 | # from whatever <tab> -> 'import ' |
|
213 | # from whatever <tab> -> 'import ' | |
214 | if nwords == 3 and words[0] == 'from': |
|
214 | if nwords == 3 and words[0] == 'from': | |
215 | return ['import '] |
|
215 | return ['import '] | |
216 |
|
216 | |||
217 | # 'from xy<tab>' or 'import xy<tab>' |
|
217 | # 'from xy<tab>' or 'import xy<tab>' | |
218 | if nwords < 3 and (words[0] in ['import','from']) : |
|
218 | if nwords < 3 and (words[0] in ['import','from']) : | |
219 | if nwords == 1: |
|
219 | if nwords == 1: | |
220 | return get_root_modules() |
|
220 | return get_root_modules() | |
221 | mod = words[1].split('.') |
|
221 | mod = words[1].split('.') | |
222 | if len(mod) < 2: |
|
222 | if len(mod) < 2: | |
223 | return get_root_modules() |
|
223 | return get_root_modules() | |
224 | completion_list = try_import('.'.join(mod[:-1]), True) |
|
224 | completion_list = try_import('.'.join(mod[:-1]), True) | |
225 | return ['.'.join(mod[:-1] + [el]) for el in completion_list] |
|
225 | return ['.'.join(mod[:-1] + [el]) for el in completion_list] | |
226 |
|
226 | |||
227 | # 'from xyz import abc<tab>' |
|
227 | # 'from xyz import abc<tab>' | |
228 | if nwords >= 3 and words[0] == 'from': |
|
228 | if nwords >= 3 and words[0] == 'from': | |
229 | mod = words[1] |
|
229 | mod = words[1] | |
230 | return try_import(mod) |
|
230 | return try_import(mod) | |
231 |
|
231 | |||
232 | #----------------------------------------------------------------------------- |
|
232 | #----------------------------------------------------------------------------- | |
233 | # Completers |
|
233 | # Completers | |
234 | #----------------------------------------------------------------------------- |
|
234 | #----------------------------------------------------------------------------- | |
235 | # These all have the func(self, event) signature to be used as custom |
|
235 | # These all have the func(self, event) signature to be used as custom | |
236 | # completers |
|
236 | # completers | |
237 |
|
237 | |||
238 | def module_completer(self,event): |
|
238 | def module_completer(self,event): | |
239 | """Give completions after user has typed 'import ...' or 'from ...'""" |
|
239 | """Give completions after user has typed 'import ...' or 'from ...'""" | |
240 |
|
240 | |||
241 | # This works in all versions of python. While 2.5 has |
|
241 | # This works in all versions of python. While 2.5 has | |
242 | # pkgutil.walk_packages(), that particular routine is fairly dangerous, |
|
242 | # pkgutil.walk_packages(), that particular routine is fairly dangerous, | |
243 | # since it imports *EVERYTHING* on sys.path. That is: a) very slow b) full |
|
243 | # since it imports *EVERYTHING* on sys.path. That is: a) very slow b) full | |
244 | # of possibly problematic side effects. |
|
244 | # of possibly problematic side effects. | |
245 | # This search the folders in the sys.path for available modules. |
|
245 | # This search the folders in the sys.path for available modules. | |
246 |
|
246 | |||
247 | return module_completion(event.line) |
|
247 | return module_completion(event.line) | |
248 |
|
248 | |||
249 | # FIXME: there's a lot of logic common to the run, cd and builtin file |
|
249 | # FIXME: there's a lot of logic common to the run, cd and builtin file | |
250 | # completers, that is currently reimplemented in each. |
|
250 | # completers, that is currently reimplemented in each. | |
251 |
|
251 | |||
252 | def magic_run_completer(self, event): |
|
252 | def magic_run_completer(self, event): | |
253 | """Complete files that end in .py or .ipy for the %run command. |
|
253 | """Complete files that end in .py or .ipy for the %run command. | |
254 | """ |
|
254 | """ | |
255 | comps = arg_split(event.line, strict=False) |
|
255 | comps = arg_split(event.line, strict=False) | |
256 | relpath = (len(comps) > 1 and comps[-1] or '').strip("'\"") |
|
256 | relpath = (len(comps) > 1 and comps[-1] or '').strip("'\"") | |
257 |
|
257 | |||
258 | #print("\nev=", event) # dbg |
|
258 | #print("\nev=", event) # dbg | |
259 | #print("rp=", relpath) # dbg |
|
259 | #print("rp=", relpath) # dbg | |
260 | #print('comps=', comps) # dbg |
|
260 | #print('comps=', comps) # dbg | |
261 |
|
261 | |||
262 | lglob = glob.glob |
|
262 | lglob = glob.glob | |
263 | isdir = os.path.isdir |
|
263 | isdir = os.path.isdir | |
264 | relpath, tilde_expand, tilde_val = expand_user(relpath) |
|
264 | relpath, tilde_expand, tilde_val = expand_user(relpath) | |
265 |
|
265 | |||
266 | dirs = [f.replace('\\','/') + "/" for f in lglob(relpath+'*') if isdir(f)] |
|
266 | dirs = [f.replace('\\','/') + "/" for f in lglob(relpath+'*') if isdir(f)] | |
267 |
|
267 | |||
268 | # Find if the user has already typed the first filename, after which we |
|
268 | # Find if the user has already typed the first filename, after which we | |
269 | # should complete on all files, since after the first one other files may |
|
269 | # should complete on all files, since after the first one other files may | |
270 | # be arguments to the input script. |
|
270 | # be arguments to the input script. | |
271 |
|
271 | |||
272 | if filter(magic_run_re.match, comps): |
|
272 | if filter(magic_run_re.match, comps): | |
273 | pys = [f.replace('\\','/') for f in lglob('*')] |
|
273 | pys = [f.replace('\\','/') for f in lglob('*')] | |
274 | else: |
|
274 | else: | |
275 | pys = [f.replace('\\','/') |
|
275 | pys = [f.replace('\\','/') | |
276 | for f in lglob(relpath+'*.py') + lglob(relpath+'*.ipy') + |
|
276 | for f in lglob(relpath+'*.py') + lglob(relpath+'*.ipy') + | |
277 | lglob(relpath + '*.pyw')] |
|
277 | lglob(relpath + '*.pyw')] | |
278 | #print('run comp:', dirs+pys) # dbg |
|
278 | #print('run comp:', dirs+pys) # dbg | |
279 | return [compress_user(p, tilde_expand, tilde_val) for p in dirs+pys] |
|
279 | return [compress_user(p, tilde_expand, tilde_val) for p in dirs+pys] | |
280 |
|
280 | |||
281 |
|
281 | |||
282 | def cd_completer(self, event): |
|
282 | def cd_completer(self, event): | |
283 | """Completer function for cd, which only returns directories.""" |
|
283 | """Completer function for cd, which only returns directories.""" | |
284 | ip = get_ipython() |
|
284 | ip = get_ipython() | |
285 | relpath = event.symbol |
|
285 | relpath = event.symbol | |
286 |
|
286 | |||
287 | #print(event) # dbg |
|
287 | #print(event) # dbg | |
288 | if event.line.endswith('-b') or ' -b ' in event.line: |
|
288 | if event.line.endswith('-b') or ' -b ' in event.line: | |
289 | # return only bookmark completions |
|
289 | # return only bookmark completions | |
290 | bkms = self.db.get('bookmarks', None) |
|
290 | bkms = self.db.get('bookmarks', None) | |
291 | if bkms: |
|
291 | if bkms: | |
292 | return bkms.keys() |
|
292 | return bkms.keys() | |
293 | else: |
|
293 | else: | |
294 | return [] |
|
294 | return [] | |
295 |
|
295 | |||
296 | if event.symbol == '-': |
|
296 | if event.symbol == '-': | |
297 | width_dh = str(len(str(len(ip.user_ns['_dh']) + 1))) |
|
297 | width_dh = str(len(str(len(ip.user_ns['_dh']) + 1))) | |
298 | # jump in directory history by number |
|
298 | # jump in directory history by number | |
299 | fmt = '-%0' + width_dh +'d [%s]' |
|
299 | fmt = '-%0' + width_dh +'d [%s]' | |
300 | ents = [ fmt % (i,s) for i,s in enumerate(ip.user_ns['_dh'])] |
|
300 | ents = [ fmt % (i,s) for i,s in enumerate(ip.user_ns['_dh'])] | |
301 | if len(ents) > 1: |
|
301 | if len(ents) > 1: | |
302 | return ents |
|
302 | return ents | |
303 | return [] |
|
303 | return [] | |
304 |
|
304 | |||
305 | if event.symbol.startswith('--'): |
|
305 | if event.symbol.startswith('--'): | |
306 | return ["--" + os.path.basename(d) for d in ip.user_ns['_dh']] |
|
306 | return ["--" + os.path.basename(d) for d in ip.user_ns['_dh']] | |
307 |
|
307 | |||
308 | # Expand ~ in path and normalize directory separators. |
|
308 | # Expand ~ in path and normalize directory separators. | |
309 | relpath, tilde_expand, tilde_val = expand_user(relpath) |
|
309 | relpath, tilde_expand, tilde_val = expand_user(relpath) | |
310 | relpath = relpath.replace('\\','/') |
|
310 | relpath = relpath.replace('\\','/') | |
311 |
|
311 | |||
312 | found = [] |
|
312 | found = [] | |
313 | for d in [f.replace('\\','/') + '/' for f in glob.glob(relpath+'*') |
|
313 | for d in [f.replace('\\','/') + '/' for f in glob.glob(relpath+'*') | |
314 | if os.path.isdir(f)]: |
|
314 | if os.path.isdir(f)]: | |
315 | if ' ' in d: |
|
315 | if ' ' in d: | |
316 | # we don't want to deal with any of that, complex code |
|
316 | # we don't want to deal with any of that, complex code | |
317 | # for this is elsewhere |
|
317 | # for this is elsewhere | |
318 | raise TryNext |
|
318 | raise TryNext | |
319 |
|
319 | |||
320 | found.append(d) |
|
320 | found.append(d) | |
321 |
|
321 | |||
322 | if not found: |
|
322 | if not found: | |
323 | if os.path.isdir(relpath): |
|
323 | if os.path.isdir(relpath): | |
324 | return [compress_user(relpath, tilde_expand, tilde_val)] |
|
324 | return [compress_user(relpath, tilde_expand, tilde_val)] | |
325 |
|
325 | |||
326 | # if no completions so far, try bookmarks |
|
326 | # if no completions so far, try bookmarks | |
327 |
bks = self.db.get('bookmarks',{}) |
|
327 | bks = self.db.get('bookmarks',{}) | |
328 | bkmatches = [s for s in bks if s.startswith(event.symbol)] |
|
328 | bkmatches = [s for s in bks if s.startswith(event.symbol)] | |
329 | if bkmatches: |
|
329 | if bkmatches: | |
330 | return bkmatches |
|
330 | return bkmatches | |
331 |
|
331 | |||
332 | raise TryNext |
|
332 | raise TryNext | |
333 |
|
333 | |||
334 | return [compress_user(p, tilde_expand, tilde_val) for p in found] |
|
334 | return [compress_user(p, tilde_expand, tilde_val) for p in found] | |
335 |
|
335 | |||
336 | def reset_completer(self, event): |
|
336 | def reset_completer(self, event): | |
337 | "A completer for %reset magic" |
|
337 | "A completer for %reset magic" | |
338 | return '-f -s in out array dhist'.split() |
|
338 | return '-f -s in out array dhist'.split() |
@@ -1,789 +1,789 b'' | |||||
1 | #!/usr/bin/env python |
|
1 | #!/usr/bin/env python | |
2 | """An interactive kernel that talks to frontends over 0MQ.""" |
|
2 | """An interactive kernel that talks to frontends over 0MQ.""" | |
3 |
|
3 | |||
4 | #----------------------------------------------------------------------------- |
|
4 | #----------------------------------------------------------------------------- | |
5 | # Imports |
|
5 | # Imports | |
6 | #----------------------------------------------------------------------------- |
|
6 | #----------------------------------------------------------------------------- | |
7 | from __future__ import print_function |
|
7 | from __future__ import print_function | |
8 |
|
8 | |||
9 | # Standard library imports |
|
9 | # Standard library imports | |
10 | import sys |
|
10 | import sys | |
11 | import time |
|
11 | import time | |
12 | import traceback |
|
12 | import traceback | |
13 | import logging |
|
13 | import logging | |
14 | import uuid |
|
14 | import uuid | |
15 |
|
15 | |||
16 | from datetime import datetime |
|
16 | from datetime import datetime | |
17 | from signal import ( |
|
17 | from signal import ( | |
18 | signal, default_int_handler, SIGINT |
|
18 | signal, default_int_handler, SIGINT | |
19 | ) |
|
19 | ) | |
20 |
|
20 | |||
21 | # System library imports |
|
21 | # System library imports | |
22 | import zmq |
|
22 | import zmq | |
23 | from zmq.eventloop import ioloop |
|
23 | from zmq.eventloop import ioloop | |
24 | from zmq.eventloop.zmqstream import ZMQStream |
|
24 | from zmq.eventloop.zmqstream import ZMQStream | |
25 |
|
25 | |||
26 | # Local imports |
|
26 | # Local imports | |
27 | from IPython.config.configurable import Configurable |
|
27 | from IPython.config.configurable import Configurable | |
28 | from IPython.core.error import StdinNotImplementedError |
|
28 | from IPython.core.error import StdinNotImplementedError | |
29 | from IPython.core import release |
|
29 | from IPython.core import release | |
30 | from IPython.utils import py3compat |
|
30 | from IPython.utils import py3compat | |
31 | from IPython.utils.py3compat import builtin_mod, unicode_type, string_types |
|
31 | from IPython.utils.py3compat import builtin_mod, unicode_type, string_types | |
32 | from IPython.utils.jsonutil import json_clean |
|
32 | from IPython.utils.jsonutil import json_clean | |
33 | from IPython.utils.traitlets import ( |
|
33 | from IPython.utils.traitlets import ( | |
34 | Any, Instance, Float, Dict, List, Set, Integer, Unicode, |
|
34 | Any, Instance, Float, Dict, List, Set, Integer, Unicode, | |
35 | Type |
|
35 | Type | |
36 | ) |
|
36 | ) | |
37 |
|
37 | |||
38 | from .serialize import serialize_object, unpack_apply_message |
|
38 | from .serialize import serialize_object, unpack_apply_message | |
39 | from .session import Session |
|
39 | from .session import Session | |
40 | from .zmqshell import ZMQInteractiveShell |
|
40 | from .zmqshell import ZMQInteractiveShell | |
41 |
|
41 | |||
42 |
|
42 | |||
43 | #----------------------------------------------------------------------------- |
|
43 | #----------------------------------------------------------------------------- | |
44 | # Main kernel class |
|
44 | # Main kernel class | |
45 | #----------------------------------------------------------------------------- |
|
45 | #----------------------------------------------------------------------------- | |
46 |
|
46 | |||
47 | protocol_version = list(release.kernel_protocol_version_info) |
|
47 | protocol_version = list(release.kernel_protocol_version_info) | |
48 | ipython_version = list(release.version_info) |
|
48 | ipython_version = list(release.version_info) | |
49 | language_version = list(sys.version_info[:3]) |
|
49 | language_version = list(sys.version_info[:3]) | |
50 |
|
50 | |||
51 |
|
51 | |||
52 | class Kernel(Configurable): |
|
52 | class Kernel(Configurable): | |
53 |
|
53 | |||
54 | #--------------------------------------------------------------------------- |
|
54 | #--------------------------------------------------------------------------- | |
55 | # Kernel interface |
|
55 | # Kernel interface | |
56 | #--------------------------------------------------------------------------- |
|
56 | #--------------------------------------------------------------------------- | |
57 |
|
57 | |||
58 | # attribute to override with a GUI |
|
58 | # attribute to override with a GUI | |
59 | eventloop = Any(None) |
|
59 | eventloop = Any(None) | |
60 | def _eventloop_changed(self, name, old, new): |
|
60 | def _eventloop_changed(self, name, old, new): | |
61 | """schedule call to eventloop from IOLoop""" |
|
61 | """schedule call to eventloop from IOLoop""" | |
62 | loop = ioloop.IOLoop.instance() |
|
62 | loop = ioloop.IOLoop.instance() | |
63 | loop.add_timeout(time.time()+0.1, self.enter_eventloop) |
|
63 | loop.add_timeout(time.time()+0.1, self.enter_eventloop) | |
64 |
|
64 | |||
65 | shell = Instance('IPython.core.interactiveshell.InteractiveShellABC') |
|
65 | shell = Instance('IPython.core.interactiveshell.InteractiveShellABC') | |
66 | shell_class = Type(ZMQInteractiveShell) |
|
66 | shell_class = Type(ZMQInteractiveShell) | |
67 |
|
67 | |||
68 | session = Instance(Session) |
|
68 | session = Instance(Session) | |
69 | profile_dir = Instance('IPython.core.profiledir.ProfileDir') |
|
69 | profile_dir = Instance('IPython.core.profiledir.ProfileDir') | |
70 | shell_streams = List() |
|
70 | shell_streams = List() | |
71 | control_stream = Instance(ZMQStream) |
|
71 | control_stream = Instance(ZMQStream) | |
72 | iopub_socket = Instance(zmq.Socket) |
|
72 | iopub_socket = Instance(zmq.Socket) | |
73 | stdin_socket = Instance(zmq.Socket) |
|
73 | stdin_socket = Instance(zmq.Socket) | |
74 | log = Instance(logging.Logger) |
|
74 | log = Instance(logging.Logger) | |
75 |
|
75 | |||
76 | user_module = Any() |
|
76 | user_module = Any() | |
77 | def _user_module_changed(self, name, old, new): |
|
77 | def _user_module_changed(self, name, old, new): | |
78 | if self.shell is not None: |
|
78 | if self.shell is not None: | |
79 | self.shell.user_module = new |
|
79 | self.shell.user_module = new | |
80 |
|
80 | |||
81 | user_ns = Instance(dict, args=None, allow_none=True) |
|
81 | user_ns = Instance(dict, args=None, allow_none=True) | |
82 | def _user_ns_changed(self, name, old, new): |
|
82 | def _user_ns_changed(self, name, old, new): | |
83 | if self.shell is not None: |
|
83 | if self.shell is not None: | |
84 | self.shell.user_ns = new |
|
84 | self.shell.user_ns = new | |
85 | self.shell.init_user_ns() |
|
85 | self.shell.init_user_ns() | |
86 |
|
86 | |||
87 | # identities: |
|
87 | # identities: | |
88 | int_id = Integer(-1) |
|
88 | int_id = Integer(-1) | |
89 | ident = Unicode() |
|
89 | ident = Unicode() | |
90 |
|
90 | |||
91 | def _ident_default(self): |
|
91 | def _ident_default(self): | |
92 | return unicode_type(uuid.uuid4()) |
|
92 | return unicode_type(uuid.uuid4()) | |
93 |
|
93 | |||
94 |
|
94 | |||
95 | # Private interface |
|
95 | # Private interface | |
96 |
|
96 | |||
97 | # Time to sleep after flushing the stdout/err buffers in each execute |
|
97 | # Time to sleep after flushing the stdout/err buffers in each execute | |
98 | # cycle. While this introduces a hard limit on the minimal latency of the |
|
98 | # cycle. While this introduces a hard limit on the minimal latency of the | |
99 | # execute cycle, it helps prevent output synchronization problems for |
|
99 | # execute cycle, it helps prevent output synchronization problems for | |
100 | # clients. |
|
100 | # clients. | |
101 | # Units are in seconds. The minimum zmq latency on local host is probably |
|
101 | # Units are in seconds. The minimum zmq latency on local host is probably | |
102 | # ~150 microseconds, set this to 500us for now. We may need to increase it |
|
102 | # ~150 microseconds, set this to 500us for now. We may need to increase it | |
103 | # a little if it's not enough after more interactive testing. |
|
103 | # a little if it's not enough after more interactive testing. | |
104 | _execute_sleep = Float(0.0005, config=True) |
|
104 | _execute_sleep = Float(0.0005, config=True) | |
105 |
|
105 | |||
106 | # Frequency of the kernel's event loop. |
|
106 | # Frequency of the kernel's event loop. | |
107 | # Units are in seconds, kernel subclasses for GUI toolkits may need to |
|
107 | # Units are in seconds, kernel subclasses for GUI toolkits may need to | |
108 | # adapt to milliseconds. |
|
108 | # adapt to milliseconds. | |
109 | _poll_interval = Float(0.05, config=True) |
|
109 | _poll_interval = Float(0.05, config=True) | |
110 |
|
110 | |||
111 | # If the shutdown was requested over the network, we leave here the |
|
111 | # If the shutdown was requested over the network, we leave here the | |
112 | # necessary reply message so it can be sent by our registered atexit |
|
112 | # necessary reply message so it can be sent by our registered atexit | |
113 | # handler. This ensures that the reply is only sent to clients truly at |
|
113 | # handler. This ensures that the reply is only sent to clients truly at | |
114 | # the end of our shutdown process (which happens after the underlying |
|
114 | # the end of our shutdown process (which happens after the underlying | |
115 | # IPython shell's own shutdown). |
|
115 | # IPython shell's own shutdown). | |
116 | _shutdown_message = None |
|
116 | _shutdown_message = None | |
117 |
|
117 | |||
118 | # This is a dict of port number that the kernel is listening on. It is set |
|
118 | # This is a dict of port number that the kernel is listening on. It is set | |
119 | # by record_ports and used by connect_request. |
|
119 | # by record_ports and used by connect_request. | |
120 | _recorded_ports = Dict() |
|
120 | _recorded_ports = Dict() | |
121 |
|
121 | |||
122 | # A reference to the Python builtin 'raw_input' function. |
|
122 | # A reference to the Python builtin 'raw_input' function. | |
123 | # (i.e., __builtin__.raw_input for Python 2.7, builtins.input for Python 3) |
|
123 | # (i.e., __builtin__.raw_input for Python 2.7, builtins.input for Python 3) | |
124 | _sys_raw_input = Any() |
|
124 | _sys_raw_input = Any() | |
125 | _sys_eval_input = Any() |
|
125 | _sys_eval_input = Any() | |
126 |
|
126 | |||
127 | # set of aborted msg_ids |
|
127 | # set of aborted msg_ids | |
128 | aborted = Set() |
|
128 | aborted = Set() | |
129 |
|
129 | |||
130 |
|
130 | |||
131 | def __init__(self, **kwargs): |
|
131 | def __init__(self, **kwargs): | |
132 | super(Kernel, self).__init__(**kwargs) |
|
132 | super(Kernel, self).__init__(**kwargs) | |
133 |
|
133 | |||
134 | # Initialize the InteractiveShell subclass |
|
134 | # Initialize the InteractiveShell subclass | |
135 | self.shell = self.shell_class.instance(parent=self, |
|
135 | self.shell = self.shell_class.instance(parent=self, | |
136 | profile_dir = self.profile_dir, |
|
136 | profile_dir = self.profile_dir, | |
137 | user_module = self.user_module, |
|
137 | user_module = self.user_module, | |
138 | user_ns = self.user_ns, |
|
138 | user_ns = self.user_ns, | |
139 | kernel = self, |
|
139 | kernel = self, | |
140 | ) |
|
140 | ) | |
141 | self.shell.displayhook.session = self.session |
|
141 | self.shell.displayhook.session = self.session | |
142 | self.shell.displayhook.pub_socket = self.iopub_socket |
|
142 | self.shell.displayhook.pub_socket = self.iopub_socket | |
143 | self.shell.displayhook.topic = self._topic('pyout') |
|
143 | self.shell.displayhook.topic = self._topic('pyout') | |
144 | self.shell.display_pub.session = self.session |
|
144 | self.shell.display_pub.session = self.session | |
145 | self.shell.display_pub.pub_socket = self.iopub_socket |
|
145 | self.shell.display_pub.pub_socket = self.iopub_socket | |
146 | self.shell.data_pub.session = self.session |
|
146 | self.shell.data_pub.session = self.session | |
147 | self.shell.data_pub.pub_socket = self.iopub_socket |
|
147 | self.shell.data_pub.pub_socket = self.iopub_socket | |
148 |
|
148 | |||
149 | # TMP - hack while developing |
|
149 | # TMP - hack while developing | |
150 | self.shell._reply_content = None |
|
150 | self.shell._reply_content = None | |
151 |
|
151 | |||
152 | # Build dict of handlers for message types |
|
152 | # Build dict of handlers for message types | |
153 | msg_types = [ 'execute_request', 'complete_request', |
|
153 | msg_types = [ 'execute_request', 'complete_request', | |
154 | 'object_info_request', 'history_request', |
|
154 | 'object_info_request', 'history_request', | |
155 | 'kernel_info_request', |
|
155 | 'kernel_info_request', | |
156 | 'connect_request', 'shutdown_request', |
|
156 | 'connect_request', 'shutdown_request', | |
157 | 'apply_request', |
|
157 | 'apply_request', | |
158 | ] |
|
158 | ] | |
159 | self.shell_handlers = {} |
|
159 | self.shell_handlers = {} | |
160 | for msg_type in msg_types: |
|
160 | for msg_type in msg_types: | |
161 | self.shell_handlers[msg_type] = getattr(self, msg_type) |
|
161 | self.shell_handlers[msg_type] = getattr(self, msg_type) | |
162 |
|
162 | |||
163 | comm_msg_types = [ 'comm_open', 'comm_msg', 'comm_close' ] |
|
163 | comm_msg_types = [ 'comm_open', 'comm_msg', 'comm_close' ] | |
164 | comm_manager = self.shell.comm_manager |
|
164 | comm_manager = self.shell.comm_manager | |
165 | for msg_type in comm_msg_types: |
|
165 | for msg_type in comm_msg_types: | |
166 | self.shell_handlers[msg_type] = getattr(comm_manager, msg_type) |
|
166 | self.shell_handlers[msg_type] = getattr(comm_manager, msg_type) | |
167 |
|
167 | |||
168 | control_msg_types = msg_types + [ 'clear_request', 'abort_request' ] |
|
168 | control_msg_types = msg_types + [ 'clear_request', 'abort_request' ] | |
169 | self.control_handlers = {} |
|
169 | self.control_handlers = {} | |
170 | for msg_type in control_msg_types: |
|
170 | for msg_type in control_msg_types: | |
171 | self.control_handlers[msg_type] = getattr(self, msg_type) |
|
171 | self.control_handlers[msg_type] = getattr(self, msg_type) | |
172 |
|
172 | |||
173 |
|
173 | |||
174 | def dispatch_control(self, msg): |
|
174 | def dispatch_control(self, msg): | |
175 | """dispatch control requests""" |
|
175 | """dispatch control requests""" | |
176 | idents,msg = self.session.feed_identities(msg, copy=False) |
|
176 | idents,msg = self.session.feed_identities(msg, copy=False) | |
177 | try: |
|
177 | try: | |
178 | msg = self.session.unserialize(msg, content=True, copy=False) |
|
178 | msg = self.session.unserialize(msg, content=True, copy=False) | |
179 | except: |
|
179 | except: | |
180 | self.log.error("Invalid Control Message", exc_info=True) |
|
180 | self.log.error("Invalid Control Message", exc_info=True) | |
181 | return |
|
181 | return | |
182 |
|
182 | |||
183 | self.log.debug("Control received: %s", msg) |
|
183 | self.log.debug("Control received: %s", msg) | |
184 |
|
184 | |||
185 | header = msg['header'] |
|
185 | header = msg['header'] | |
186 | msg_id = header['msg_id'] |
|
186 | msg_id = header['msg_id'] | |
187 | msg_type = header['msg_type'] |
|
187 | msg_type = header['msg_type'] | |
188 |
|
188 | |||
189 | handler = self.control_handlers.get(msg_type, None) |
|
189 | handler = self.control_handlers.get(msg_type, None) | |
190 | if handler is None: |
|
190 | if handler is None: | |
191 | self.log.error("UNKNOWN CONTROL MESSAGE TYPE: %r", msg_type) |
|
191 | self.log.error("UNKNOWN CONTROL MESSAGE TYPE: %r", msg_type) | |
192 | else: |
|
192 | else: | |
193 | try: |
|
193 | try: | |
194 | handler(self.control_stream, idents, msg) |
|
194 | handler(self.control_stream, idents, msg) | |
195 | except Exception: |
|
195 | except Exception: | |
196 | self.log.error("Exception in control handler:", exc_info=True) |
|
196 | self.log.error("Exception in control handler:", exc_info=True) | |
197 |
|
197 | |||
198 | def dispatch_shell(self, stream, msg): |
|
198 | def dispatch_shell(self, stream, msg): | |
199 | """dispatch shell requests""" |
|
199 | """dispatch shell requests""" | |
200 | # flush control requests first |
|
200 | # flush control requests first | |
201 | if self.control_stream: |
|
201 | if self.control_stream: | |
202 | self.control_stream.flush() |
|
202 | self.control_stream.flush() | |
203 |
|
203 | |||
204 | idents,msg = self.session.feed_identities(msg, copy=False) |
|
204 | idents,msg = self.session.feed_identities(msg, copy=False) | |
205 | try: |
|
205 | try: | |
206 | msg = self.session.unserialize(msg, content=True, copy=False) |
|
206 | msg = self.session.unserialize(msg, content=True, copy=False) | |
207 | except: |
|
207 | except: | |
208 | self.log.error("Invalid Message", exc_info=True) |
|
208 | self.log.error("Invalid Message", exc_info=True) | |
209 | return |
|
209 | return | |
210 |
|
210 | |||
211 | header = msg['header'] |
|
211 | header = msg['header'] | |
212 | msg_id = header['msg_id'] |
|
212 | msg_id = header['msg_id'] | |
213 | msg_type = msg['header']['msg_type'] |
|
213 | msg_type = msg['header']['msg_type'] | |
214 |
|
214 | |||
215 | # Print some info about this message and leave a '--->' marker, so it's |
|
215 | # Print some info about this message and leave a '--->' marker, so it's | |
216 | # easier to trace visually the message chain when debugging. Each |
|
216 | # easier to trace visually the message chain when debugging. Each | |
217 | # handler prints its message at the end. |
|
217 | # handler prints its message at the end. | |
218 | self.log.debug('\n*** MESSAGE TYPE:%s***', msg_type) |
|
218 | self.log.debug('\n*** MESSAGE TYPE:%s***', msg_type) | |
219 | self.log.debug(' Content: %s\n --->\n ', msg['content']) |
|
219 | self.log.debug(' Content: %s\n --->\n ', msg['content']) | |
220 |
|
220 | |||
221 | if msg_id in self.aborted: |
|
221 | if msg_id in self.aborted: | |
222 | self.aborted.remove(msg_id) |
|
222 | self.aborted.remove(msg_id) | |
223 | # is it safe to assume a msg_id will not be resubmitted? |
|
223 | # is it safe to assume a msg_id will not be resubmitted? | |
224 | reply_type = msg_type.split('_')[0] + '_reply' |
|
224 | reply_type = msg_type.split('_')[0] + '_reply' | |
225 | status = {'status' : 'aborted'} |
|
225 | status = {'status' : 'aborted'} | |
226 | md = {'engine' : self.ident} |
|
226 | md = {'engine' : self.ident} | |
227 | md.update(status) |
|
227 | md.update(status) | |
228 | reply_msg = self.session.send(stream, reply_type, metadata=md, |
|
228 | reply_msg = self.session.send(stream, reply_type, metadata=md, | |
229 | content=status, parent=msg, ident=idents) |
|
229 | content=status, parent=msg, ident=idents) | |
230 | return |
|
230 | return | |
231 |
|
231 | |||
232 | handler = self.shell_handlers.get(msg_type, None) |
|
232 | handler = self.shell_handlers.get(msg_type, None) | |
233 | if handler is None: |
|
233 | if handler is None: | |
234 | self.log.error("UNKNOWN MESSAGE TYPE: %r", msg_type) |
|
234 | self.log.error("UNKNOWN MESSAGE TYPE: %r", msg_type) | |
235 | else: |
|
235 | else: | |
236 | # ensure default_int_handler during handler call |
|
236 | # ensure default_int_handler during handler call | |
237 | sig = signal(SIGINT, default_int_handler) |
|
237 | sig = signal(SIGINT, default_int_handler) | |
238 | try: |
|
238 | try: | |
239 | handler(stream, idents, msg) |
|
239 | handler(stream, idents, msg) | |
240 | except Exception: |
|
240 | except Exception: | |
241 | self.log.error("Exception in message handler:", exc_info=True) |
|
241 | self.log.error("Exception in message handler:", exc_info=True) | |
242 | finally: |
|
242 | finally: | |
243 | signal(SIGINT, sig) |
|
243 | signal(SIGINT, sig) | |
244 |
|
244 | |||
245 | def enter_eventloop(self): |
|
245 | def enter_eventloop(self): | |
246 | """enter eventloop""" |
|
246 | """enter eventloop""" | |
247 | self.log.info("entering eventloop") |
|
247 | self.log.info("entering eventloop") | |
248 | # restore default_int_handler |
|
248 | # restore default_int_handler | |
249 | signal(SIGINT, default_int_handler) |
|
249 | signal(SIGINT, default_int_handler) | |
250 | while self.eventloop is not None: |
|
250 | while self.eventloop is not None: | |
251 | try: |
|
251 | try: | |
252 | self.eventloop(self) |
|
252 | self.eventloop(self) | |
253 | except KeyboardInterrupt: |
|
253 | except KeyboardInterrupt: | |
254 | # Ctrl-C shouldn't crash the kernel |
|
254 | # Ctrl-C shouldn't crash the kernel | |
255 | self.log.error("KeyboardInterrupt caught in kernel") |
|
255 | self.log.error("KeyboardInterrupt caught in kernel") | |
256 | continue |
|
256 | continue | |
257 | else: |
|
257 | else: | |
258 | # eventloop exited cleanly, this means we should stop (right?) |
|
258 | # eventloop exited cleanly, this means we should stop (right?) | |
259 | self.eventloop = None |
|
259 | self.eventloop = None | |
260 | break |
|
260 | break | |
261 | self.log.info("exiting eventloop") |
|
261 | self.log.info("exiting eventloop") | |
262 |
|
262 | |||
263 | def start(self): |
|
263 | def start(self): | |
264 | """register dispatchers for streams""" |
|
264 | """register dispatchers for streams""" | |
265 | self.shell.exit_now = False |
|
265 | self.shell.exit_now = False | |
266 | if self.control_stream: |
|
266 | if self.control_stream: | |
267 | self.control_stream.on_recv(self.dispatch_control, copy=False) |
|
267 | self.control_stream.on_recv(self.dispatch_control, copy=False) | |
268 |
|
268 | |||
269 | def make_dispatcher(stream): |
|
269 | def make_dispatcher(stream): | |
270 | def dispatcher(msg): |
|
270 | def dispatcher(msg): | |
271 | return self.dispatch_shell(stream, msg) |
|
271 | return self.dispatch_shell(stream, msg) | |
272 | return dispatcher |
|
272 | return dispatcher | |
273 |
|
273 | |||
274 | for s in self.shell_streams: |
|
274 | for s in self.shell_streams: | |
275 | s.on_recv(make_dispatcher(s), copy=False) |
|
275 | s.on_recv(make_dispatcher(s), copy=False) | |
276 |
|
276 | |||
277 | # publish idle status |
|
277 | # publish idle status | |
278 | self._publish_status('starting') |
|
278 | self._publish_status('starting') | |
279 |
|
279 | |||
280 | def do_one_iteration(self): |
|
280 | def do_one_iteration(self): | |
281 | """step eventloop just once""" |
|
281 | """step eventloop just once""" | |
282 | if self.control_stream: |
|
282 | if self.control_stream: | |
283 | self.control_stream.flush() |
|
283 | self.control_stream.flush() | |
284 | for stream in self.shell_streams: |
|
284 | for stream in self.shell_streams: | |
285 | # handle at most one request per iteration |
|
285 | # handle at most one request per iteration | |
286 | stream.flush(zmq.POLLIN, 1) |
|
286 | stream.flush(zmq.POLLIN, 1) | |
287 | stream.flush(zmq.POLLOUT) |
|
287 | stream.flush(zmq.POLLOUT) | |
288 |
|
288 | |||
289 |
|
289 | |||
290 | def record_ports(self, ports): |
|
290 | def record_ports(self, ports): | |
291 | """Record the ports that this kernel is using. |
|
291 | """Record the ports that this kernel is using. | |
292 |
|
292 | |||
293 | The creator of the Kernel instance must call this methods if they |
|
293 | The creator of the Kernel instance must call this methods if they | |
294 | want the :meth:`connect_request` method to return the port numbers. |
|
294 | want the :meth:`connect_request` method to return the port numbers. | |
295 | """ |
|
295 | """ | |
296 | self._recorded_ports = ports |
|
296 | self._recorded_ports = ports | |
297 |
|
297 | |||
298 | #--------------------------------------------------------------------------- |
|
298 | #--------------------------------------------------------------------------- | |
299 | # Kernel request handlers |
|
299 | # Kernel request handlers | |
300 | #--------------------------------------------------------------------------- |
|
300 | #--------------------------------------------------------------------------- | |
301 |
|
301 | |||
302 | def _make_metadata(self, other=None): |
|
302 | def _make_metadata(self, other=None): | |
303 | """init metadata dict, for execute/apply_reply""" |
|
303 | """init metadata dict, for execute/apply_reply""" | |
304 | new_md = { |
|
304 | new_md = { | |
305 | 'dependencies_met' : True, |
|
305 | 'dependencies_met' : True, | |
306 | 'engine' : self.ident, |
|
306 | 'engine' : self.ident, | |
307 | 'started': datetime.now(), |
|
307 | 'started': datetime.now(), | |
308 | } |
|
308 | } | |
309 | if other: |
|
309 | if other: | |
310 | new_md.update(other) |
|
310 | new_md.update(other) | |
311 | return new_md |
|
311 | return new_md | |
312 |
|
312 | |||
313 | def _publish_pyin(self, code, parent, execution_count): |
|
313 | def _publish_pyin(self, code, parent, execution_count): | |
314 | """Publish the code request on the pyin stream.""" |
|
314 | """Publish the code request on the pyin stream.""" | |
315 |
|
315 | |||
316 | self.session.send(self.iopub_socket, u'pyin', |
|
316 | self.session.send(self.iopub_socket, u'pyin', | |
317 | {u'code':code, u'execution_count': execution_count}, |
|
317 | {u'code':code, u'execution_count': execution_count}, | |
318 | parent=parent, ident=self._topic('pyin') |
|
318 | parent=parent, ident=self._topic('pyin') | |
319 | ) |
|
319 | ) | |
320 |
|
320 | |||
321 | def _publish_status(self, status, parent=None): |
|
321 | def _publish_status(self, status, parent=None): | |
322 | """send status (busy/idle) on IOPub""" |
|
322 | """send status (busy/idle) on IOPub""" | |
323 | self.session.send(self.iopub_socket, |
|
323 | self.session.send(self.iopub_socket, | |
324 | u'status', |
|
324 | u'status', | |
325 | {u'execution_state': status}, |
|
325 | {u'execution_state': status}, | |
326 | parent=parent, |
|
326 | parent=parent, | |
327 | ident=self._topic('status'), |
|
327 | ident=self._topic('status'), | |
328 | ) |
|
328 | ) | |
329 |
|
329 | |||
330 |
|
330 | |||
331 | def execute_request(self, stream, ident, parent): |
|
331 | def execute_request(self, stream, ident, parent): | |
332 | """handle an execute_request""" |
|
332 | """handle an execute_request""" | |
333 |
|
333 | |||
334 | self._publish_status(u'busy', parent) |
|
334 | self._publish_status(u'busy', parent) | |
335 |
|
335 | |||
336 | try: |
|
336 | try: | |
337 | content = parent[u'content'] |
|
337 | content = parent[u'content'] | |
338 | code = content[u'code'] |
|
338 | code = content[u'code'] | |
339 | silent = content[u'silent'] |
|
339 | silent = content[u'silent'] | |
340 | store_history = content.get(u'store_history', not silent) |
|
340 | store_history = content.get(u'store_history', not silent) | |
341 | except: |
|
341 | except: | |
342 | self.log.error("Got bad msg: ") |
|
342 | self.log.error("Got bad msg: ") | |
343 | self.log.error("%s", parent) |
|
343 | self.log.error("%s", parent) | |
344 | return |
|
344 | return | |
345 |
|
345 | |||
346 | md = self._make_metadata(parent['metadata']) |
|
346 | md = self._make_metadata(parent['metadata']) | |
347 |
|
347 | |||
348 | shell = self.shell # we'll need this a lot here |
|
348 | shell = self.shell # we'll need this a lot here | |
349 |
|
349 | |||
350 | # Replace raw_input. Note that is not sufficient to replace |
|
350 | # Replace raw_input. Note that is not sufficient to replace | |
351 | # raw_input in the user namespace. |
|
351 | # raw_input in the user namespace. | |
352 | if content.get('allow_stdin', False): |
|
352 | if content.get('allow_stdin', False): | |
353 | raw_input = lambda prompt='': self._raw_input(prompt, ident, parent) |
|
353 | raw_input = lambda prompt='': self._raw_input(prompt, ident, parent) | |
354 | input = lambda prompt='': eval(raw_input(prompt)) |
|
354 | input = lambda prompt='': eval(raw_input(prompt)) | |
355 | else: |
|
355 | else: | |
356 | raw_input = input = lambda prompt='' : self._no_raw_input() |
|
356 | raw_input = input = lambda prompt='' : self._no_raw_input() | |
357 |
|
357 | |||
358 | if py3compat.PY3: |
|
358 | if py3compat.PY3: | |
359 | self._sys_raw_input = builtin_mod.input |
|
359 | self._sys_raw_input = builtin_mod.input | |
360 | builtin_mod.input = raw_input |
|
360 | builtin_mod.input = raw_input | |
361 | else: |
|
361 | else: | |
362 | self._sys_raw_input = builtin_mod.raw_input |
|
362 | self._sys_raw_input = builtin_mod.raw_input | |
363 | self._sys_eval_input = builtin_mod.input |
|
363 | self._sys_eval_input = builtin_mod.input | |
364 | builtin_mod.raw_input = raw_input |
|
364 | builtin_mod.raw_input = raw_input | |
365 | builtin_mod.input = input |
|
365 | builtin_mod.input = input | |
366 |
|
366 | |||
367 | # Set the parent message of the display hook and out streams. |
|
367 | # Set the parent message of the display hook and out streams. | |
368 | shell.set_parent(parent) |
|
368 | shell.set_parent(parent) | |
369 |
|
369 | |||
370 | # Re-broadcast our input for the benefit of listening clients, and |
|
370 | # Re-broadcast our input for the benefit of listening clients, and | |
371 | # start computing output |
|
371 | # start computing output | |
372 | if not silent: |
|
372 | if not silent: | |
373 | self._publish_pyin(code, parent, shell.execution_count) |
|
373 | self._publish_pyin(code, parent, shell.execution_count) | |
374 |
|
374 | |||
375 | reply_content = {} |
|
375 | reply_content = {} | |
376 | try: |
|
376 | try: | |
377 | # FIXME: the shell calls the exception handler itself. |
|
377 | # FIXME: the shell calls the exception handler itself. | |
378 | shell.run_cell(code, store_history=store_history, silent=silent) |
|
378 | shell.run_cell(code, store_history=store_history, silent=silent) | |
379 | except: |
|
379 | except: | |
380 | status = u'error' |
|
380 | status = u'error' | |
381 | # FIXME: this code right now isn't being used yet by default, |
|
381 | # FIXME: this code right now isn't being used yet by default, | |
382 | # because the run_cell() call above directly fires off exception |
|
382 | # because the run_cell() call above directly fires off exception | |
383 | # reporting. This code, therefore, is only active in the scenario |
|
383 | # reporting. This code, therefore, is only active in the scenario | |
384 | # where runlines itself has an unhandled exception. We need to |
|
384 | # where runlines itself has an unhandled exception. We need to | |
385 | # uniformize this, for all exception construction to come from a |
|
385 | # uniformize this, for all exception construction to come from a | |
386 | # single location in the codbase. |
|
386 | # single location in the codbase. | |
387 | etype, evalue, tb = sys.exc_info() |
|
387 | etype, evalue, tb = sys.exc_info() | |
388 | tb_list = traceback.format_exception(etype, evalue, tb) |
|
388 | tb_list = traceback.format_exception(etype, evalue, tb) | |
389 | reply_content.update(shell._showtraceback(etype, evalue, tb_list)) |
|
389 | reply_content.update(shell._showtraceback(etype, evalue, tb_list)) | |
390 | else: |
|
390 | else: | |
391 | status = u'ok' |
|
391 | status = u'ok' | |
392 | finally: |
|
392 | finally: | |
393 | # Restore raw_input. |
|
393 | # Restore raw_input. | |
394 | if py3compat.PY3: |
|
394 | if py3compat.PY3: | |
395 | builtin_mod.input = self._sys_raw_input |
|
395 | builtin_mod.input = self._sys_raw_input | |
396 | else: |
|
396 | else: | |
397 | builtin_mod.raw_input = self._sys_raw_input |
|
397 | builtin_mod.raw_input = self._sys_raw_input | |
398 | builtin_mod.input = self._sys_eval_input |
|
398 | builtin_mod.input = self._sys_eval_input | |
399 |
|
399 | |||
400 | reply_content[u'status'] = status |
|
400 | reply_content[u'status'] = status | |
401 |
|
401 | |||
402 | # Return the execution counter so clients can display prompts |
|
402 | # Return the execution counter so clients can display prompts | |
403 | reply_content['execution_count'] = shell.execution_count - 1 |
|
403 | reply_content['execution_count'] = shell.execution_count - 1 | |
404 |
|
404 | |||
405 | # FIXME - fish exception info out of shell, possibly left there by |
|
405 | # FIXME - fish exception info out of shell, possibly left there by | |
406 | # runlines. We'll need to clean up this logic later. |
|
406 | # runlines. We'll need to clean up this logic later. | |
407 | if shell._reply_content is not None: |
|
407 | if shell._reply_content is not None: | |
408 | reply_content.update(shell._reply_content) |
|
408 | reply_content.update(shell._reply_content) | |
409 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method='execute') |
|
409 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method='execute') | |
410 | reply_content['engine_info'] = e_info |
|
410 | reply_content['engine_info'] = e_info | |
411 | # reset after use |
|
411 | # reset after use | |
412 | shell._reply_content = None |
|
412 | shell._reply_content = None | |
413 |
|
413 | |||
414 | if 'traceback' in reply_content: |
|
414 | if 'traceback' in reply_content: | |
415 | self.log.info("Exception in execute request:\n%s", '\n'.join(reply_content['traceback'])) |
|
415 | self.log.info("Exception in execute request:\n%s", '\n'.join(reply_content['traceback'])) | |
416 |
|
416 | |||
417 |
|
417 | |||
418 | # At this point, we can tell whether the main code execution succeeded |
|
418 | # At this point, we can tell whether the main code execution succeeded | |
419 | # or not. If it did, we proceed to evaluate user_variables/expressions |
|
419 | # or not. If it did, we proceed to evaluate user_variables/expressions | |
420 | if reply_content['status'] == 'ok': |
|
420 | if reply_content['status'] == 'ok': | |
421 | reply_content[u'user_variables'] = \ |
|
421 | reply_content[u'user_variables'] = \ | |
422 | shell.user_variables(content.get(u'user_variables', [])) |
|
422 | shell.user_variables(content.get(u'user_variables', [])) | |
423 | reply_content[u'user_expressions'] = \ |
|
423 | reply_content[u'user_expressions'] = \ | |
424 | shell.user_expressions(content.get(u'user_expressions', {})) |
|
424 | shell.user_expressions(content.get(u'user_expressions', {})) | |
425 | else: |
|
425 | else: | |
426 | # If there was an error, don't even try to compute variables or |
|
426 | # If there was an error, don't even try to compute variables or | |
427 | # expressions |
|
427 | # expressions | |
428 | reply_content[u'user_variables'] = {} |
|
428 | reply_content[u'user_variables'] = {} | |
429 | reply_content[u'user_expressions'] = {} |
|
429 | reply_content[u'user_expressions'] = {} | |
430 |
|
430 | |||
431 | # Payloads should be retrieved regardless of outcome, so we can both |
|
431 | # Payloads should be retrieved regardless of outcome, so we can both | |
432 | # recover partial output (that could have been generated early in a |
|
432 | # recover partial output (that could have been generated early in a | |
433 | # block, before an error) and clear the payload system always. |
|
433 | # block, before an error) and clear the payload system always. | |
434 | reply_content[u'payload'] = shell.payload_manager.read_payload() |
|
434 | reply_content[u'payload'] = shell.payload_manager.read_payload() | |
435 | # Be agressive about clearing the payload because we don't want |
|
435 | # Be agressive about clearing the payload because we don't want | |
436 | # it to sit in memory until the next execute_request comes in. |
|
436 | # it to sit in memory until the next execute_request comes in. | |
437 | shell.payload_manager.clear_payload() |
|
437 | shell.payload_manager.clear_payload() | |
438 |
|
438 | |||
439 | # Flush output before sending the reply. |
|
439 | # Flush output before sending the reply. | |
440 | sys.stdout.flush() |
|
440 | sys.stdout.flush() | |
441 | sys.stderr.flush() |
|
441 | sys.stderr.flush() | |
442 | # FIXME: on rare occasions, the flush doesn't seem to make it to the |
|
442 | # FIXME: on rare occasions, the flush doesn't seem to make it to the | |
443 | # clients... This seems to mitigate the problem, but we definitely need |
|
443 | # clients... This seems to mitigate the problem, but we definitely need | |
444 | # to better understand what's going on. |
|
444 | # to better understand what's going on. | |
445 | if self._execute_sleep: |
|
445 | if self._execute_sleep: | |
446 | time.sleep(self._execute_sleep) |
|
446 | time.sleep(self._execute_sleep) | |
447 |
|
447 | |||
448 | # Send the reply. |
|
448 | # Send the reply. | |
449 | reply_content = json_clean(reply_content) |
|
449 | reply_content = json_clean(reply_content) | |
450 |
|
450 | |||
451 | md['status'] = reply_content['status'] |
|
451 | md['status'] = reply_content['status'] | |
452 | if reply_content['status'] == 'error' and \ |
|
452 | if reply_content['status'] == 'error' and \ | |
453 | reply_content['ename'] == 'UnmetDependency': |
|
453 | reply_content['ename'] == 'UnmetDependency': | |
454 | md['dependencies_met'] = False |
|
454 | md['dependencies_met'] = False | |
455 |
|
455 | |||
456 | reply_msg = self.session.send(stream, u'execute_reply', |
|
456 | reply_msg = self.session.send(stream, u'execute_reply', | |
457 | reply_content, parent, metadata=md, |
|
457 | reply_content, parent, metadata=md, | |
458 | ident=ident) |
|
458 | ident=ident) | |
459 |
|
459 | |||
460 | self.log.debug("%s", reply_msg) |
|
460 | self.log.debug("%s", reply_msg) | |
461 |
|
461 | |||
462 | if not silent and reply_msg['content']['status'] == u'error': |
|
462 | if not silent and reply_msg['content']['status'] == u'error': | |
463 | self._abort_queues() |
|
463 | self._abort_queues() | |
464 |
|
464 | |||
465 | self._publish_status(u'idle', parent) |
|
465 | self._publish_status(u'idle', parent) | |
466 |
|
466 | |||
467 | def complete_request(self, stream, ident, parent): |
|
467 | def complete_request(self, stream, ident, parent): | |
468 | txt, matches = self._complete(parent) |
|
468 | txt, matches = self._complete(parent) | |
469 | matches = {'matches' : matches, |
|
469 | matches = {'matches' : matches, | |
470 | 'matched_text' : txt, |
|
470 | 'matched_text' : txt, | |
471 | 'status' : 'ok'} |
|
471 | 'status' : 'ok'} | |
472 | matches = json_clean(matches) |
|
472 | matches = json_clean(matches) | |
473 | completion_msg = self.session.send(stream, 'complete_reply', |
|
473 | completion_msg = self.session.send(stream, 'complete_reply', | |
474 | matches, parent, ident) |
|
474 | matches, parent, ident) | |
475 | self.log.debug("%s", completion_msg) |
|
475 | self.log.debug("%s", completion_msg) | |
476 |
|
476 | |||
477 | def object_info_request(self, stream, ident, parent): |
|
477 | def object_info_request(self, stream, ident, parent): | |
478 | content = parent['content'] |
|
478 | content = parent['content'] | |
479 | object_info = self.shell.object_inspect(content['oname'], |
|
479 | object_info = self.shell.object_inspect(content['oname'], | |
480 | detail_level = content.get('detail_level', 0) |
|
480 | detail_level = content.get('detail_level', 0) | |
481 | ) |
|
481 | ) | |
482 | # Before we send this object over, we scrub it for JSON usage |
|
482 | # Before we send this object over, we scrub it for JSON usage | |
483 | oinfo = json_clean(object_info) |
|
483 | oinfo = json_clean(object_info) | |
484 | msg = self.session.send(stream, 'object_info_reply', |
|
484 | msg = self.session.send(stream, 'object_info_reply', | |
485 | oinfo, parent, ident) |
|
485 | oinfo, parent, ident) | |
486 | self.log.debug("%s", msg) |
|
486 | self.log.debug("%s", msg) | |
487 |
|
487 | |||
488 | def history_request(self, stream, ident, parent): |
|
488 | def history_request(self, stream, ident, parent): | |
489 | # We need to pull these out, as passing **kwargs doesn't work with |
|
489 | # We need to pull these out, as passing **kwargs doesn't work with | |
490 | # unicode keys before Python 2.6.5. |
|
490 | # unicode keys before Python 2.6.5. | |
491 | hist_access_type = parent['content']['hist_access_type'] |
|
491 | hist_access_type = parent['content']['hist_access_type'] | |
492 | raw = parent['content']['raw'] |
|
492 | raw = parent['content']['raw'] | |
493 | output = parent['content']['output'] |
|
493 | output = parent['content']['output'] | |
494 | if hist_access_type == 'tail': |
|
494 | if hist_access_type == 'tail': | |
495 | n = parent['content']['n'] |
|
495 | n = parent['content']['n'] | |
496 | hist = self.shell.history_manager.get_tail(n, raw=raw, output=output, |
|
496 | hist = self.shell.history_manager.get_tail(n, raw=raw, output=output, | |
497 | include_latest=True) |
|
497 | include_latest=True) | |
498 |
|
498 | |||
499 | elif hist_access_type == 'range': |
|
499 | elif hist_access_type == 'range': | |
500 | session = parent['content']['session'] |
|
500 | session = parent['content']['session'] | |
501 | start = parent['content']['start'] |
|
501 | start = parent['content']['start'] | |
502 | stop = parent['content']['stop'] |
|
502 | stop = parent['content']['stop'] | |
503 | hist = self.shell.history_manager.get_range(session, start, stop, |
|
503 | hist = self.shell.history_manager.get_range(session, start, stop, | |
504 | raw=raw, output=output) |
|
504 | raw=raw, output=output) | |
505 |
|
505 | |||
506 | elif hist_access_type == 'search': |
|
506 | elif hist_access_type == 'search': | |
507 | n = parent['content'].get('n') |
|
507 | n = parent['content'].get('n') | |
508 | unique = parent['content'].get('unique', False) |
|
508 | unique = parent['content'].get('unique', False) | |
509 | pattern = parent['content']['pattern'] |
|
509 | pattern = parent['content']['pattern'] | |
510 | hist = self.shell.history_manager.search( |
|
510 | hist = self.shell.history_manager.search( | |
511 | pattern, raw=raw, output=output, n=n, unique=unique) |
|
511 | pattern, raw=raw, output=output, n=n, unique=unique) | |
512 |
|
512 | |||
513 | else: |
|
513 | else: | |
514 | hist = [] |
|
514 | hist = [] | |
515 | hist = list(hist) |
|
515 | hist = list(hist) | |
516 | content = {'history' : hist} |
|
516 | content = {'history' : hist} | |
517 | content = json_clean(content) |
|
517 | content = json_clean(content) | |
518 | msg = self.session.send(stream, 'history_reply', |
|
518 | msg = self.session.send(stream, 'history_reply', | |
519 | content, parent, ident) |
|
519 | content, parent, ident) | |
520 | self.log.debug("Sending history reply with %i entries", len(hist)) |
|
520 | self.log.debug("Sending history reply with %i entries", len(hist)) | |
521 |
|
521 | |||
522 | def connect_request(self, stream, ident, parent): |
|
522 | def connect_request(self, stream, ident, parent): | |
523 | if self._recorded_ports is not None: |
|
523 | if self._recorded_ports is not None: | |
524 | content = self._recorded_ports.copy() |
|
524 | content = self._recorded_ports.copy() | |
525 | else: |
|
525 | else: | |
526 | content = {} |
|
526 | content = {} | |
527 | msg = self.session.send(stream, 'connect_reply', |
|
527 | msg = self.session.send(stream, 'connect_reply', | |
528 | content, parent, ident) |
|
528 | content, parent, ident) | |
529 | self.log.debug("%s", msg) |
|
529 | self.log.debug("%s", msg) | |
530 |
|
530 | |||
531 | def kernel_info_request(self, stream, ident, parent): |
|
531 | def kernel_info_request(self, stream, ident, parent): | |
532 | vinfo = { |
|
532 | vinfo = { | |
533 | 'protocol_version': protocol_version, |
|
533 | 'protocol_version': protocol_version, | |
534 | 'ipython_version': ipython_version, |
|
534 | 'ipython_version': ipython_version, | |
535 | 'language_version': language_version, |
|
535 | 'language_version': language_version, | |
536 | 'language': 'python', |
|
536 | 'language': 'python', | |
537 | } |
|
537 | } | |
538 | msg = self.session.send(stream, 'kernel_info_reply', |
|
538 | msg = self.session.send(stream, 'kernel_info_reply', | |
539 | vinfo, parent, ident) |
|
539 | vinfo, parent, ident) | |
540 | self.log.debug("%s", msg) |
|
540 | self.log.debug("%s", msg) | |
541 |
|
541 | |||
542 | def shutdown_request(self, stream, ident, parent): |
|
542 | def shutdown_request(self, stream, ident, parent): | |
543 | self.shell.exit_now = True |
|
543 | self.shell.exit_now = True | |
544 | content = dict(status='ok') |
|
544 | content = dict(status='ok') | |
545 | content.update(parent['content']) |
|
545 | content.update(parent['content']) | |
546 | self.session.send(stream, u'shutdown_reply', content, parent, ident=ident) |
|
546 | self.session.send(stream, u'shutdown_reply', content, parent, ident=ident) | |
547 | # same content, but different msg_id for broadcasting on IOPub |
|
547 | # same content, but different msg_id for broadcasting on IOPub | |
548 | self._shutdown_message = self.session.msg(u'shutdown_reply', |
|
548 | self._shutdown_message = self.session.msg(u'shutdown_reply', | |
549 | content, parent |
|
549 | content, parent | |
550 | ) |
|
550 | ) | |
551 |
|
551 | |||
552 | self._at_shutdown() |
|
552 | self._at_shutdown() | |
553 | # call sys.exit after a short delay |
|
553 | # call sys.exit after a short delay | |
554 | loop = ioloop.IOLoop.instance() |
|
554 | loop = ioloop.IOLoop.instance() | |
555 | loop.add_timeout(time.time()+0.1, loop.stop) |
|
555 | loop.add_timeout(time.time()+0.1, loop.stop) | |
556 |
|
556 | |||
557 | #--------------------------------------------------------------------------- |
|
557 | #--------------------------------------------------------------------------- | |
558 | # Engine methods |
|
558 | # Engine methods | |
559 | #--------------------------------------------------------------------------- |
|
559 | #--------------------------------------------------------------------------- | |
560 |
|
560 | |||
561 | def apply_request(self, stream, ident, parent): |
|
561 | def apply_request(self, stream, ident, parent): | |
562 | try: |
|
562 | try: | |
563 | content = parent[u'content'] |
|
563 | content = parent[u'content'] | |
564 | bufs = parent[u'buffers'] |
|
564 | bufs = parent[u'buffers'] | |
565 | msg_id = parent['header']['msg_id'] |
|
565 | msg_id = parent['header']['msg_id'] | |
566 | except: |
|
566 | except: | |
567 | self.log.error("Got bad msg: %s", parent, exc_info=True) |
|
567 | self.log.error("Got bad msg: %s", parent, exc_info=True) | |
568 | return |
|
568 | return | |
569 |
|
569 | |||
570 | self._publish_status(u'busy', parent) |
|
570 | self._publish_status(u'busy', parent) | |
571 |
|
571 | |||
572 | # Set the parent message of the display hook and out streams. |
|
572 | # Set the parent message of the display hook and out streams. | |
573 | shell = self.shell |
|
573 | shell = self.shell | |
574 | shell.set_parent(parent) |
|
574 | shell.set_parent(parent) | |
575 |
|
575 | |||
576 | # pyin_msg = self.session.msg(u'pyin',{u'code':code}, parent=parent) |
|
576 | # pyin_msg = self.session.msg(u'pyin',{u'code':code}, parent=parent) | |
577 | # self.iopub_socket.send(pyin_msg) |
|
577 | # self.iopub_socket.send(pyin_msg) | |
578 | # self.session.send(self.iopub_socket, u'pyin', {u'code':code},parent=parent) |
|
578 | # self.session.send(self.iopub_socket, u'pyin', {u'code':code},parent=parent) | |
579 | md = self._make_metadata(parent['metadata']) |
|
579 | md = self._make_metadata(parent['metadata']) | |
580 | try: |
|
580 | try: | |
581 | working = shell.user_ns |
|
581 | working = shell.user_ns | |
582 |
|
582 | |||
583 | prefix = "_"+str(msg_id).replace("-","")+"_" |
|
583 | prefix = "_"+str(msg_id).replace("-","")+"_" | |
584 |
|
584 | |||
585 | f,args,kwargs = unpack_apply_message(bufs, working, copy=False) |
|
585 | f,args,kwargs = unpack_apply_message(bufs, working, copy=False) | |
586 |
|
586 | |||
587 | fname = getattr(f, '__name__', 'f') |
|
587 | fname = getattr(f, '__name__', 'f') | |
588 |
|
588 | |||
589 | fname = prefix+"f" |
|
589 | fname = prefix+"f" | |
590 | argname = prefix+"args" |
|
590 | argname = prefix+"args" | |
591 | kwargname = prefix+"kwargs" |
|
591 | kwargname = prefix+"kwargs" | |
592 | resultname = prefix+"result" |
|
592 | resultname = prefix+"result" | |
593 |
|
593 | |||
594 | ns = { fname : f, argname : args, kwargname : kwargs , resultname : None } |
|
594 | ns = { fname : f, argname : args, kwargname : kwargs , resultname : None } | |
595 | # print ns |
|
595 | # print ns | |
596 | working.update(ns) |
|
596 | working.update(ns) | |
597 | code = "%s = %s(*%s,**%s)" % (resultname, fname, argname, kwargname) |
|
597 | code = "%s = %s(*%s,**%s)" % (resultname, fname, argname, kwargname) | |
598 | try: |
|
598 | try: | |
599 | exec(code, shell.user_global_ns, shell.user_ns) |
|
599 | exec(code, shell.user_global_ns, shell.user_ns) | |
600 | result = working.get(resultname) |
|
600 | result = working.get(resultname) | |
601 | finally: |
|
601 | finally: | |
602 |
for key in ns |
|
602 | for key in ns: | |
603 | working.pop(key) |
|
603 | working.pop(key) | |
604 |
|
604 | |||
605 | result_buf = serialize_object(result, |
|
605 | result_buf = serialize_object(result, | |
606 | buffer_threshold=self.session.buffer_threshold, |
|
606 | buffer_threshold=self.session.buffer_threshold, | |
607 | item_threshold=self.session.item_threshold, |
|
607 | item_threshold=self.session.item_threshold, | |
608 | ) |
|
608 | ) | |
609 |
|
609 | |||
610 | except: |
|
610 | except: | |
611 | # invoke IPython traceback formatting |
|
611 | # invoke IPython traceback formatting | |
612 | shell.showtraceback() |
|
612 | shell.showtraceback() | |
613 | # FIXME - fish exception info out of shell, possibly left there by |
|
613 | # FIXME - fish exception info out of shell, possibly left there by | |
614 | # run_code. We'll need to clean up this logic later. |
|
614 | # run_code. We'll need to clean up this logic later. | |
615 | reply_content = {} |
|
615 | reply_content = {} | |
616 | if shell._reply_content is not None: |
|
616 | if shell._reply_content is not None: | |
617 | reply_content.update(shell._reply_content) |
|
617 | reply_content.update(shell._reply_content) | |
618 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method='apply') |
|
618 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method='apply') | |
619 | reply_content['engine_info'] = e_info |
|
619 | reply_content['engine_info'] = e_info | |
620 | # reset after use |
|
620 | # reset after use | |
621 | shell._reply_content = None |
|
621 | shell._reply_content = None | |
622 |
|
622 | |||
623 | self.session.send(self.iopub_socket, u'pyerr', reply_content, parent=parent, |
|
623 | self.session.send(self.iopub_socket, u'pyerr', reply_content, parent=parent, | |
624 | ident=self._topic('pyerr')) |
|
624 | ident=self._topic('pyerr')) | |
625 | self.log.info("Exception in apply request:\n%s", '\n'.join(reply_content['traceback'])) |
|
625 | self.log.info("Exception in apply request:\n%s", '\n'.join(reply_content['traceback'])) | |
626 | result_buf = [] |
|
626 | result_buf = [] | |
627 |
|
627 | |||
628 | if reply_content['ename'] == 'UnmetDependency': |
|
628 | if reply_content['ename'] == 'UnmetDependency': | |
629 | md['dependencies_met'] = False |
|
629 | md['dependencies_met'] = False | |
630 | else: |
|
630 | else: | |
631 | reply_content = {'status' : 'ok'} |
|
631 | reply_content = {'status' : 'ok'} | |
632 |
|
632 | |||
633 | # put 'ok'/'error' status in header, for scheduler introspection: |
|
633 | # put 'ok'/'error' status in header, for scheduler introspection: | |
634 | md['status'] = reply_content['status'] |
|
634 | md['status'] = reply_content['status'] | |
635 |
|
635 | |||
636 | # flush i/o |
|
636 | # flush i/o | |
637 | sys.stdout.flush() |
|
637 | sys.stdout.flush() | |
638 | sys.stderr.flush() |
|
638 | sys.stderr.flush() | |
639 |
|
639 | |||
640 | reply_msg = self.session.send(stream, u'apply_reply', reply_content, |
|
640 | reply_msg = self.session.send(stream, u'apply_reply', reply_content, | |
641 | parent=parent, ident=ident,buffers=result_buf, metadata=md) |
|
641 | parent=parent, ident=ident,buffers=result_buf, metadata=md) | |
642 |
|
642 | |||
643 | self._publish_status(u'idle', parent) |
|
643 | self._publish_status(u'idle', parent) | |
644 |
|
644 | |||
645 | #--------------------------------------------------------------------------- |
|
645 | #--------------------------------------------------------------------------- | |
646 | # Control messages |
|
646 | # Control messages | |
647 | #--------------------------------------------------------------------------- |
|
647 | #--------------------------------------------------------------------------- | |
648 |
|
648 | |||
649 | def abort_request(self, stream, ident, parent): |
|
649 | def abort_request(self, stream, ident, parent): | |
650 | """abort a specifig msg by id""" |
|
650 | """abort a specifig msg by id""" | |
651 | msg_ids = parent['content'].get('msg_ids', None) |
|
651 | msg_ids = parent['content'].get('msg_ids', None) | |
652 | if isinstance(msg_ids, string_types): |
|
652 | if isinstance(msg_ids, string_types): | |
653 | msg_ids = [msg_ids] |
|
653 | msg_ids = [msg_ids] | |
654 | if not msg_ids: |
|
654 | if not msg_ids: | |
655 | self.abort_queues() |
|
655 | self.abort_queues() | |
656 | for mid in msg_ids: |
|
656 | for mid in msg_ids: | |
657 | self.aborted.add(str(mid)) |
|
657 | self.aborted.add(str(mid)) | |
658 |
|
658 | |||
659 | content = dict(status='ok') |
|
659 | content = dict(status='ok') | |
660 | reply_msg = self.session.send(stream, 'abort_reply', content=content, |
|
660 | reply_msg = self.session.send(stream, 'abort_reply', content=content, | |
661 | parent=parent, ident=ident) |
|
661 | parent=parent, ident=ident) | |
662 | self.log.debug("%s", reply_msg) |
|
662 | self.log.debug("%s", reply_msg) | |
663 |
|
663 | |||
664 | def clear_request(self, stream, idents, parent): |
|
664 | def clear_request(self, stream, idents, parent): | |
665 | """Clear our namespace.""" |
|
665 | """Clear our namespace.""" | |
666 | self.shell.reset(False) |
|
666 | self.shell.reset(False) | |
667 | msg = self.session.send(stream, 'clear_reply', ident=idents, parent=parent, |
|
667 | msg = self.session.send(stream, 'clear_reply', ident=idents, parent=parent, | |
668 | content = dict(status='ok')) |
|
668 | content = dict(status='ok')) | |
669 |
|
669 | |||
670 |
|
670 | |||
671 | #--------------------------------------------------------------------------- |
|
671 | #--------------------------------------------------------------------------- | |
672 | # Protected interface |
|
672 | # Protected interface | |
673 | #--------------------------------------------------------------------------- |
|
673 | #--------------------------------------------------------------------------- | |
674 |
|
674 | |||
675 | def _wrap_exception(self, method=None): |
|
675 | def _wrap_exception(self, method=None): | |
676 | # import here, because _wrap_exception is only used in parallel, |
|
676 | # import here, because _wrap_exception is only used in parallel, | |
677 | # and parallel has higher min pyzmq version |
|
677 | # and parallel has higher min pyzmq version | |
678 | from IPython.parallel.error import wrap_exception |
|
678 | from IPython.parallel.error import wrap_exception | |
679 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method=method) |
|
679 | e_info = dict(engine_uuid=self.ident, engine_id=self.int_id, method=method) | |
680 | content = wrap_exception(e_info) |
|
680 | content = wrap_exception(e_info) | |
681 | return content |
|
681 | return content | |
682 |
|
682 | |||
683 | def _topic(self, topic): |
|
683 | def _topic(self, topic): | |
684 | """prefixed topic for IOPub messages""" |
|
684 | """prefixed topic for IOPub messages""" | |
685 | if self.int_id >= 0: |
|
685 | if self.int_id >= 0: | |
686 | base = "engine.%i" % self.int_id |
|
686 | base = "engine.%i" % self.int_id | |
687 | else: |
|
687 | else: | |
688 | base = "kernel.%s" % self.ident |
|
688 | base = "kernel.%s" % self.ident | |
689 |
|
689 | |||
690 | return py3compat.cast_bytes("%s.%s" % (base, topic)) |
|
690 | return py3compat.cast_bytes("%s.%s" % (base, topic)) | |
691 |
|
691 | |||
692 | def _abort_queues(self): |
|
692 | def _abort_queues(self): | |
693 | for stream in self.shell_streams: |
|
693 | for stream in self.shell_streams: | |
694 | if stream: |
|
694 | if stream: | |
695 | self._abort_queue(stream) |
|
695 | self._abort_queue(stream) | |
696 |
|
696 | |||
697 | def _abort_queue(self, stream): |
|
697 | def _abort_queue(self, stream): | |
698 | poller = zmq.Poller() |
|
698 | poller = zmq.Poller() | |
699 | poller.register(stream.socket, zmq.POLLIN) |
|
699 | poller.register(stream.socket, zmq.POLLIN) | |
700 | while True: |
|
700 | while True: | |
701 | idents,msg = self.session.recv(stream, zmq.NOBLOCK, content=True) |
|
701 | idents,msg = self.session.recv(stream, zmq.NOBLOCK, content=True) | |
702 | if msg is None: |
|
702 | if msg is None: | |
703 | return |
|
703 | return | |
704 |
|
704 | |||
705 | self.log.info("Aborting:") |
|
705 | self.log.info("Aborting:") | |
706 | self.log.info("%s", msg) |
|
706 | self.log.info("%s", msg) | |
707 | msg_type = msg['header']['msg_type'] |
|
707 | msg_type = msg['header']['msg_type'] | |
708 | reply_type = msg_type.split('_')[0] + '_reply' |
|
708 | reply_type = msg_type.split('_')[0] + '_reply' | |
709 |
|
709 | |||
710 | status = {'status' : 'aborted'} |
|
710 | status = {'status' : 'aborted'} | |
711 | md = {'engine' : self.ident} |
|
711 | md = {'engine' : self.ident} | |
712 | md.update(status) |
|
712 | md.update(status) | |
713 | reply_msg = self.session.send(stream, reply_type, metadata=md, |
|
713 | reply_msg = self.session.send(stream, reply_type, metadata=md, | |
714 | content=status, parent=msg, ident=idents) |
|
714 | content=status, parent=msg, ident=idents) | |
715 | self.log.debug("%s", reply_msg) |
|
715 | self.log.debug("%s", reply_msg) | |
716 | # We need to wait a bit for requests to come in. This can probably |
|
716 | # We need to wait a bit for requests to come in. This can probably | |
717 | # be set shorter for true asynchronous clients. |
|
717 | # be set shorter for true asynchronous clients. | |
718 | poller.poll(50) |
|
718 | poller.poll(50) | |
719 |
|
719 | |||
720 |
|
720 | |||
721 | def _no_raw_input(self): |
|
721 | def _no_raw_input(self): | |
722 | """Raise StdinNotImplentedError if active frontend doesn't support |
|
722 | """Raise StdinNotImplentedError if active frontend doesn't support | |
723 | stdin.""" |
|
723 | stdin.""" | |
724 | raise StdinNotImplementedError("raw_input was called, but this " |
|
724 | raise StdinNotImplementedError("raw_input was called, but this " | |
725 | "frontend does not support stdin.") |
|
725 | "frontend does not support stdin.") | |
726 |
|
726 | |||
727 | def _raw_input(self, prompt, ident, parent): |
|
727 | def _raw_input(self, prompt, ident, parent): | |
728 | # Flush output before making the request. |
|
728 | # Flush output before making the request. | |
729 | sys.stderr.flush() |
|
729 | sys.stderr.flush() | |
730 | sys.stdout.flush() |
|
730 | sys.stdout.flush() | |
731 | # flush the stdin socket, to purge stale replies |
|
731 | # flush the stdin socket, to purge stale replies | |
732 | while True: |
|
732 | while True: | |
733 | try: |
|
733 | try: | |
734 | self.stdin_socket.recv_multipart(zmq.NOBLOCK) |
|
734 | self.stdin_socket.recv_multipart(zmq.NOBLOCK) | |
735 | except zmq.ZMQError as e: |
|
735 | except zmq.ZMQError as e: | |
736 | if e.errno == zmq.EAGAIN: |
|
736 | if e.errno == zmq.EAGAIN: | |
737 | break |
|
737 | break | |
738 | else: |
|
738 | else: | |
739 | raise |
|
739 | raise | |
740 |
|
740 | |||
741 | # Send the input request. |
|
741 | # Send the input request. | |
742 | content = json_clean(dict(prompt=prompt)) |
|
742 | content = json_clean(dict(prompt=prompt)) | |
743 | self.session.send(self.stdin_socket, u'input_request', content, parent, |
|
743 | self.session.send(self.stdin_socket, u'input_request', content, parent, | |
744 | ident=ident) |
|
744 | ident=ident) | |
745 |
|
745 | |||
746 | # Await a response. |
|
746 | # Await a response. | |
747 | while True: |
|
747 | while True: | |
748 | try: |
|
748 | try: | |
749 | ident, reply = self.session.recv(self.stdin_socket, 0) |
|
749 | ident, reply = self.session.recv(self.stdin_socket, 0) | |
750 | except Exception: |
|
750 | except Exception: | |
751 | self.log.warn("Invalid Message:", exc_info=True) |
|
751 | self.log.warn("Invalid Message:", exc_info=True) | |
752 | except KeyboardInterrupt: |
|
752 | except KeyboardInterrupt: | |
753 | # re-raise KeyboardInterrupt, to truncate traceback |
|
753 | # re-raise KeyboardInterrupt, to truncate traceback | |
754 | raise KeyboardInterrupt |
|
754 | raise KeyboardInterrupt | |
755 | else: |
|
755 | else: | |
756 | break |
|
756 | break | |
757 | try: |
|
757 | try: | |
758 | value = py3compat.unicode_to_str(reply['content']['value']) |
|
758 | value = py3compat.unicode_to_str(reply['content']['value']) | |
759 | except: |
|
759 | except: | |
760 | self.log.error("Got bad raw_input reply: ") |
|
760 | self.log.error("Got bad raw_input reply: ") | |
761 | self.log.error("%s", parent) |
|
761 | self.log.error("%s", parent) | |
762 | value = '' |
|
762 | value = '' | |
763 | if value == '\x04': |
|
763 | if value == '\x04': | |
764 | # EOF |
|
764 | # EOF | |
765 | raise EOFError |
|
765 | raise EOFError | |
766 | return value |
|
766 | return value | |
767 |
|
767 | |||
768 | def _complete(self, msg): |
|
768 | def _complete(self, msg): | |
769 | c = msg['content'] |
|
769 | c = msg['content'] | |
770 | try: |
|
770 | try: | |
771 | cpos = int(c['cursor_pos']) |
|
771 | cpos = int(c['cursor_pos']) | |
772 | except: |
|
772 | except: | |
773 | # If we don't get something that we can convert to an integer, at |
|
773 | # If we don't get something that we can convert to an integer, at | |
774 | # least attempt the completion guessing the cursor is at the end of |
|
774 | # least attempt the completion guessing the cursor is at the end of | |
775 | # the text, if there's any, and otherwise of the line |
|
775 | # the text, if there's any, and otherwise of the line | |
776 | cpos = len(c['text']) |
|
776 | cpos = len(c['text']) | |
777 | if cpos==0: |
|
777 | if cpos==0: | |
778 | cpos = len(c['line']) |
|
778 | cpos = len(c['line']) | |
779 | return self.shell.complete(c['text'], c['line'], cpos) |
|
779 | return self.shell.complete(c['text'], c['line'], cpos) | |
780 |
|
780 | |||
781 | def _at_shutdown(self): |
|
781 | def _at_shutdown(self): | |
782 | """Actions taken at shutdown by the kernel, called by python's atexit. |
|
782 | """Actions taken at shutdown by the kernel, called by python's atexit. | |
783 | """ |
|
783 | """ | |
784 | # io.rprint("Kernel at_shutdown") # dbg |
|
784 | # io.rprint("Kernel at_shutdown") # dbg | |
785 | if self._shutdown_message is not None: |
|
785 | if self._shutdown_message is not None: | |
786 | self.session.send(self.iopub_socket, self._shutdown_message, ident=self._topic('shutdown')) |
|
786 | self.session.send(self.iopub_socket, self._shutdown_message, ident=self._topic('shutdown')) | |
787 | self.log.debug("%s", self._shutdown_message) |
|
787 | self.log.debug("%s", self._shutdown_message) | |
788 | [ s.flush(zmq.POLLOUT) for s in self.shell_streams ] |
|
788 | [ s.flush(zmq.POLLOUT) for s in self.shell_streams ] | |
789 |
|
789 |
@@ -1,198 +1,198 b'' | |||||
1 | """serialization utilities for apply messages |
|
1 | """serialization utilities for apply messages | |
2 |
|
2 | |||
3 | Authors: |
|
3 | Authors: | |
4 |
|
4 | |||
5 | * Min RK |
|
5 | * Min RK | |
6 | """ |
|
6 | """ | |
7 | #----------------------------------------------------------------------------- |
|
7 | #----------------------------------------------------------------------------- | |
8 | # Copyright (C) 2010-2011 The IPython Development Team |
|
8 | # Copyright (C) 2010-2011 The IPython Development Team | |
9 | # |
|
9 | # | |
10 | # Distributed under the terms of the BSD License. The full license is in |
|
10 | # Distributed under the terms of the BSD License. The full license is in | |
11 | # the file COPYING, distributed as part of this software. |
|
11 | # the file COPYING, distributed as part of this software. | |
12 | #----------------------------------------------------------------------------- |
|
12 | #----------------------------------------------------------------------------- | |
13 |
|
13 | |||
14 | #----------------------------------------------------------------------------- |
|
14 | #----------------------------------------------------------------------------- | |
15 | # Imports |
|
15 | # Imports | |
16 | #----------------------------------------------------------------------------- |
|
16 | #----------------------------------------------------------------------------- | |
17 |
|
17 | |||
18 | try: |
|
18 | try: | |
19 | import cPickle |
|
19 | import cPickle | |
20 | pickle = cPickle |
|
20 | pickle = cPickle | |
21 | except: |
|
21 | except: | |
22 | cPickle = None |
|
22 | cPickle = None | |
23 | import pickle |
|
23 | import pickle | |
24 |
|
24 | |||
25 |
|
25 | |||
26 | # IPython imports |
|
26 | # IPython imports | |
27 | from IPython.utils import py3compat |
|
27 | from IPython.utils import py3compat | |
28 | from IPython.utils.data import flatten |
|
28 | from IPython.utils.data import flatten | |
29 | from IPython.utils.pickleutil import ( |
|
29 | from IPython.utils.pickleutil import ( | |
30 | can, uncan, can_sequence, uncan_sequence, CannedObject, |
|
30 | can, uncan, can_sequence, uncan_sequence, CannedObject, | |
31 | istype, sequence_types, |
|
31 | istype, sequence_types, | |
32 | ) |
|
32 | ) | |
33 |
|
33 | |||
34 | if py3compat.PY3: |
|
34 | if py3compat.PY3: | |
35 | buffer = memoryview |
|
35 | buffer = memoryview | |
36 |
|
36 | |||
37 | #----------------------------------------------------------------------------- |
|
37 | #----------------------------------------------------------------------------- | |
38 | # Serialization Functions |
|
38 | # Serialization Functions | |
39 | #----------------------------------------------------------------------------- |
|
39 | #----------------------------------------------------------------------------- | |
40 |
|
40 | |||
41 | # default values for the thresholds: |
|
41 | # default values for the thresholds: | |
42 | MAX_ITEMS = 64 |
|
42 | MAX_ITEMS = 64 | |
43 | MAX_BYTES = 1024 |
|
43 | MAX_BYTES = 1024 | |
44 |
|
44 | |||
45 | def _extract_buffers(obj, threshold=MAX_BYTES): |
|
45 | def _extract_buffers(obj, threshold=MAX_BYTES): | |
46 | """extract buffers larger than a certain threshold""" |
|
46 | """extract buffers larger than a certain threshold""" | |
47 | buffers = [] |
|
47 | buffers = [] | |
48 | if isinstance(obj, CannedObject) and obj.buffers: |
|
48 | if isinstance(obj, CannedObject) and obj.buffers: | |
49 | for i,buf in enumerate(obj.buffers): |
|
49 | for i,buf in enumerate(obj.buffers): | |
50 | if len(buf) > threshold: |
|
50 | if len(buf) > threshold: | |
51 | # buffer larger than threshold, prevent pickling |
|
51 | # buffer larger than threshold, prevent pickling | |
52 | obj.buffers[i] = None |
|
52 | obj.buffers[i] = None | |
53 | buffers.append(buf) |
|
53 | buffers.append(buf) | |
54 | elif isinstance(buf, buffer): |
|
54 | elif isinstance(buf, buffer): | |
55 | # buffer too small for separate send, coerce to bytes |
|
55 | # buffer too small for separate send, coerce to bytes | |
56 | # because pickling buffer objects just results in broken pointers |
|
56 | # because pickling buffer objects just results in broken pointers | |
57 | obj.buffers[i] = bytes(buf) |
|
57 | obj.buffers[i] = bytes(buf) | |
58 | return buffers |
|
58 | return buffers | |
59 |
|
59 | |||
60 | def _restore_buffers(obj, buffers): |
|
60 | def _restore_buffers(obj, buffers): | |
61 | """restore buffers extracted by """ |
|
61 | """restore buffers extracted by """ | |
62 | if isinstance(obj, CannedObject) and obj.buffers: |
|
62 | if isinstance(obj, CannedObject) and obj.buffers: | |
63 | for i,buf in enumerate(obj.buffers): |
|
63 | for i,buf in enumerate(obj.buffers): | |
64 | if buf is None: |
|
64 | if buf is None: | |
65 | obj.buffers[i] = buffers.pop(0) |
|
65 | obj.buffers[i] = buffers.pop(0) | |
66 |
|
66 | |||
67 | def serialize_object(obj, buffer_threshold=MAX_BYTES, item_threshold=MAX_ITEMS): |
|
67 | def serialize_object(obj, buffer_threshold=MAX_BYTES, item_threshold=MAX_ITEMS): | |
68 | """Serialize an object into a list of sendable buffers. |
|
68 | """Serialize an object into a list of sendable buffers. | |
69 |
|
69 | |||
70 | Parameters |
|
70 | Parameters | |
71 | ---------- |
|
71 | ---------- | |
72 |
|
72 | |||
73 | obj : object |
|
73 | obj : object | |
74 | The object to be serialized |
|
74 | The object to be serialized | |
75 | buffer_threshold : int |
|
75 | buffer_threshold : int | |
76 | The threshold (in bytes) for pulling out data buffers |
|
76 | The threshold (in bytes) for pulling out data buffers | |
77 | to avoid pickling them. |
|
77 | to avoid pickling them. | |
78 | item_threshold : int |
|
78 | item_threshold : int | |
79 | The maximum number of items over which canning will iterate. |
|
79 | The maximum number of items over which canning will iterate. | |
80 | Containers (lists, dicts) larger than this will be pickled without |
|
80 | Containers (lists, dicts) larger than this will be pickled without | |
81 | introspection. |
|
81 | introspection. | |
82 |
|
82 | |||
83 | Returns |
|
83 | Returns | |
84 | ------- |
|
84 | ------- | |
85 | [bufs] : list of buffers representing the serialized object. |
|
85 | [bufs] : list of buffers representing the serialized object. | |
86 | """ |
|
86 | """ | |
87 | buffers = [] |
|
87 | buffers = [] | |
88 | if istype(obj, sequence_types) and len(obj) < item_threshold: |
|
88 | if istype(obj, sequence_types) and len(obj) < item_threshold: | |
89 | cobj = can_sequence(obj) |
|
89 | cobj = can_sequence(obj) | |
90 | for c in cobj: |
|
90 | for c in cobj: | |
91 | buffers.extend(_extract_buffers(c, buffer_threshold)) |
|
91 | buffers.extend(_extract_buffers(c, buffer_threshold)) | |
92 | elif istype(obj, dict) and len(obj) < item_threshold: |
|
92 | elif istype(obj, dict) and len(obj) < item_threshold: | |
93 | cobj = {} |
|
93 | cobj = {} | |
94 |
for k in sorted(obj |
|
94 | for k in sorted(obj): | |
95 | c = can(obj[k]) |
|
95 | c = can(obj[k]) | |
96 | buffers.extend(_extract_buffers(c, buffer_threshold)) |
|
96 | buffers.extend(_extract_buffers(c, buffer_threshold)) | |
97 | cobj[k] = c |
|
97 | cobj[k] = c | |
98 | else: |
|
98 | else: | |
99 | cobj = can(obj) |
|
99 | cobj = can(obj) | |
100 | buffers.extend(_extract_buffers(cobj, buffer_threshold)) |
|
100 | buffers.extend(_extract_buffers(cobj, buffer_threshold)) | |
101 |
|
101 | |||
102 | buffers.insert(0, pickle.dumps(cobj,-1)) |
|
102 | buffers.insert(0, pickle.dumps(cobj,-1)) | |
103 | return buffers |
|
103 | return buffers | |
104 |
|
104 | |||
105 | def unserialize_object(buffers, g=None): |
|
105 | def unserialize_object(buffers, g=None): | |
106 | """reconstruct an object serialized by serialize_object from data buffers. |
|
106 | """reconstruct an object serialized by serialize_object from data buffers. | |
107 |
|
107 | |||
108 | Parameters |
|
108 | Parameters | |
109 | ---------- |
|
109 | ---------- | |
110 |
|
110 | |||
111 | bufs : list of buffers/bytes |
|
111 | bufs : list of buffers/bytes | |
112 |
|
112 | |||
113 | g : globals to be used when uncanning |
|
113 | g : globals to be used when uncanning | |
114 |
|
114 | |||
115 | Returns |
|
115 | Returns | |
116 | ------- |
|
116 | ------- | |
117 |
|
117 | |||
118 | (newobj, bufs) : unpacked object, and the list of remaining unused buffers. |
|
118 | (newobj, bufs) : unpacked object, and the list of remaining unused buffers. | |
119 | """ |
|
119 | """ | |
120 | bufs = list(buffers) |
|
120 | bufs = list(buffers) | |
121 | pobj = bufs.pop(0) |
|
121 | pobj = bufs.pop(0) | |
122 | if not isinstance(pobj, bytes): |
|
122 | if not isinstance(pobj, bytes): | |
123 | # a zmq message |
|
123 | # a zmq message | |
124 | pobj = bytes(pobj) |
|
124 | pobj = bytes(pobj) | |
125 | canned = pickle.loads(pobj) |
|
125 | canned = pickle.loads(pobj) | |
126 | if istype(canned, sequence_types) and len(canned) < MAX_ITEMS: |
|
126 | if istype(canned, sequence_types) and len(canned) < MAX_ITEMS: | |
127 | for c in canned: |
|
127 | for c in canned: | |
128 | _restore_buffers(c, bufs) |
|
128 | _restore_buffers(c, bufs) | |
129 | newobj = uncan_sequence(canned, g) |
|
129 | newobj = uncan_sequence(canned, g) | |
130 | elif istype(canned, dict) and len(canned) < MAX_ITEMS: |
|
130 | elif istype(canned, dict) and len(canned) < MAX_ITEMS: | |
131 | newobj = {} |
|
131 | newobj = {} | |
132 |
for k in sorted(canned |
|
132 | for k in sorted(canned): | |
133 | c = canned[k] |
|
133 | c = canned[k] | |
134 | _restore_buffers(c, bufs) |
|
134 | _restore_buffers(c, bufs) | |
135 | newobj[k] = uncan(c, g) |
|
135 | newobj[k] = uncan(c, g) | |
136 | else: |
|
136 | else: | |
137 | _restore_buffers(canned, bufs) |
|
137 | _restore_buffers(canned, bufs) | |
138 | newobj = uncan(canned, g) |
|
138 | newobj = uncan(canned, g) | |
139 |
|
139 | |||
140 | return newobj, bufs |
|
140 | return newobj, bufs | |
141 |
|
141 | |||
142 | def pack_apply_message(f, args, kwargs, buffer_threshold=MAX_BYTES, item_threshold=MAX_ITEMS): |
|
142 | def pack_apply_message(f, args, kwargs, buffer_threshold=MAX_BYTES, item_threshold=MAX_ITEMS): | |
143 | """pack up a function, args, and kwargs to be sent over the wire |
|
143 | """pack up a function, args, and kwargs to be sent over the wire | |
144 |
|
144 | |||
145 | Each element of args/kwargs will be canned for special treatment, |
|
145 | Each element of args/kwargs will be canned for special treatment, | |
146 | but inspection will not go any deeper than that. |
|
146 | but inspection will not go any deeper than that. | |
147 |
|
147 | |||
148 | Any object whose data is larger than `threshold` will not have their data copied |
|
148 | Any object whose data is larger than `threshold` will not have their data copied | |
149 | (only numpy arrays and bytes/buffers support zero-copy) |
|
149 | (only numpy arrays and bytes/buffers support zero-copy) | |
150 |
|
150 | |||
151 | Message will be a list of bytes/buffers of the format: |
|
151 | Message will be a list of bytes/buffers of the format: | |
152 |
|
152 | |||
153 | [ cf, pinfo, <arg_bufs>, <kwarg_bufs> ] |
|
153 | [ cf, pinfo, <arg_bufs>, <kwarg_bufs> ] | |
154 |
|
154 | |||
155 | With length at least two + len(args) + len(kwargs) |
|
155 | With length at least two + len(args) + len(kwargs) | |
156 | """ |
|
156 | """ | |
157 |
|
157 | |||
158 | arg_bufs = flatten(serialize_object(arg, buffer_threshold, item_threshold) for arg in args) |
|
158 | arg_bufs = flatten(serialize_object(arg, buffer_threshold, item_threshold) for arg in args) | |
159 |
|
159 | |||
160 | kw_keys = sorted(kwargs.keys()) |
|
160 | kw_keys = sorted(kwargs.keys()) | |
161 | kwarg_bufs = flatten(serialize_object(kwargs[key], buffer_threshold, item_threshold) for key in kw_keys) |
|
161 | kwarg_bufs = flatten(serialize_object(kwargs[key], buffer_threshold, item_threshold) for key in kw_keys) | |
162 |
|
162 | |||
163 | info = dict(nargs=len(args), narg_bufs=len(arg_bufs), kw_keys=kw_keys) |
|
163 | info = dict(nargs=len(args), narg_bufs=len(arg_bufs), kw_keys=kw_keys) | |
164 |
|
164 | |||
165 | msg = [pickle.dumps(can(f),-1)] |
|
165 | msg = [pickle.dumps(can(f),-1)] | |
166 | msg.append(pickle.dumps(info, -1)) |
|
166 | msg.append(pickle.dumps(info, -1)) | |
167 | msg.extend(arg_bufs) |
|
167 | msg.extend(arg_bufs) | |
168 | msg.extend(kwarg_bufs) |
|
168 | msg.extend(kwarg_bufs) | |
169 |
|
169 | |||
170 | return msg |
|
170 | return msg | |
171 |
|
171 | |||
172 | def unpack_apply_message(bufs, g=None, copy=True): |
|
172 | def unpack_apply_message(bufs, g=None, copy=True): | |
173 | """unpack f,args,kwargs from buffers packed by pack_apply_message() |
|
173 | """unpack f,args,kwargs from buffers packed by pack_apply_message() | |
174 | Returns: original f,args,kwargs""" |
|
174 | Returns: original f,args,kwargs""" | |
175 | bufs = list(bufs) # allow us to pop |
|
175 | bufs = list(bufs) # allow us to pop | |
176 | assert len(bufs) >= 2, "not enough buffers!" |
|
176 | assert len(bufs) >= 2, "not enough buffers!" | |
177 | if not copy: |
|
177 | if not copy: | |
178 | for i in range(2): |
|
178 | for i in range(2): | |
179 | bufs[i] = bufs[i].bytes |
|
179 | bufs[i] = bufs[i].bytes | |
180 | f = uncan(pickle.loads(bufs.pop(0)), g) |
|
180 | f = uncan(pickle.loads(bufs.pop(0)), g) | |
181 | info = pickle.loads(bufs.pop(0)) |
|
181 | info = pickle.loads(bufs.pop(0)) | |
182 | arg_bufs, kwarg_bufs = bufs[:info['narg_bufs']], bufs[info['narg_bufs']:] |
|
182 | arg_bufs, kwarg_bufs = bufs[:info['narg_bufs']], bufs[info['narg_bufs']:] | |
183 |
|
183 | |||
184 | args = [] |
|
184 | args = [] | |
185 | for i in range(info['nargs']): |
|
185 | for i in range(info['nargs']): | |
186 | arg, arg_bufs = unserialize_object(arg_bufs, g) |
|
186 | arg, arg_bufs = unserialize_object(arg_bufs, g) | |
187 | args.append(arg) |
|
187 | args.append(arg) | |
188 | args = tuple(args) |
|
188 | args = tuple(args) | |
189 | assert not arg_bufs, "Shouldn't be any arg bufs left over" |
|
189 | assert not arg_bufs, "Shouldn't be any arg bufs left over" | |
190 |
|
190 | |||
191 | kwargs = {} |
|
191 | kwargs = {} | |
192 | for key in info['kw_keys']: |
|
192 | for key in info['kw_keys']: | |
193 | kwarg, kwarg_bufs = unserialize_object(kwarg_bufs, g) |
|
193 | kwarg, kwarg_bufs = unserialize_object(kwarg_bufs, g) | |
194 | kwargs[key] = kwarg |
|
194 | kwargs[key] = kwarg | |
195 | assert not kwarg_bufs, "Shouldn't be any kwarg bufs left over" |
|
195 | assert not kwarg_bufs, "Shouldn't be any kwarg bufs left over" | |
196 |
|
196 | |||
197 | return f,args,kwargs |
|
197 | return f,args,kwargs | |
198 |
|
198 |
@@ -1,1854 +1,1854 b'' | |||||
1 | """A semi-synchronous Client for the ZMQ cluster |
|
1 | """A semi-synchronous Client for the ZMQ cluster | |
2 |
|
2 | |||
3 | Authors: |
|
3 | Authors: | |
4 |
|
4 | |||
5 | * MinRK |
|
5 | * MinRK | |
6 | """ |
|
6 | """ | |
7 | from __future__ import print_function |
|
7 | from __future__ import print_function | |
8 | #----------------------------------------------------------------------------- |
|
8 | #----------------------------------------------------------------------------- | |
9 | # Copyright (C) 2010-2011 The IPython Development Team |
|
9 | # Copyright (C) 2010-2011 The IPython Development Team | |
10 | # |
|
10 | # | |
11 | # Distributed under the terms of the BSD License. The full license is in |
|
11 | # Distributed under the terms of the BSD License. The full license is in | |
12 | # the file COPYING, distributed as part of this software. |
|
12 | # the file COPYING, distributed as part of this software. | |
13 | #----------------------------------------------------------------------------- |
|
13 | #----------------------------------------------------------------------------- | |
14 |
|
14 | |||
15 | #----------------------------------------------------------------------------- |
|
15 | #----------------------------------------------------------------------------- | |
16 | # Imports |
|
16 | # Imports | |
17 | #----------------------------------------------------------------------------- |
|
17 | #----------------------------------------------------------------------------- | |
18 |
|
18 | |||
19 | import os |
|
19 | import os | |
20 | import json |
|
20 | import json | |
21 | import sys |
|
21 | import sys | |
22 | from threading import Thread, Event |
|
22 | from threading import Thread, Event | |
23 | import time |
|
23 | import time | |
24 | import warnings |
|
24 | import warnings | |
25 | from datetime import datetime |
|
25 | from datetime import datetime | |
26 | from getpass import getpass |
|
26 | from getpass import getpass | |
27 | from pprint import pprint |
|
27 | from pprint import pprint | |
28 |
|
28 | |||
29 | pjoin = os.path.join |
|
29 | pjoin = os.path.join | |
30 |
|
30 | |||
31 | import zmq |
|
31 | import zmq | |
32 | # from zmq.eventloop import ioloop, zmqstream |
|
32 | # from zmq.eventloop import ioloop, zmqstream | |
33 |
|
33 | |||
34 | from IPython.config.configurable import MultipleInstanceError |
|
34 | from IPython.config.configurable import MultipleInstanceError | |
35 | from IPython.core.application import BaseIPythonApplication |
|
35 | from IPython.core.application import BaseIPythonApplication | |
36 | from IPython.core.profiledir import ProfileDir, ProfileDirError |
|
36 | from IPython.core.profiledir import ProfileDir, ProfileDirError | |
37 |
|
37 | |||
38 | from IPython.utils.capture import RichOutput |
|
38 | from IPython.utils.capture import RichOutput | |
39 | from IPython.utils.coloransi import TermColors |
|
39 | from IPython.utils.coloransi import TermColors | |
40 | from IPython.utils.jsonutil import rekey |
|
40 | from IPython.utils.jsonutil import rekey | |
41 | from IPython.utils.localinterfaces import localhost, is_local_ip |
|
41 | from IPython.utils.localinterfaces import localhost, is_local_ip | |
42 | from IPython.utils.path import get_ipython_dir |
|
42 | from IPython.utils.path import get_ipython_dir | |
43 | from IPython.utils.py3compat import cast_bytes, string_types, xrange |
|
43 | from IPython.utils.py3compat import cast_bytes, string_types, xrange | |
44 | from IPython.utils.traitlets import (HasTraits, Integer, Instance, Unicode, |
|
44 | from IPython.utils.traitlets import (HasTraits, Integer, Instance, Unicode, | |
45 | Dict, List, Bool, Set, Any) |
|
45 | Dict, List, Bool, Set, Any) | |
46 | from IPython.external.decorator import decorator |
|
46 | from IPython.external.decorator import decorator | |
47 | from IPython.external.ssh import tunnel |
|
47 | from IPython.external.ssh import tunnel | |
48 |
|
48 | |||
49 | from IPython.parallel import Reference |
|
49 | from IPython.parallel import Reference | |
50 | from IPython.parallel import error |
|
50 | from IPython.parallel import error | |
51 | from IPython.parallel import util |
|
51 | from IPython.parallel import util | |
52 |
|
52 | |||
53 | from IPython.kernel.zmq.session import Session, Message |
|
53 | from IPython.kernel.zmq.session import Session, Message | |
54 | from IPython.kernel.zmq import serialize |
|
54 | from IPython.kernel.zmq import serialize | |
55 |
|
55 | |||
56 | from .asyncresult import AsyncResult, AsyncHubResult |
|
56 | from .asyncresult import AsyncResult, AsyncHubResult | |
57 | from .view import DirectView, LoadBalancedView |
|
57 | from .view import DirectView, LoadBalancedView | |
58 |
|
58 | |||
59 | #-------------------------------------------------------------------------- |
|
59 | #-------------------------------------------------------------------------- | |
60 | # Decorators for Client methods |
|
60 | # Decorators for Client methods | |
61 | #-------------------------------------------------------------------------- |
|
61 | #-------------------------------------------------------------------------- | |
62 |
|
62 | |||
63 | @decorator |
|
63 | @decorator | |
64 | def spin_first(f, self, *args, **kwargs): |
|
64 | def spin_first(f, self, *args, **kwargs): | |
65 | """Call spin() to sync state prior to calling the method.""" |
|
65 | """Call spin() to sync state prior to calling the method.""" | |
66 | self.spin() |
|
66 | self.spin() | |
67 | return f(self, *args, **kwargs) |
|
67 | return f(self, *args, **kwargs) | |
68 |
|
68 | |||
69 |
|
69 | |||
70 | #-------------------------------------------------------------------------- |
|
70 | #-------------------------------------------------------------------------- | |
71 | # Classes |
|
71 | # Classes | |
72 | #-------------------------------------------------------------------------- |
|
72 | #-------------------------------------------------------------------------- | |
73 |
|
73 | |||
74 |
|
74 | |||
75 | class ExecuteReply(RichOutput): |
|
75 | class ExecuteReply(RichOutput): | |
76 | """wrapper for finished Execute results""" |
|
76 | """wrapper for finished Execute results""" | |
77 | def __init__(self, msg_id, content, metadata): |
|
77 | def __init__(self, msg_id, content, metadata): | |
78 | self.msg_id = msg_id |
|
78 | self.msg_id = msg_id | |
79 | self._content = content |
|
79 | self._content = content | |
80 | self.execution_count = content['execution_count'] |
|
80 | self.execution_count = content['execution_count'] | |
81 | self.metadata = metadata |
|
81 | self.metadata = metadata | |
82 |
|
82 | |||
83 | # RichOutput overrides |
|
83 | # RichOutput overrides | |
84 |
|
84 | |||
85 | @property |
|
85 | @property | |
86 | def source(self): |
|
86 | def source(self): | |
87 | pyout = self.metadata['pyout'] |
|
87 | pyout = self.metadata['pyout'] | |
88 | if pyout: |
|
88 | if pyout: | |
89 | return pyout.get('source', '') |
|
89 | return pyout.get('source', '') | |
90 |
|
90 | |||
91 | @property |
|
91 | @property | |
92 | def data(self): |
|
92 | def data(self): | |
93 | pyout = self.metadata['pyout'] |
|
93 | pyout = self.metadata['pyout'] | |
94 | if pyout: |
|
94 | if pyout: | |
95 | return pyout.get('data', {}) |
|
95 | return pyout.get('data', {}) | |
96 |
|
96 | |||
97 | @property |
|
97 | @property | |
98 | def _metadata(self): |
|
98 | def _metadata(self): | |
99 | pyout = self.metadata['pyout'] |
|
99 | pyout = self.metadata['pyout'] | |
100 | if pyout: |
|
100 | if pyout: | |
101 | return pyout.get('metadata', {}) |
|
101 | return pyout.get('metadata', {}) | |
102 |
|
102 | |||
103 | def display(self): |
|
103 | def display(self): | |
104 | from IPython.display import publish_display_data |
|
104 | from IPython.display import publish_display_data | |
105 | publish_display_data(self.source, self.data, self.metadata) |
|
105 | publish_display_data(self.source, self.data, self.metadata) | |
106 |
|
106 | |||
107 | def _repr_mime_(self, mime): |
|
107 | def _repr_mime_(self, mime): | |
108 | if mime not in self.data: |
|
108 | if mime not in self.data: | |
109 | return |
|
109 | return | |
110 | data = self.data[mime] |
|
110 | data = self.data[mime] | |
111 | if mime in self._metadata: |
|
111 | if mime in self._metadata: | |
112 | return data, self._metadata[mime] |
|
112 | return data, self._metadata[mime] | |
113 | else: |
|
113 | else: | |
114 | return data |
|
114 | return data | |
115 |
|
115 | |||
116 | def __getitem__(self, key): |
|
116 | def __getitem__(self, key): | |
117 | return self.metadata[key] |
|
117 | return self.metadata[key] | |
118 |
|
118 | |||
119 | def __getattr__(self, key): |
|
119 | def __getattr__(self, key): | |
120 | if key not in self.metadata: |
|
120 | if key not in self.metadata: | |
121 | raise AttributeError(key) |
|
121 | raise AttributeError(key) | |
122 | return self.metadata[key] |
|
122 | return self.metadata[key] | |
123 |
|
123 | |||
124 | def __repr__(self): |
|
124 | def __repr__(self): | |
125 | pyout = self.metadata['pyout'] or {'data':{}} |
|
125 | pyout = self.metadata['pyout'] or {'data':{}} | |
126 | text_out = pyout['data'].get('text/plain', '') |
|
126 | text_out = pyout['data'].get('text/plain', '') | |
127 | if len(text_out) > 32: |
|
127 | if len(text_out) > 32: | |
128 | text_out = text_out[:29] + '...' |
|
128 | text_out = text_out[:29] + '...' | |
129 |
|
129 | |||
130 | return "<ExecuteReply[%i]: %s>" % (self.execution_count, text_out) |
|
130 | return "<ExecuteReply[%i]: %s>" % (self.execution_count, text_out) | |
131 |
|
131 | |||
132 | def _repr_pretty_(self, p, cycle): |
|
132 | def _repr_pretty_(self, p, cycle): | |
133 | pyout = self.metadata['pyout'] or {'data':{}} |
|
133 | pyout = self.metadata['pyout'] or {'data':{}} | |
134 | text_out = pyout['data'].get('text/plain', '') |
|
134 | text_out = pyout['data'].get('text/plain', '') | |
135 |
|
135 | |||
136 | if not text_out: |
|
136 | if not text_out: | |
137 | return |
|
137 | return | |
138 |
|
138 | |||
139 | try: |
|
139 | try: | |
140 | ip = get_ipython() |
|
140 | ip = get_ipython() | |
141 | except NameError: |
|
141 | except NameError: | |
142 | colors = "NoColor" |
|
142 | colors = "NoColor" | |
143 | else: |
|
143 | else: | |
144 | colors = ip.colors |
|
144 | colors = ip.colors | |
145 |
|
145 | |||
146 | if colors == "NoColor": |
|
146 | if colors == "NoColor": | |
147 | out = normal = "" |
|
147 | out = normal = "" | |
148 | else: |
|
148 | else: | |
149 | out = TermColors.Red |
|
149 | out = TermColors.Red | |
150 | normal = TermColors.Normal |
|
150 | normal = TermColors.Normal | |
151 |
|
151 | |||
152 | if '\n' in text_out and not text_out.startswith('\n'): |
|
152 | if '\n' in text_out and not text_out.startswith('\n'): | |
153 | # add newline for multiline reprs |
|
153 | # add newline for multiline reprs | |
154 | text_out = '\n' + text_out |
|
154 | text_out = '\n' + text_out | |
155 |
|
155 | |||
156 | p.text( |
|
156 | p.text( | |
157 | out + u'Out[%i:%i]: ' % ( |
|
157 | out + u'Out[%i:%i]: ' % ( | |
158 | self.metadata['engine_id'], self.execution_count |
|
158 | self.metadata['engine_id'], self.execution_count | |
159 | ) + normal + text_out |
|
159 | ) + normal + text_out | |
160 | ) |
|
160 | ) | |
161 |
|
161 | |||
162 |
|
162 | |||
163 | class Metadata(dict): |
|
163 | class Metadata(dict): | |
164 | """Subclass of dict for initializing metadata values. |
|
164 | """Subclass of dict for initializing metadata values. | |
165 |
|
165 | |||
166 | Attribute access works on keys. |
|
166 | Attribute access works on keys. | |
167 |
|
167 | |||
168 | These objects have a strict set of keys - errors will raise if you try |
|
168 | These objects have a strict set of keys - errors will raise if you try | |
169 | to add new keys. |
|
169 | to add new keys. | |
170 | """ |
|
170 | """ | |
171 | def __init__(self, *args, **kwargs): |
|
171 | def __init__(self, *args, **kwargs): | |
172 | dict.__init__(self) |
|
172 | dict.__init__(self) | |
173 | md = {'msg_id' : None, |
|
173 | md = {'msg_id' : None, | |
174 | 'submitted' : None, |
|
174 | 'submitted' : None, | |
175 | 'started' : None, |
|
175 | 'started' : None, | |
176 | 'completed' : None, |
|
176 | 'completed' : None, | |
177 | 'received' : None, |
|
177 | 'received' : None, | |
178 | 'engine_uuid' : None, |
|
178 | 'engine_uuid' : None, | |
179 | 'engine_id' : None, |
|
179 | 'engine_id' : None, | |
180 | 'follow' : None, |
|
180 | 'follow' : None, | |
181 | 'after' : None, |
|
181 | 'after' : None, | |
182 | 'status' : None, |
|
182 | 'status' : None, | |
183 |
|
183 | |||
184 | 'pyin' : None, |
|
184 | 'pyin' : None, | |
185 | 'pyout' : None, |
|
185 | 'pyout' : None, | |
186 | 'pyerr' : None, |
|
186 | 'pyerr' : None, | |
187 | 'stdout' : '', |
|
187 | 'stdout' : '', | |
188 | 'stderr' : '', |
|
188 | 'stderr' : '', | |
189 | 'outputs' : [], |
|
189 | 'outputs' : [], | |
190 | 'data': {}, |
|
190 | 'data': {}, | |
191 | 'outputs_ready' : False, |
|
191 | 'outputs_ready' : False, | |
192 | } |
|
192 | } | |
193 | self.update(md) |
|
193 | self.update(md) | |
194 | self.update(dict(*args, **kwargs)) |
|
194 | self.update(dict(*args, **kwargs)) | |
195 |
|
195 | |||
196 | def __getattr__(self, key): |
|
196 | def __getattr__(self, key): | |
197 | """getattr aliased to getitem""" |
|
197 | """getattr aliased to getitem""" | |
198 |
if key in self |
|
198 | if key in self: | |
199 | return self[key] |
|
199 | return self[key] | |
200 | else: |
|
200 | else: | |
201 | raise AttributeError(key) |
|
201 | raise AttributeError(key) | |
202 |
|
202 | |||
203 | def __setattr__(self, key, value): |
|
203 | def __setattr__(self, key, value): | |
204 | """setattr aliased to setitem, with strict""" |
|
204 | """setattr aliased to setitem, with strict""" | |
205 |
if key in self |
|
205 | if key in self: | |
206 | self[key] = value |
|
206 | self[key] = value | |
207 | else: |
|
207 | else: | |
208 | raise AttributeError(key) |
|
208 | raise AttributeError(key) | |
209 |
|
209 | |||
210 | def __setitem__(self, key, value): |
|
210 | def __setitem__(self, key, value): | |
211 | """strict static key enforcement""" |
|
211 | """strict static key enforcement""" | |
212 |
if key in self |
|
212 | if key in self: | |
213 | dict.__setitem__(self, key, value) |
|
213 | dict.__setitem__(self, key, value) | |
214 | else: |
|
214 | else: | |
215 | raise KeyError(key) |
|
215 | raise KeyError(key) | |
216 |
|
216 | |||
217 |
|
217 | |||
218 | class Client(HasTraits): |
|
218 | class Client(HasTraits): | |
219 | """A semi-synchronous client to the IPython ZMQ cluster |
|
219 | """A semi-synchronous client to the IPython ZMQ cluster | |
220 |
|
220 | |||
221 | Parameters |
|
221 | Parameters | |
222 | ---------- |
|
222 | ---------- | |
223 |
|
223 | |||
224 | url_file : str/unicode; path to ipcontroller-client.json |
|
224 | url_file : str/unicode; path to ipcontroller-client.json | |
225 | This JSON file should contain all the information needed to connect to a cluster, |
|
225 | This JSON file should contain all the information needed to connect to a cluster, | |
226 | and is likely the only argument needed. |
|
226 | and is likely the only argument needed. | |
227 | Connection information for the Hub's registration. If a json connector |
|
227 | Connection information for the Hub's registration. If a json connector | |
228 | file is given, then likely no further configuration is necessary. |
|
228 | file is given, then likely no further configuration is necessary. | |
229 | [Default: use profile] |
|
229 | [Default: use profile] | |
230 | profile : bytes |
|
230 | profile : bytes | |
231 | The name of the Cluster profile to be used to find connector information. |
|
231 | The name of the Cluster profile to be used to find connector information. | |
232 | If run from an IPython application, the default profile will be the same |
|
232 | If run from an IPython application, the default profile will be the same | |
233 | as the running application, otherwise it will be 'default'. |
|
233 | as the running application, otherwise it will be 'default'. | |
234 | cluster_id : str |
|
234 | cluster_id : str | |
235 | String id to added to runtime files, to prevent name collisions when using |
|
235 | String id to added to runtime files, to prevent name collisions when using | |
236 | multiple clusters with a single profile simultaneously. |
|
236 | multiple clusters with a single profile simultaneously. | |
237 | When set, will look for files named like: 'ipcontroller-<cluster_id>-client.json' |
|
237 | When set, will look for files named like: 'ipcontroller-<cluster_id>-client.json' | |
238 | Since this is text inserted into filenames, typical recommendations apply: |
|
238 | Since this is text inserted into filenames, typical recommendations apply: | |
239 | Simple character strings are ideal, and spaces are not recommended (but |
|
239 | Simple character strings are ideal, and spaces are not recommended (but | |
240 | should generally work) |
|
240 | should generally work) | |
241 | context : zmq.Context |
|
241 | context : zmq.Context | |
242 | Pass an existing zmq.Context instance, otherwise the client will create its own. |
|
242 | Pass an existing zmq.Context instance, otherwise the client will create its own. | |
243 | debug : bool |
|
243 | debug : bool | |
244 | flag for lots of message printing for debug purposes |
|
244 | flag for lots of message printing for debug purposes | |
245 | timeout : int/float |
|
245 | timeout : int/float | |
246 | time (in seconds) to wait for connection replies from the Hub |
|
246 | time (in seconds) to wait for connection replies from the Hub | |
247 | [Default: 10] |
|
247 | [Default: 10] | |
248 |
|
248 | |||
249 | #-------------- session related args ---------------- |
|
249 | #-------------- session related args ---------------- | |
250 |
|
250 | |||
251 | config : Config object |
|
251 | config : Config object | |
252 | If specified, this will be relayed to the Session for configuration |
|
252 | If specified, this will be relayed to the Session for configuration | |
253 | username : str |
|
253 | username : str | |
254 | set username for the session object |
|
254 | set username for the session object | |
255 |
|
255 | |||
256 | #-------------- ssh related args ---------------- |
|
256 | #-------------- ssh related args ---------------- | |
257 | # These are args for configuring the ssh tunnel to be used |
|
257 | # These are args for configuring the ssh tunnel to be used | |
258 | # credentials are used to forward connections over ssh to the Controller |
|
258 | # credentials are used to forward connections over ssh to the Controller | |
259 | # Note that the ip given in `addr` needs to be relative to sshserver |
|
259 | # Note that the ip given in `addr` needs to be relative to sshserver | |
260 | # The most basic case is to leave addr as pointing to localhost (127.0.0.1), |
|
260 | # The most basic case is to leave addr as pointing to localhost (127.0.0.1), | |
261 | # and set sshserver as the same machine the Controller is on. However, |
|
261 | # and set sshserver as the same machine the Controller is on. However, | |
262 | # the only requirement is that sshserver is able to see the Controller |
|
262 | # the only requirement is that sshserver is able to see the Controller | |
263 | # (i.e. is within the same trusted network). |
|
263 | # (i.e. is within the same trusted network). | |
264 |
|
264 | |||
265 | sshserver : str |
|
265 | sshserver : str | |
266 | A string of the form passed to ssh, i.e. 'server.tld' or 'user@server.tld:port' |
|
266 | A string of the form passed to ssh, i.e. 'server.tld' or 'user@server.tld:port' | |
267 | If keyfile or password is specified, and this is not, it will default to |
|
267 | If keyfile or password is specified, and this is not, it will default to | |
268 | the ip given in addr. |
|
268 | the ip given in addr. | |
269 | sshkey : str; path to ssh private key file |
|
269 | sshkey : str; path to ssh private key file | |
270 | This specifies a key to be used in ssh login, default None. |
|
270 | This specifies a key to be used in ssh login, default None. | |
271 | Regular default ssh keys will be used without specifying this argument. |
|
271 | Regular default ssh keys will be used without specifying this argument. | |
272 | password : str |
|
272 | password : str | |
273 | Your ssh password to sshserver. Note that if this is left None, |
|
273 | Your ssh password to sshserver. Note that if this is left None, | |
274 | you will be prompted for it if passwordless key based login is unavailable. |
|
274 | you will be prompted for it if passwordless key based login is unavailable. | |
275 | paramiko : bool |
|
275 | paramiko : bool | |
276 | flag for whether to use paramiko instead of shell ssh for tunneling. |
|
276 | flag for whether to use paramiko instead of shell ssh for tunneling. | |
277 | [default: True on win32, False else] |
|
277 | [default: True on win32, False else] | |
278 |
|
278 | |||
279 |
|
279 | |||
280 | Attributes |
|
280 | Attributes | |
281 | ---------- |
|
281 | ---------- | |
282 |
|
282 | |||
283 | ids : list of int engine IDs |
|
283 | ids : list of int engine IDs | |
284 | requesting the ids attribute always synchronizes |
|
284 | requesting the ids attribute always synchronizes | |
285 | the registration state. To request ids without synchronization, |
|
285 | the registration state. To request ids without synchronization, | |
286 | use semi-private _ids attributes. |
|
286 | use semi-private _ids attributes. | |
287 |
|
287 | |||
288 | history : list of msg_ids |
|
288 | history : list of msg_ids | |
289 | a list of msg_ids, keeping track of all the execution |
|
289 | a list of msg_ids, keeping track of all the execution | |
290 | messages you have submitted in order. |
|
290 | messages you have submitted in order. | |
291 |
|
291 | |||
292 | outstanding : set of msg_ids |
|
292 | outstanding : set of msg_ids | |
293 | a set of msg_ids that have been submitted, but whose |
|
293 | a set of msg_ids that have been submitted, but whose | |
294 | results have not yet been received. |
|
294 | results have not yet been received. | |
295 |
|
295 | |||
296 | results : dict |
|
296 | results : dict | |
297 | a dict of all our results, keyed by msg_id |
|
297 | a dict of all our results, keyed by msg_id | |
298 |
|
298 | |||
299 | block : bool |
|
299 | block : bool | |
300 | determines default behavior when block not specified |
|
300 | determines default behavior when block not specified | |
301 | in execution methods |
|
301 | in execution methods | |
302 |
|
302 | |||
303 | Methods |
|
303 | Methods | |
304 | ------- |
|
304 | ------- | |
305 |
|
305 | |||
306 | spin |
|
306 | spin | |
307 | flushes incoming results and registration state changes |
|
307 | flushes incoming results and registration state changes | |
308 | control methods spin, and requesting `ids` also ensures up to date |
|
308 | control methods spin, and requesting `ids` also ensures up to date | |
309 |
|
309 | |||
310 | wait |
|
310 | wait | |
311 | wait on one or more msg_ids |
|
311 | wait on one or more msg_ids | |
312 |
|
312 | |||
313 | execution methods |
|
313 | execution methods | |
314 | apply |
|
314 | apply | |
315 | legacy: execute, run |
|
315 | legacy: execute, run | |
316 |
|
316 | |||
317 | data movement |
|
317 | data movement | |
318 | push, pull, scatter, gather |
|
318 | push, pull, scatter, gather | |
319 |
|
319 | |||
320 | query methods |
|
320 | query methods | |
321 | queue_status, get_result, purge, result_status |
|
321 | queue_status, get_result, purge, result_status | |
322 |
|
322 | |||
323 | control methods |
|
323 | control methods | |
324 | abort, shutdown |
|
324 | abort, shutdown | |
325 |
|
325 | |||
326 | """ |
|
326 | """ | |
327 |
|
327 | |||
328 |
|
328 | |||
329 | block = Bool(False) |
|
329 | block = Bool(False) | |
330 | outstanding = Set() |
|
330 | outstanding = Set() | |
331 | results = Instance('collections.defaultdict', (dict,)) |
|
331 | results = Instance('collections.defaultdict', (dict,)) | |
332 | metadata = Instance('collections.defaultdict', (Metadata,)) |
|
332 | metadata = Instance('collections.defaultdict', (Metadata,)) | |
333 | history = List() |
|
333 | history = List() | |
334 | debug = Bool(False) |
|
334 | debug = Bool(False) | |
335 | _spin_thread = Any() |
|
335 | _spin_thread = Any() | |
336 | _stop_spinning = Any() |
|
336 | _stop_spinning = Any() | |
337 |
|
337 | |||
338 | profile=Unicode() |
|
338 | profile=Unicode() | |
339 | def _profile_default(self): |
|
339 | def _profile_default(self): | |
340 | if BaseIPythonApplication.initialized(): |
|
340 | if BaseIPythonApplication.initialized(): | |
341 | # an IPython app *might* be running, try to get its profile |
|
341 | # an IPython app *might* be running, try to get its profile | |
342 | try: |
|
342 | try: | |
343 | return BaseIPythonApplication.instance().profile |
|
343 | return BaseIPythonApplication.instance().profile | |
344 | except (AttributeError, MultipleInstanceError): |
|
344 | except (AttributeError, MultipleInstanceError): | |
345 | # could be a *different* subclass of config.Application, |
|
345 | # could be a *different* subclass of config.Application, | |
346 | # which would raise one of these two errors. |
|
346 | # which would raise one of these two errors. | |
347 | return u'default' |
|
347 | return u'default' | |
348 | else: |
|
348 | else: | |
349 | return u'default' |
|
349 | return u'default' | |
350 |
|
350 | |||
351 |
|
351 | |||
352 | _outstanding_dict = Instance('collections.defaultdict', (set,)) |
|
352 | _outstanding_dict = Instance('collections.defaultdict', (set,)) | |
353 | _ids = List() |
|
353 | _ids = List() | |
354 | _connected=Bool(False) |
|
354 | _connected=Bool(False) | |
355 | _ssh=Bool(False) |
|
355 | _ssh=Bool(False) | |
356 | _context = Instance('zmq.Context') |
|
356 | _context = Instance('zmq.Context') | |
357 | _config = Dict() |
|
357 | _config = Dict() | |
358 | _engines=Instance(util.ReverseDict, (), {}) |
|
358 | _engines=Instance(util.ReverseDict, (), {}) | |
359 | # _hub_socket=Instance('zmq.Socket') |
|
359 | # _hub_socket=Instance('zmq.Socket') | |
360 | _query_socket=Instance('zmq.Socket') |
|
360 | _query_socket=Instance('zmq.Socket') | |
361 | _control_socket=Instance('zmq.Socket') |
|
361 | _control_socket=Instance('zmq.Socket') | |
362 | _iopub_socket=Instance('zmq.Socket') |
|
362 | _iopub_socket=Instance('zmq.Socket') | |
363 | _notification_socket=Instance('zmq.Socket') |
|
363 | _notification_socket=Instance('zmq.Socket') | |
364 | _mux_socket=Instance('zmq.Socket') |
|
364 | _mux_socket=Instance('zmq.Socket') | |
365 | _task_socket=Instance('zmq.Socket') |
|
365 | _task_socket=Instance('zmq.Socket') | |
366 | _task_scheme=Unicode() |
|
366 | _task_scheme=Unicode() | |
367 | _closed = False |
|
367 | _closed = False | |
368 | _ignored_control_replies=Integer(0) |
|
368 | _ignored_control_replies=Integer(0) | |
369 | _ignored_hub_replies=Integer(0) |
|
369 | _ignored_hub_replies=Integer(0) | |
370 |
|
370 | |||
371 | def __new__(self, *args, **kw): |
|
371 | def __new__(self, *args, **kw): | |
372 | # don't raise on positional args |
|
372 | # don't raise on positional args | |
373 | return HasTraits.__new__(self, **kw) |
|
373 | return HasTraits.__new__(self, **kw) | |
374 |
|
374 | |||
375 | def __init__(self, url_file=None, profile=None, profile_dir=None, ipython_dir=None, |
|
375 | def __init__(self, url_file=None, profile=None, profile_dir=None, ipython_dir=None, | |
376 | context=None, debug=False, |
|
376 | context=None, debug=False, | |
377 | sshserver=None, sshkey=None, password=None, paramiko=None, |
|
377 | sshserver=None, sshkey=None, password=None, paramiko=None, | |
378 | timeout=10, cluster_id=None, **extra_args |
|
378 | timeout=10, cluster_id=None, **extra_args | |
379 | ): |
|
379 | ): | |
380 | if profile: |
|
380 | if profile: | |
381 | super(Client, self).__init__(debug=debug, profile=profile) |
|
381 | super(Client, self).__init__(debug=debug, profile=profile) | |
382 | else: |
|
382 | else: | |
383 | super(Client, self).__init__(debug=debug) |
|
383 | super(Client, self).__init__(debug=debug) | |
384 | if context is None: |
|
384 | if context is None: | |
385 | context = zmq.Context.instance() |
|
385 | context = zmq.Context.instance() | |
386 | self._context = context |
|
386 | self._context = context | |
387 | self._stop_spinning = Event() |
|
387 | self._stop_spinning = Event() | |
388 |
|
388 | |||
389 | if 'url_or_file' in extra_args: |
|
389 | if 'url_or_file' in extra_args: | |
390 | url_file = extra_args['url_or_file'] |
|
390 | url_file = extra_args['url_or_file'] | |
391 | warnings.warn("url_or_file arg no longer supported, use url_file", DeprecationWarning) |
|
391 | warnings.warn("url_or_file arg no longer supported, use url_file", DeprecationWarning) | |
392 |
|
392 | |||
393 | if url_file and util.is_url(url_file): |
|
393 | if url_file and util.is_url(url_file): | |
394 | raise ValueError("single urls cannot be specified, url-files must be used.") |
|
394 | raise ValueError("single urls cannot be specified, url-files must be used.") | |
395 |
|
395 | |||
396 | self._setup_profile_dir(self.profile, profile_dir, ipython_dir) |
|
396 | self._setup_profile_dir(self.profile, profile_dir, ipython_dir) | |
397 |
|
397 | |||
398 | if self._cd is not None: |
|
398 | if self._cd is not None: | |
399 | if url_file is None: |
|
399 | if url_file is None: | |
400 | if not cluster_id: |
|
400 | if not cluster_id: | |
401 | client_json = 'ipcontroller-client.json' |
|
401 | client_json = 'ipcontroller-client.json' | |
402 | else: |
|
402 | else: | |
403 | client_json = 'ipcontroller-%s-client.json' % cluster_id |
|
403 | client_json = 'ipcontroller-%s-client.json' % cluster_id | |
404 | url_file = pjoin(self._cd.security_dir, client_json) |
|
404 | url_file = pjoin(self._cd.security_dir, client_json) | |
405 | if url_file is None: |
|
405 | if url_file is None: | |
406 | raise ValueError( |
|
406 | raise ValueError( | |
407 | "I can't find enough information to connect to a hub!" |
|
407 | "I can't find enough information to connect to a hub!" | |
408 | " Please specify at least one of url_file or profile." |
|
408 | " Please specify at least one of url_file or profile." | |
409 | ) |
|
409 | ) | |
410 |
|
410 | |||
411 | with open(url_file) as f: |
|
411 | with open(url_file) as f: | |
412 | cfg = json.load(f) |
|
412 | cfg = json.load(f) | |
413 |
|
413 | |||
414 | self._task_scheme = cfg['task_scheme'] |
|
414 | self._task_scheme = cfg['task_scheme'] | |
415 |
|
415 | |||
416 | # sync defaults from args, json: |
|
416 | # sync defaults from args, json: | |
417 | if sshserver: |
|
417 | if sshserver: | |
418 | cfg['ssh'] = sshserver |
|
418 | cfg['ssh'] = sshserver | |
419 |
|
419 | |||
420 | location = cfg.setdefault('location', None) |
|
420 | location = cfg.setdefault('location', None) | |
421 |
|
421 | |||
422 | proto,addr = cfg['interface'].split('://') |
|
422 | proto,addr = cfg['interface'].split('://') | |
423 | addr = util.disambiguate_ip_address(addr, location) |
|
423 | addr = util.disambiguate_ip_address(addr, location) | |
424 | cfg['interface'] = "%s://%s" % (proto, addr) |
|
424 | cfg['interface'] = "%s://%s" % (proto, addr) | |
425 |
|
425 | |||
426 | # turn interface,port into full urls: |
|
426 | # turn interface,port into full urls: | |
427 | for key in ('control', 'task', 'mux', 'iopub', 'notification', 'registration'): |
|
427 | for key in ('control', 'task', 'mux', 'iopub', 'notification', 'registration'): | |
428 | cfg[key] = cfg['interface'] + ':%i' % cfg[key] |
|
428 | cfg[key] = cfg['interface'] + ':%i' % cfg[key] | |
429 |
|
429 | |||
430 | url = cfg['registration'] |
|
430 | url = cfg['registration'] | |
431 |
|
431 | |||
432 | if location is not None and addr == localhost(): |
|
432 | if location is not None and addr == localhost(): | |
433 | # location specified, and connection is expected to be local |
|
433 | # location specified, and connection is expected to be local | |
434 | if not is_local_ip(location) and not sshserver: |
|
434 | if not is_local_ip(location) and not sshserver: | |
435 | # load ssh from JSON *only* if the controller is not on |
|
435 | # load ssh from JSON *only* if the controller is not on | |
436 | # this machine |
|
436 | # this machine | |
437 | sshserver=cfg['ssh'] |
|
437 | sshserver=cfg['ssh'] | |
438 | if not is_local_ip(location) and not sshserver: |
|
438 | if not is_local_ip(location) and not sshserver: | |
439 | # warn if no ssh specified, but SSH is probably needed |
|
439 | # warn if no ssh specified, but SSH is probably needed | |
440 | # This is only a warning, because the most likely cause |
|
440 | # This is only a warning, because the most likely cause | |
441 | # is a local Controller on a laptop whose IP is dynamic |
|
441 | # is a local Controller on a laptop whose IP is dynamic | |
442 | warnings.warn(""" |
|
442 | warnings.warn(""" | |
443 | Controller appears to be listening on localhost, but not on this machine. |
|
443 | Controller appears to be listening on localhost, but not on this machine. | |
444 | If this is true, you should specify Client(...,sshserver='you@%s') |
|
444 | If this is true, you should specify Client(...,sshserver='you@%s') | |
445 | or instruct your controller to listen on an external IP."""%location, |
|
445 | or instruct your controller to listen on an external IP."""%location, | |
446 | RuntimeWarning) |
|
446 | RuntimeWarning) | |
447 | elif not sshserver: |
|
447 | elif not sshserver: | |
448 | # otherwise sync with cfg |
|
448 | # otherwise sync with cfg | |
449 | sshserver = cfg['ssh'] |
|
449 | sshserver = cfg['ssh'] | |
450 |
|
450 | |||
451 | self._config = cfg |
|
451 | self._config = cfg | |
452 |
|
452 | |||
453 | self._ssh = bool(sshserver or sshkey or password) |
|
453 | self._ssh = bool(sshserver or sshkey or password) | |
454 | if self._ssh and sshserver is None: |
|
454 | if self._ssh and sshserver is None: | |
455 | # default to ssh via localhost |
|
455 | # default to ssh via localhost | |
456 | sshserver = addr |
|
456 | sshserver = addr | |
457 | if self._ssh and password is None: |
|
457 | if self._ssh and password is None: | |
458 | if tunnel.try_passwordless_ssh(sshserver, sshkey, paramiko): |
|
458 | if tunnel.try_passwordless_ssh(sshserver, sshkey, paramiko): | |
459 | password=False |
|
459 | password=False | |
460 | else: |
|
460 | else: | |
461 | password = getpass("SSH Password for %s: "%sshserver) |
|
461 | password = getpass("SSH Password for %s: "%sshserver) | |
462 | ssh_kwargs = dict(keyfile=sshkey, password=password, paramiko=paramiko) |
|
462 | ssh_kwargs = dict(keyfile=sshkey, password=password, paramiko=paramiko) | |
463 |
|
463 | |||
464 | # configure and construct the session |
|
464 | # configure and construct the session | |
465 | try: |
|
465 | try: | |
466 | extra_args['packer'] = cfg['pack'] |
|
466 | extra_args['packer'] = cfg['pack'] | |
467 | extra_args['unpacker'] = cfg['unpack'] |
|
467 | extra_args['unpacker'] = cfg['unpack'] | |
468 | extra_args['key'] = cast_bytes(cfg['key']) |
|
468 | extra_args['key'] = cast_bytes(cfg['key']) | |
469 | extra_args['signature_scheme'] = cfg['signature_scheme'] |
|
469 | extra_args['signature_scheme'] = cfg['signature_scheme'] | |
470 | except KeyError as exc: |
|
470 | except KeyError as exc: | |
471 | msg = '\n'.join([ |
|
471 | msg = '\n'.join([ | |
472 | "Connection file is invalid (missing '{}'), possibly from an old version of IPython.", |
|
472 | "Connection file is invalid (missing '{}'), possibly from an old version of IPython.", | |
473 | "If you are reusing connection files, remove them and start ipcontroller again." |
|
473 | "If you are reusing connection files, remove them and start ipcontroller again." | |
474 | ]) |
|
474 | ]) | |
475 | raise ValueError(msg.format(exc.message)) |
|
475 | raise ValueError(msg.format(exc.message)) | |
476 |
|
476 | |||
477 | self.session = Session(**extra_args) |
|
477 | self.session = Session(**extra_args) | |
478 |
|
478 | |||
479 | self._query_socket = self._context.socket(zmq.DEALER) |
|
479 | self._query_socket = self._context.socket(zmq.DEALER) | |
480 |
|
480 | |||
481 | if self._ssh: |
|
481 | if self._ssh: | |
482 | tunnel.tunnel_connection(self._query_socket, cfg['registration'], sshserver, **ssh_kwargs) |
|
482 | tunnel.tunnel_connection(self._query_socket, cfg['registration'], sshserver, **ssh_kwargs) | |
483 | else: |
|
483 | else: | |
484 | self._query_socket.connect(cfg['registration']) |
|
484 | self._query_socket.connect(cfg['registration']) | |
485 |
|
485 | |||
486 | self.session.debug = self.debug |
|
486 | self.session.debug = self.debug | |
487 |
|
487 | |||
488 | self._notification_handlers = {'registration_notification' : self._register_engine, |
|
488 | self._notification_handlers = {'registration_notification' : self._register_engine, | |
489 | 'unregistration_notification' : self._unregister_engine, |
|
489 | 'unregistration_notification' : self._unregister_engine, | |
490 | 'shutdown_notification' : lambda msg: self.close(), |
|
490 | 'shutdown_notification' : lambda msg: self.close(), | |
491 | } |
|
491 | } | |
492 | self._queue_handlers = {'execute_reply' : self._handle_execute_reply, |
|
492 | self._queue_handlers = {'execute_reply' : self._handle_execute_reply, | |
493 | 'apply_reply' : self._handle_apply_reply} |
|
493 | 'apply_reply' : self._handle_apply_reply} | |
494 |
|
494 | |||
495 | try: |
|
495 | try: | |
496 | self._connect(sshserver, ssh_kwargs, timeout) |
|
496 | self._connect(sshserver, ssh_kwargs, timeout) | |
497 | except: |
|
497 | except: | |
498 | self.close(linger=0) |
|
498 | self.close(linger=0) | |
499 | raise |
|
499 | raise | |
500 |
|
500 | |||
501 | # last step: setup magics, if we are in IPython: |
|
501 | # last step: setup magics, if we are in IPython: | |
502 |
|
502 | |||
503 | try: |
|
503 | try: | |
504 | ip = get_ipython() |
|
504 | ip = get_ipython() | |
505 | except NameError: |
|
505 | except NameError: | |
506 | return |
|
506 | return | |
507 | else: |
|
507 | else: | |
508 | if 'px' not in ip.magics_manager.magics: |
|
508 | if 'px' not in ip.magics_manager.magics: | |
509 | # in IPython but we are the first Client. |
|
509 | # in IPython but we are the first Client. | |
510 | # activate a default view for parallel magics. |
|
510 | # activate a default view for parallel magics. | |
511 | self.activate() |
|
511 | self.activate() | |
512 |
|
512 | |||
513 | def __del__(self): |
|
513 | def __del__(self): | |
514 | """cleanup sockets, but _not_ context.""" |
|
514 | """cleanup sockets, but _not_ context.""" | |
515 | self.close() |
|
515 | self.close() | |
516 |
|
516 | |||
517 | def _setup_profile_dir(self, profile, profile_dir, ipython_dir): |
|
517 | def _setup_profile_dir(self, profile, profile_dir, ipython_dir): | |
518 | if ipython_dir is None: |
|
518 | if ipython_dir is None: | |
519 | ipython_dir = get_ipython_dir() |
|
519 | ipython_dir = get_ipython_dir() | |
520 | if profile_dir is not None: |
|
520 | if profile_dir is not None: | |
521 | try: |
|
521 | try: | |
522 | self._cd = ProfileDir.find_profile_dir(profile_dir) |
|
522 | self._cd = ProfileDir.find_profile_dir(profile_dir) | |
523 | return |
|
523 | return | |
524 | except ProfileDirError: |
|
524 | except ProfileDirError: | |
525 | pass |
|
525 | pass | |
526 | elif profile is not None: |
|
526 | elif profile is not None: | |
527 | try: |
|
527 | try: | |
528 | self._cd = ProfileDir.find_profile_dir_by_name( |
|
528 | self._cd = ProfileDir.find_profile_dir_by_name( | |
529 | ipython_dir, profile) |
|
529 | ipython_dir, profile) | |
530 | return |
|
530 | return | |
531 | except ProfileDirError: |
|
531 | except ProfileDirError: | |
532 | pass |
|
532 | pass | |
533 | self._cd = None |
|
533 | self._cd = None | |
534 |
|
534 | |||
535 | def _update_engines(self, engines): |
|
535 | def _update_engines(self, engines): | |
536 | """Update our engines dict and _ids from a dict of the form: {id:uuid}.""" |
|
536 | """Update our engines dict and _ids from a dict of the form: {id:uuid}.""" | |
537 | for k,v in engines.iteritems(): |
|
537 | for k,v in engines.iteritems(): | |
538 | eid = int(k) |
|
538 | eid = int(k) | |
539 | if eid not in self._engines: |
|
539 | if eid not in self._engines: | |
540 | self._ids.append(eid) |
|
540 | self._ids.append(eid) | |
541 | self._engines[eid] = v |
|
541 | self._engines[eid] = v | |
542 | self._ids = sorted(self._ids) |
|
542 | self._ids = sorted(self._ids) | |
543 | if sorted(self._engines.keys()) != range(len(self._engines)) and \ |
|
543 | if sorted(self._engines.keys()) != range(len(self._engines)) and \ | |
544 | self._task_scheme == 'pure' and self._task_socket: |
|
544 | self._task_scheme == 'pure' and self._task_socket: | |
545 | self._stop_scheduling_tasks() |
|
545 | self._stop_scheduling_tasks() | |
546 |
|
546 | |||
547 | def _stop_scheduling_tasks(self): |
|
547 | def _stop_scheduling_tasks(self): | |
548 | """Stop scheduling tasks because an engine has been unregistered |
|
548 | """Stop scheduling tasks because an engine has been unregistered | |
549 | from a pure ZMQ scheduler. |
|
549 | from a pure ZMQ scheduler. | |
550 | """ |
|
550 | """ | |
551 | self._task_socket.close() |
|
551 | self._task_socket.close() | |
552 | self._task_socket = None |
|
552 | self._task_socket = None | |
553 | msg = "An engine has been unregistered, and we are using pure " +\ |
|
553 | msg = "An engine has been unregistered, and we are using pure " +\ | |
554 | "ZMQ task scheduling. Task farming will be disabled." |
|
554 | "ZMQ task scheduling. Task farming will be disabled." | |
555 | if self.outstanding: |
|
555 | if self.outstanding: | |
556 | msg += " If you were running tasks when this happened, " +\ |
|
556 | msg += " If you were running tasks when this happened, " +\ | |
557 | "some `outstanding` msg_ids may never resolve." |
|
557 | "some `outstanding` msg_ids may never resolve." | |
558 | warnings.warn(msg, RuntimeWarning) |
|
558 | warnings.warn(msg, RuntimeWarning) | |
559 |
|
559 | |||
560 | def _build_targets(self, targets): |
|
560 | def _build_targets(self, targets): | |
561 | """Turn valid target IDs or 'all' into two lists: |
|
561 | """Turn valid target IDs or 'all' into two lists: | |
562 | (int_ids, uuids). |
|
562 | (int_ids, uuids). | |
563 | """ |
|
563 | """ | |
564 | if not self._ids: |
|
564 | if not self._ids: | |
565 | # flush notification socket if no engines yet, just in case |
|
565 | # flush notification socket if no engines yet, just in case | |
566 | if not self.ids: |
|
566 | if not self.ids: | |
567 | raise error.NoEnginesRegistered("Can't build targets without any engines") |
|
567 | raise error.NoEnginesRegistered("Can't build targets without any engines") | |
568 |
|
568 | |||
569 | if targets is None: |
|
569 | if targets is None: | |
570 | targets = self._ids |
|
570 | targets = self._ids | |
571 | elif isinstance(targets, string_types): |
|
571 | elif isinstance(targets, string_types): | |
572 | if targets.lower() == 'all': |
|
572 | if targets.lower() == 'all': | |
573 | targets = self._ids |
|
573 | targets = self._ids | |
574 | else: |
|
574 | else: | |
575 | raise TypeError("%r not valid str target, must be 'all'"%(targets)) |
|
575 | raise TypeError("%r not valid str target, must be 'all'"%(targets)) | |
576 | elif isinstance(targets, int): |
|
576 | elif isinstance(targets, int): | |
577 | if targets < 0: |
|
577 | if targets < 0: | |
578 | targets = self.ids[targets] |
|
578 | targets = self.ids[targets] | |
579 | if targets not in self._ids: |
|
579 | if targets not in self._ids: | |
580 | raise IndexError("No such engine: %i"%targets) |
|
580 | raise IndexError("No such engine: %i"%targets) | |
581 | targets = [targets] |
|
581 | targets = [targets] | |
582 |
|
582 | |||
583 | if isinstance(targets, slice): |
|
583 | if isinstance(targets, slice): | |
584 | indices = range(len(self._ids))[targets] |
|
584 | indices = range(len(self._ids))[targets] | |
585 | ids = self.ids |
|
585 | ids = self.ids | |
586 | targets = [ ids[i] for i in indices ] |
|
586 | targets = [ ids[i] for i in indices ] | |
587 |
|
587 | |||
588 | if not isinstance(targets, (tuple, list, xrange)): |
|
588 | if not isinstance(targets, (tuple, list, xrange)): | |
589 | raise TypeError("targets by int/slice/collection of ints only, not %s"%(type(targets))) |
|
589 | raise TypeError("targets by int/slice/collection of ints only, not %s"%(type(targets))) | |
590 |
|
590 | |||
591 | return [cast_bytes(self._engines[t]) for t in targets], list(targets) |
|
591 | return [cast_bytes(self._engines[t]) for t in targets], list(targets) | |
592 |
|
592 | |||
593 | def _connect(self, sshserver, ssh_kwargs, timeout): |
|
593 | def _connect(self, sshserver, ssh_kwargs, timeout): | |
594 | """setup all our socket connections to the cluster. This is called from |
|
594 | """setup all our socket connections to the cluster. This is called from | |
595 | __init__.""" |
|
595 | __init__.""" | |
596 |
|
596 | |||
597 | # Maybe allow reconnecting? |
|
597 | # Maybe allow reconnecting? | |
598 | if self._connected: |
|
598 | if self._connected: | |
599 | return |
|
599 | return | |
600 | self._connected=True |
|
600 | self._connected=True | |
601 |
|
601 | |||
602 | def connect_socket(s, url): |
|
602 | def connect_socket(s, url): | |
603 | if self._ssh: |
|
603 | if self._ssh: | |
604 | return tunnel.tunnel_connection(s, url, sshserver, **ssh_kwargs) |
|
604 | return tunnel.tunnel_connection(s, url, sshserver, **ssh_kwargs) | |
605 | else: |
|
605 | else: | |
606 | return s.connect(url) |
|
606 | return s.connect(url) | |
607 |
|
607 | |||
608 | self.session.send(self._query_socket, 'connection_request') |
|
608 | self.session.send(self._query_socket, 'connection_request') | |
609 | # use Poller because zmq.select has wrong units in pyzmq 2.1.7 |
|
609 | # use Poller because zmq.select has wrong units in pyzmq 2.1.7 | |
610 | poller = zmq.Poller() |
|
610 | poller = zmq.Poller() | |
611 | poller.register(self._query_socket, zmq.POLLIN) |
|
611 | poller.register(self._query_socket, zmq.POLLIN) | |
612 | # poll expects milliseconds, timeout is seconds |
|
612 | # poll expects milliseconds, timeout is seconds | |
613 | evts = poller.poll(timeout*1000) |
|
613 | evts = poller.poll(timeout*1000) | |
614 | if not evts: |
|
614 | if not evts: | |
615 | raise error.TimeoutError("Hub connection request timed out") |
|
615 | raise error.TimeoutError("Hub connection request timed out") | |
616 | idents,msg = self.session.recv(self._query_socket,mode=0) |
|
616 | idents,msg = self.session.recv(self._query_socket,mode=0) | |
617 | if self.debug: |
|
617 | if self.debug: | |
618 | pprint(msg) |
|
618 | pprint(msg) | |
619 | content = msg['content'] |
|
619 | content = msg['content'] | |
620 | # self._config['registration'] = dict(content) |
|
620 | # self._config['registration'] = dict(content) | |
621 | cfg = self._config |
|
621 | cfg = self._config | |
622 | if content['status'] == 'ok': |
|
622 | if content['status'] == 'ok': | |
623 | self._mux_socket = self._context.socket(zmq.DEALER) |
|
623 | self._mux_socket = self._context.socket(zmq.DEALER) | |
624 | connect_socket(self._mux_socket, cfg['mux']) |
|
624 | connect_socket(self._mux_socket, cfg['mux']) | |
625 |
|
625 | |||
626 | self._task_socket = self._context.socket(zmq.DEALER) |
|
626 | self._task_socket = self._context.socket(zmq.DEALER) | |
627 | connect_socket(self._task_socket, cfg['task']) |
|
627 | connect_socket(self._task_socket, cfg['task']) | |
628 |
|
628 | |||
629 | self._notification_socket = self._context.socket(zmq.SUB) |
|
629 | self._notification_socket = self._context.socket(zmq.SUB) | |
630 | self._notification_socket.setsockopt(zmq.SUBSCRIBE, b'') |
|
630 | self._notification_socket.setsockopt(zmq.SUBSCRIBE, b'') | |
631 | connect_socket(self._notification_socket, cfg['notification']) |
|
631 | connect_socket(self._notification_socket, cfg['notification']) | |
632 |
|
632 | |||
633 | self._control_socket = self._context.socket(zmq.DEALER) |
|
633 | self._control_socket = self._context.socket(zmq.DEALER) | |
634 | connect_socket(self._control_socket, cfg['control']) |
|
634 | connect_socket(self._control_socket, cfg['control']) | |
635 |
|
635 | |||
636 | self._iopub_socket = self._context.socket(zmq.SUB) |
|
636 | self._iopub_socket = self._context.socket(zmq.SUB) | |
637 | self._iopub_socket.setsockopt(zmq.SUBSCRIBE, b'') |
|
637 | self._iopub_socket.setsockopt(zmq.SUBSCRIBE, b'') | |
638 | connect_socket(self._iopub_socket, cfg['iopub']) |
|
638 | connect_socket(self._iopub_socket, cfg['iopub']) | |
639 |
|
639 | |||
640 | self._update_engines(dict(content['engines'])) |
|
640 | self._update_engines(dict(content['engines'])) | |
641 | else: |
|
641 | else: | |
642 | self._connected = False |
|
642 | self._connected = False | |
643 | raise Exception("Failed to connect!") |
|
643 | raise Exception("Failed to connect!") | |
644 |
|
644 | |||
645 | #-------------------------------------------------------------------------- |
|
645 | #-------------------------------------------------------------------------- | |
646 | # handlers and callbacks for incoming messages |
|
646 | # handlers and callbacks for incoming messages | |
647 | #-------------------------------------------------------------------------- |
|
647 | #-------------------------------------------------------------------------- | |
648 |
|
648 | |||
649 | def _unwrap_exception(self, content): |
|
649 | def _unwrap_exception(self, content): | |
650 | """unwrap exception, and remap engine_id to int.""" |
|
650 | """unwrap exception, and remap engine_id to int.""" | |
651 | e = error.unwrap_exception(content) |
|
651 | e = error.unwrap_exception(content) | |
652 | # print e.traceback |
|
652 | # print e.traceback | |
653 | if e.engine_info: |
|
653 | if e.engine_info: | |
654 | e_uuid = e.engine_info['engine_uuid'] |
|
654 | e_uuid = e.engine_info['engine_uuid'] | |
655 | eid = self._engines[e_uuid] |
|
655 | eid = self._engines[e_uuid] | |
656 | e.engine_info['engine_id'] = eid |
|
656 | e.engine_info['engine_id'] = eid | |
657 | return e |
|
657 | return e | |
658 |
|
658 | |||
659 | def _extract_metadata(self, msg): |
|
659 | def _extract_metadata(self, msg): | |
660 | header = msg['header'] |
|
660 | header = msg['header'] | |
661 | parent = msg['parent_header'] |
|
661 | parent = msg['parent_header'] | |
662 | msg_meta = msg['metadata'] |
|
662 | msg_meta = msg['metadata'] | |
663 | content = msg['content'] |
|
663 | content = msg['content'] | |
664 | md = {'msg_id' : parent['msg_id'], |
|
664 | md = {'msg_id' : parent['msg_id'], | |
665 | 'received' : datetime.now(), |
|
665 | 'received' : datetime.now(), | |
666 | 'engine_uuid' : msg_meta.get('engine', None), |
|
666 | 'engine_uuid' : msg_meta.get('engine', None), | |
667 | 'follow' : msg_meta.get('follow', []), |
|
667 | 'follow' : msg_meta.get('follow', []), | |
668 | 'after' : msg_meta.get('after', []), |
|
668 | 'after' : msg_meta.get('after', []), | |
669 | 'status' : content['status'], |
|
669 | 'status' : content['status'], | |
670 | } |
|
670 | } | |
671 |
|
671 | |||
672 | if md['engine_uuid'] is not None: |
|
672 | if md['engine_uuid'] is not None: | |
673 | md['engine_id'] = self._engines.get(md['engine_uuid'], None) |
|
673 | md['engine_id'] = self._engines.get(md['engine_uuid'], None) | |
674 |
|
674 | |||
675 | if 'date' in parent: |
|
675 | if 'date' in parent: | |
676 | md['submitted'] = parent['date'] |
|
676 | md['submitted'] = parent['date'] | |
677 | if 'started' in msg_meta: |
|
677 | if 'started' in msg_meta: | |
678 | md['started'] = msg_meta['started'] |
|
678 | md['started'] = msg_meta['started'] | |
679 | if 'date' in header: |
|
679 | if 'date' in header: | |
680 | md['completed'] = header['date'] |
|
680 | md['completed'] = header['date'] | |
681 | return md |
|
681 | return md | |
682 |
|
682 | |||
683 | def _register_engine(self, msg): |
|
683 | def _register_engine(self, msg): | |
684 | """Register a new engine, and update our connection info.""" |
|
684 | """Register a new engine, and update our connection info.""" | |
685 | content = msg['content'] |
|
685 | content = msg['content'] | |
686 | eid = content['id'] |
|
686 | eid = content['id'] | |
687 | d = {eid : content['uuid']} |
|
687 | d = {eid : content['uuid']} | |
688 | self._update_engines(d) |
|
688 | self._update_engines(d) | |
689 |
|
689 | |||
690 | def _unregister_engine(self, msg): |
|
690 | def _unregister_engine(self, msg): | |
691 | """Unregister an engine that has died.""" |
|
691 | """Unregister an engine that has died.""" | |
692 | content = msg['content'] |
|
692 | content = msg['content'] | |
693 | eid = int(content['id']) |
|
693 | eid = int(content['id']) | |
694 | if eid in self._ids: |
|
694 | if eid in self._ids: | |
695 | self._ids.remove(eid) |
|
695 | self._ids.remove(eid) | |
696 | uuid = self._engines.pop(eid) |
|
696 | uuid = self._engines.pop(eid) | |
697 |
|
697 | |||
698 | self._handle_stranded_msgs(eid, uuid) |
|
698 | self._handle_stranded_msgs(eid, uuid) | |
699 |
|
699 | |||
700 | if self._task_socket and self._task_scheme == 'pure': |
|
700 | if self._task_socket and self._task_scheme == 'pure': | |
701 | self._stop_scheduling_tasks() |
|
701 | self._stop_scheduling_tasks() | |
702 |
|
702 | |||
703 | def _handle_stranded_msgs(self, eid, uuid): |
|
703 | def _handle_stranded_msgs(self, eid, uuid): | |
704 | """Handle messages known to be on an engine when the engine unregisters. |
|
704 | """Handle messages known to be on an engine when the engine unregisters. | |
705 |
|
705 | |||
706 | It is possible that this will fire prematurely - that is, an engine will |
|
706 | It is possible that this will fire prematurely - that is, an engine will | |
707 | go down after completing a result, and the client will be notified |
|
707 | go down after completing a result, and the client will be notified | |
708 | of the unregistration and later receive the successful result. |
|
708 | of the unregistration and later receive the successful result. | |
709 | """ |
|
709 | """ | |
710 |
|
710 | |||
711 | outstanding = self._outstanding_dict[uuid] |
|
711 | outstanding = self._outstanding_dict[uuid] | |
712 |
|
712 | |||
713 | for msg_id in list(outstanding): |
|
713 | for msg_id in list(outstanding): | |
714 | if msg_id in self.results: |
|
714 | if msg_id in self.results: | |
715 | # we already |
|
715 | # we already | |
716 | continue |
|
716 | continue | |
717 | try: |
|
717 | try: | |
718 | raise error.EngineError("Engine %r died while running task %r"%(eid, msg_id)) |
|
718 | raise error.EngineError("Engine %r died while running task %r"%(eid, msg_id)) | |
719 | except: |
|
719 | except: | |
720 | content = error.wrap_exception() |
|
720 | content = error.wrap_exception() | |
721 | # build a fake message: |
|
721 | # build a fake message: | |
722 | msg = self.session.msg('apply_reply', content=content) |
|
722 | msg = self.session.msg('apply_reply', content=content) | |
723 | msg['parent_header']['msg_id'] = msg_id |
|
723 | msg['parent_header']['msg_id'] = msg_id | |
724 | msg['metadata']['engine'] = uuid |
|
724 | msg['metadata']['engine'] = uuid | |
725 | self._handle_apply_reply(msg) |
|
725 | self._handle_apply_reply(msg) | |
726 |
|
726 | |||
727 | def _handle_execute_reply(self, msg): |
|
727 | def _handle_execute_reply(self, msg): | |
728 | """Save the reply to an execute_request into our results. |
|
728 | """Save the reply to an execute_request into our results. | |
729 |
|
729 | |||
730 | execute messages are never actually used. apply is used instead. |
|
730 | execute messages are never actually used. apply is used instead. | |
731 | """ |
|
731 | """ | |
732 |
|
732 | |||
733 | parent = msg['parent_header'] |
|
733 | parent = msg['parent_header'] | |
734 | msg_id = parent['msg_id'] |
|
734 | msg_id = parent['msg_id'] | |
735 | if msg_id not in self.outstanding: |
|
735 | if msg_id not in self.outstanding: | |
736 | if msg_id in self.history: |
|
736 | if msg_id in self.history: | |
737 | print(("got stale result: %s"%msg_id)) |
|
737 | print(("got stale result: %s"%msg_id)) | |
738 | else: |
|
738 | else: | |
739 | print(("got unknown result: %s"%msg_id)) |
|
739 | print(("got unknown result: %s"%msg_id)) | |
740 | else: |
|
740 | else: | |
741 | self.outstanding.remove(msg_id) |
|
741 | self.outstanding.remove(msg_id) | |
742 |
|
742 | |||
743 | content = msg['content'] |
|
743 | content = msg['content'] | |
744 | header = msg['header'] |
|
744 | header = msg['header'] | |
745 |
|
745 | |||
746 | # construct metadata: |
|
746 | # construct metadata: | |
747 | md = self.metadata[msg_id] |
|
747 | md = self.metadata[msg_id] | |
748 | md.update(self._extract_metadata(msg)) |
|
748 | md.update(self._extract_metadata(msg)) | |
749 | # is this redundant? |
|
749 | # is this redundant? | |
750 | self.metadata[msg_id] = md |
|
750 | self.metadata[msg_id] = md | |
751 |
|
751 | |||
752 | e_outstanding = self._outstanding_dict[md['engine_uuid']] |
|
752 | e_outstanding = self._outstanding_dict[md['engine_uuid']] | |
753 | if msg_id in e_outstanding: |
|
753 | if msg_id in e_outstanding: | |
754 | e_outstanding.remove(msg_id) |
|
754 | e_outstanding.remove(msg_id) | |
755 |
|
755 | |||
756 | # construct result: |
|
756 | # construct result: | |
757 | if content['status'] == 'ok': |
|
757 | if content['status'] == 'ok': | |
758 | self.results[msg_id] = ExecuteReply(msg_id, content, md) |
|
758 | self.results[msg_id] = ExecuteReply(msg_id, content, md) | |
759 | elif content['status'] == 'aborted': |
|
759 | elif content['status'] == 'aborted': | |
760 | self.results[msg_id] = error.TaskAborted(msg_id) |
|
760 | self.results[msg_id] = error.TaskAborted(msg_id) | |
761 | elif content['status'] == 'resubmitted': |
|
761 | elif content['status'] == 'resubmitted': | |
762 | # TODO: handle resubmission |
|
762 | # TODO: handle resubmission | |
763 | pass |
|
763 | pass | |
764 | else: |
|
764 | else: | |
765 | self.results[msg_id] = self._unwrap_exception(content) |
|
765 | self.results[msg_id] = self._unwrap_exception(content) | |
766 |
|
766 | |||
767 | def _handle_apply_reply(self, msg): |
|
767 | def _handle_apply_reply(self, msg): | |
768 | """Save the reply to an apply_request into our results.""" |
|
768 | """Save the reply to an apply_request into our results.""" | |
769 | parent = msg['parent_header'] |
|
769 | parent = msg['parent_header'] | |
770 | msg_id = parent['msg_id'] |
|
770 | msg_id = parent['msg_id'] | |
771 | if msg_id not in self.outstanding: |
|
771 | if msg_id not in self.outstanding: | |
772 | if msg_id in self.history: |
|
772 | if msg_id in self.history: | |
773 | print(("got stale result: %s"%msg_id)) |
|
773 | print(("got stale result: %s"%msg_id)) | |
774 | print(self.results[msg_id]) |
|
774 | print(self.results[msg_id]) | |
775 | print(msg) |
|
775 | print(msg) | |
776 | else: |
|
776 | else: | |
777 | print(("got unknown result: %s"%msg_id)) |
|
777 | print(("got unknown result: %s"%msg_id)) | |
778 | else: |
|
778 | else: | |
779 | self.outstanding.remove(msg_id) |
|
779 | self.outstanding.remove(msg_id) | |
780 | content = msg['content'] |
|
780 | content = msg['content'] | |
781 | header = msg['header'] |
|
781 | header = msg['header'] | |
782 |
|
782 | |||
783 | # construct metadata: |
|
783 | # construct metadata: | |
784 | md = self.metadata[msg_id] |
|
784 | md = self.metadata[msg_id] | |
785 | md.update(self._extract_metadata(msg)) |
|
785 | md.update(self._extract_metadata(msg)) | |
786 | # is this redundant? |
|
786 | # is this redundant? | |
787 | self.metadata[msg_id] = md |
|
787 | self.metadata[msg_id] = md | |
788 |
|
788 | |||
789 | e_outstanding = self._outstanding_dict[md['engine_uuid']] |
|
789 | e_outstanding = self._outstanding_dict[md['engine_uuid']] | |
790 | if msg_id in e_outstanding: |
|
790 | if msg_id in e_outstanding: | |
791 | e_outstanding.remove(msg_id) |
|
791 | e_outstanding.remove(msg_id) | |
792 |
|
792 | |||
793 | # construct result: |
|
793 | # construct result: | |
794 | if content['status'] == 'ok': |
|
794 | if content['status'] == 'ok': | |
795 | self.results[msg_id] = serialize.unserialize_object(msg['buffers'])[0] |
|
795 | self.results[msg_id] = serialize.unserialize_object(msg['buffers'])[0] | |
796 | elif content['status'] == 'aborted': |
|
796 | elif content['status'] == 'aborted': | |
797 | self.results[msg_id] = error.TaskAborted(msg_id) |
|
797 | self.results[msg_id] = error.TaskAborted(msg_id) | |
798 | elif content['status'] == 'resubmitted': |
|
798 | elif content['status'] == 'resubmitted': | |
799 | # TODO: handle resubmission |
|
799 | # TODO: handle resubmission | |
800 | pass |
|
800 | pass | |
801 | else: |
|
801 | else: | |
802 | self.results[msg_id] = self._unwrap_exception(content) |
|
802 | self.results[msg_id] = self._unwrap_exception(content) | |
803 |
|
803 | |||
804 | def _flush_notifications(self): |
|
804 | def _flush_notifications(self): | |
805 | """Flush notifications of engine registrations waiting |
|
805 | """Flush notifications of engine registrations waiting | |
806 | in ZMQ queue.""" |
|
806 | in ZMQ queue.""" | |
807 | idents,msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK) |
|
807 | idents,msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK) | |
808 | while msg is not None: |
|
808 | while msg is not None: | |
809 | if self.debug: |
|
809 | if self.debug: | |
810 | pprint(msg) |
|
810 | pprint(msg) | |
811 | msg_type = msg['header']['msg_type'] |
|
811 | msg_type = msg['header']['msg_type'] | |
812 | handler = self._notification_handlers.get(msg_type, None) |
|
812 | handler = self._notification_handlers.get(msg_type, None) | |
813 | if handler is None: |
|
813 | if handler is None: | |
814 | raise Exception("Unhandled message type: %s" % msg_type) |
|
814 | raise Exception("Unhandled message type: %s" % msg_type) | |
815 | else: |
|
815 | else: | |
816 | handler(msg) |
|
816 | handler(msg) | |
817 | idents,msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK) |
|
817 | idents,msg = self.session.recv(self._notification_socket, mode=zmq.NOBLOCK) | |
818 |
|
818 | |||
819 | def _flush_results(self, sock): |
|
819 | def _flush_results(self, sock): | |
820 | """Flush task or queue results waiting in ZMQ queue.""" |
|
820 | """Flush task or queue results waiting in ZMQ queue.""" | |
821 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
821 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
822 | while msg is not None: |
|
822 | while msg is not None: | |
823 | if self.debug: |
|
823 | if self.debug: | |
824 | pprint(msg) |
|
824 | pprint(msg) | |
825 | msg_type = msg['header']['msg_type'] |
|
825 | msg_type = msg['header']['msg_type'] | |
826 | handler = self._queue_handlers.get(msg_type, None) |
|
826 | handler = self._queue_handlers.get(msg_type, None) | |
827 | if handler is None: |
|
827 | if handler is None: | |
828 | raise Exception("Unhandled message type: %s" % msg_type) |
|
828 | raise Exception("Unhandled message type: %s" % msg_type) | |
829 | else: |
|
829 | else: | |
830 | handler(msg) |
|
830 | handler(msg) | |
831 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
831 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
832 |
|
832 | |||
833 | def _flush_control(self, sock): |
|
833 | def _flush_control(self, sock): | |
834 | """Flush replies from the control channel waiting |
|
834 | """Flush replies from the control channel waiting | |
835 | in the ZMQ queue. |
|
835 | in the ZMQ queue. | |
836 |
|
836 | |||
837 | Currently: ignore them.""" |
|
837 | Currently: ignore them.""" | |
838 | if self._ignored_control_replies <= 0: |
|
838 | if self._ignored_control_replies <= 0: | |
839 | return |
|
839 | return | |
840 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
840 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
841 | while msg is not None: |
|
841 | while msg is not None: | |
842 | self._ignored_control_replies -= 1 |
|
842 | self._ignored_control_replies -= 1 | |
843 | if self.debug: |
|
843 | if self.debug: | |
844 | pprint(msg) |
|
844 | pprint(msg) | |
845 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
845 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
846 |
|
846 | |||
847 | def _flush_ignored_control(self): |
|
847 | def _flush_ignored_control(self): | |
848 | """flush ignored control replies""" |
|
848 | """flush ignored control replies""" | |
849 | while self._ignored_control_replies > 0: |
|
849 | while self._ignored_control_replies > 0: | |
850 | self.session.recv(self._control_socket) |
|
850 | self.session.recv(self._control_socket) | |
851 | self._ignored_control_replies -= 1 |
|
851 | self._ignored_control_replies -= 1 | |
852 |
|
852 | |||
853 | def _flush_ignored_hub_replies(self): |
|
853 | def _flush_ignored_hub_replies(self): | |
854 | ident,msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK) |
|
854 | ident,msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK) | |
855 | while msg is not None: |
|
855 | while msg is not None: | |
856 | ident,msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK) |
|
856 | ident,msg = self.session.recv(self._query_socket, mode=zmq.NOBLOCK) | |
857 |
|
857 | |||
858 | def _flush_iopub(self, sock): |
|
858 | def _flush_iopub(self, sock): | |
859 | """Flush replies from the iopub channel waiting |
|
859 | """Flush replies from the iopub channel waiting | |
860 | in the ZMQ queue. |
|
860 | in the ZMQ queue. | |
861 | """ |
|
861 | """ | |
862 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
862 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
863 | while msg is not None: |
|
863 | while msg is not None: | |
864 | if self.debug: |
|
864 | if self.debug: | |
865 | pprint(msg) |
|
865 | pprint(msg) | |
866 | parent = msg['parent_header'] |
|
866 | parent = msg['parent_header'] | |
867 | # ignore IOPub messages with no parent. |
|
867 | # ignore IOPub messages with no parent. | |
868 | # Caused by print statements or warnings from before the first execution. |
|
868 | # Caused by print statements or warnings from before the first execution. | |
869 | if not parent: |
|
869 | if not parent: | |
870 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
870 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
871 | continue |
|
871 | continue | |
872 | msg_id = parent['msg_id'] |
|
872 | msg_id = parent['msg_id'] | |
873 | content = msg['content'] |
|
873 | content = msg['content'] | |
874 | header = msg['header'] |
|
874 | header = msg['header'] | |
875 | msg_type = msg['header']['msg_type'] |
|
875 | msg_type = msg['header']['msg_type'] | |
876 |
|
876 | |||
877 | # init metadata: |
|
877 | # init metadata: | |
878 | md = self.metadata[msg_id] |
|
878 | md = self.metadata[msg_id] | |
879 |
|
879 | |||
880 | if msg_type == 'stream': |
|
880 | if msg_type == 'stream': | |
881 | name = content['name'] |
|
881 | name = content['name'] | |
882 | s = md[name] or '' |
|
882 | s = md[name] or '' | |
883 | md[name] = s + content['data'] |
|
883 | md[name] = s + content['data'] | |
884 | elif msg_type == 'pyerr': |
|
884 | elif msg_type == 'pyerr': | |
885 | md.update({'pyerr' : self._unwrap_exception(content)}) |
|
885 | md.update({'pyerr' : self._unwrap_exception(content)}) | |
886 | elif msg_type == 'pyin': |
|
886 | elif msg_type == 'pyin': | |
887 | md.update({'pyin' : content['code']}) |
|
887 | md.update({'pyin' : content['code']}) | |
888 | elif msg_type == 'display_data': |
|
888 | elif msg_type == 'display_data': | |
889 | md['outputs'].append(content) |
|
889 | md['outputs'].append(content) | |
890 | elif msg_type == 'pyout': |
|
890 | elif msg_type == 'pyout': | |
891 | md['pyout'] = content |
|
891 | md['pyout'] = content | |
892 | elif msg_type == 'data_message': |
|
892 | elif msg_type == 'data_message': | |
893 | data, remainder = serialize.unserialize_object(msg['buffers']) |
|
893 | data, remainder = serialize.unserialize_object(msg['buffers']) | |
894 | md['data'].update(data) |
|
894 | md['data'].update(data) | |
895 | elif msg_type == 'status': |
|
895 | elif msg_type == 'status': | |
896 | # idle message comes after all outputs |
|
896 | # idle message comes after all outputs | |
897 | if content['execution_state'] == 'idle': |
|
897 | if content['execution_state'] == 'idle': | |
898 | md['outputs_ready'] = True |
|
898 | md['outputs_ready'] = True | |
899 | else: |
|
899 | else: | |
900 | # unhandled msg_type (status, etc.) |
|
900 | # unhandled msg_type (status, etc.) | |
901 | pass |
|
901 | pass | |
902 |
|
902 | |||
903 | # reduntant? |
|
903 | # reduntant? | |
904 | self.metadata[msg_id] = md |
|
904 | self.metadata[msg_id] = md | |
905 |
|
905 | |||
906 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) |
|
906 | idents,msg = self.session.recv(sock, mode=zmq.NOBLOCK) | |
907 |
|
907 | |||
908 | #-------------------------------------------------------------------------- |
|
908 | #-------------------------------------------------------------------------- | |
909 | # len, getitem |
|
909 | # len, getitem | |
910 | #-------------------------------------------------------------------------- |
|
910 | #-------------------------------------------------------------------------- | |
911 |
|
911 | |||
912 | def __len__(self): |
|
912 | def __len__(self): | |
913 | """len(client) returns # of engines.""" |
|
913 | """len(client) returns # of engines.""" | |
914 | return len(self.ids) |
|
914 | return len(self.ids) | |
915 |
|
915 | |||
916 | def __getitem__(self, key): |
|
916 | def __getitem__(self, key): | |
917 | """index access returns DirectView multiplexer objects |
|
917 | """index access returns DirectView multiplexer objects | |
918 |
|
918 | |||
919 | Must be int, slice, or list/tuple/xrange of ints""" |
|
919 | Must be int, slice, or list/tuple/xrange of ints""" | |
920 | if not isinstance(key, (int, slice, tuple, list, xrange)): |
|
920 | if not isinstance(key, (int, slice, tuple, list, xrange)): | |
921 | raise TypeError("key by int/slice/iterable of ints only, not %s"%(type(key))) |
|
921 | raise TypeError("key by int/slice/iterable of ints only, not %s"%(type(key))) | |
922 | else: |
|
922 | else: | |
923 | return self.direct_view(key) |
|
923 | return self.direct_view(key) | |
924 |
|
924 | |||
925 | #-------------------------------------------------------------------------- |
|
925 | #-------------------------------------------------------------------------- | |
926 | # Begin public methods |
|
926 | # Begin public methods | |
927 | #-------------------------------------------------------------------------- |
|
927 | #-------------------------------------------------------------------------- | |
928 |
|
928 | |||
929 | @property |
|
929 | @property | |
930 | def ids(self): |
|
930 | def ids(self): | |
931 | """Always up-to-date ids property.""" |
|
931 | """Always up-to-date ids property.""" | |
932 | self._flush_notifications() |
|
932 | self._flush_notifications() | |
933 | # always copy: |
|
933 | # always copy: | |
934 | return list(self._ids) |
|
934 | return list(self._ids) | |
935 |
|
935 | |||
936 | def activate(self, targets='all', suffix=''): |
|
936 | def activate(self, targets='all', suffix=''): | |
937 | """Create a DirectView and register it with IPython magics |
|
937 | """Create a DirectView and register it with IPython magics | |
938 |
|
938 | |||
939 | Defines the magics `%px, %autopx, %pxresult, %%px` |
|
939 | Defines the magics `%px, %autopx, %pxresult, %%px` | |
940 |
|
940 | |||
941 | Parameters |
|
941 | Parameters | |
942 | ---------- |
|
942 | ---------- | |
943 |
|
943 | |||
944 | targets: int, list of ints, or 'all' |
|
944 | targets: int, list of ints, or 'all' | |
945 | The engines on which the view's magics will run |
|
945 | The engines on which the view's magics will run | |
946 | suffix: str [default: ''] |
|
946 | suffix: str [default: ''] | |
947 | The suffix, if any, for the magics. This allows you to have |
|
947 | The suffix, if any, for the magics. This allows you to have | |
948 | multiple views associated with parallel magics at the same time. |
|
948 | multiple views associated with parallel magics at the same time. | |
949 |
|
949 | |||
950 | e.g. ``rc.activate(targets=0, suffix='0')`` will give you |
|
950 | e.g. ``rc.activate(targets=0, suffix='0')`` will give you | |
951 | the magics ``%px0``, ``%pxresult0``, etc. for running magics just |
|
951 | the magics ``%px0``, ``%pxresult0``, etc. for running magics just | |
952 | on engine 0. |
|
952 | on engine 0. | |
953 | """ |
|
953 | """ | |
954 | view = self.direct_view(targets) |
|
954 | view = self.direct_view(targets) | |
955 | view.block = True |
|
955 | view.block = True | |
956 | view.activate(suffix) |
|
956 | view.activate(suffix) | |
957 | return view |
|
957 | return view | |
958 |
|
958 | |||
959 | def close(self, linger=None): |
|
959 | def close(self, linger=None): | |
960 | """Close my zmq Sockets |
|
960 | """Close my zmq Sockets | |
961 |
|
961 | |||
962 | If `linger`, set the zmq LINGER socket option, |
|
962 | If `linger`, set the zmq LINGER socket option, | |
963 | which allows discarding of messages. |
|
963 | which allows discarding of messages. | |
964 | """ |
|
964 | """ | |
965 | if self._closed: |
|
965 | if self._closed: | |
966 | return |
|
966 | return | |
967 | self.stop_spin_thread() |
|
967 | self.stop_spin_thread() | |
968 | snames = [ trait for trait in self.trait_names() if trait.endswith("socket") ] |
|
968 | snames = [ trait for trait in self.trait_names() if trait.endswith("socket") ] | |
969 | for name in snames: |
|
969 | for name in snames: | |
970 | socket = getattr(self, name) |
|
970 | socket = getattr(self, name) | |
971 | if socket is not None and not socket.closed: |
|
971 | if socket is not None and not socket.closed: | |
972 | if linger is not None: |
|
972 | if linger is not None: | |
973 | socket.close(linger=linger) |
|
973 | socket.close(linger=linger) | |
974 | else: |
|
974 | else: | |
975 | socket.close() |
|
975 | socket.close() | |
976 | self._closed = True |
|
976 | self._closed = True | |
977 |
|
977 | |||
978 | def _spin_every(self, interval=1): |
|
978 | def _spin_every(self, interval=1): | |
979 | """target func for use in spin_thread""" |
|
979 | """target func for use in spin_thread""" | |
980 | while True: |
|
980 | while True: | |
981 | if self._stop_spinning.is_set(): |
|
981 | if self._stop_spinning.is_set(): | |
982 | return |
|
982 | return | |
983 | time.sleep(interval) |
|
983 | time.sleep(interval) | |
984 | self.spin() |
|
984 | self.spin() | |
985 |
|
985 | |||
986 | def spin_thread(self, interval=1): |
|
986 | def spin_thread(self, interval=1): | |
987 | """call Client.spin() in a background thread on some regular interval |
|
987 | """call Client.spin() in a background thread on some regular interval | |
988 |
|
988 | |||
989 | This helps ensure that messages don't pile up too much in the zmq queue |
|
989 | This helps ensure that messages don't pile up too much in the zmq queue | |
990 | while you are working on other things, or just leaving an idle terminal. |
|
990 | while you are working on other things, or just leaving an idle terminal. | |
991 |
|
991 | |||
992 | It also helps limit potential padding of the `received` timestamp |
|
992 | It also helps limit potential padding of the `received` timestamp | |
993 | on AsyncResult objects, used for timings. |
|
993 | on AsyncResult objects, used for timings. | |
994 |
|
994 | |||
995 | Parameters |
|
995 | Parameters | |
996 | ---------- |
|
996 | ---------- | |
997 |
|
997 | |||
998 | interval : float, optional |
|
998 | interval : float, optional | |
999 | The interval on which to spin the client in the background thread |
|
999 | The interval on which to spin the client in the background thread | |
1000 | (simply passed to time.sleep). |
|
1000 | (simply passed to time.sleep). | |
1001 |
|
1001 | |||
1002 | Notes |
|
1002 | Notes | |
1003 | ----- |
|
1003 | ----- | |
1004 |
|
1004 | |||
1005 | For precision timing, you may want to use this method to put a bound |
|
1005 | For precision timing, you may want to use this method to put a bound | |
1006 | on the jitter (in seconds) in `received` timestamps used |
|
1006 | on the jitter (in seconds) in `received` timestamps used | |
1007 | in AsyncResult.wall_time. |
|
1007 | in AsyncResult.wall_time. | |
1008 |
|
1008 | |||
1009 | """ |
|
1009 | """ | |
1010 | if self._spin_thread is not None: |
|
1010 | if self._spin_thread is not None: | |
1011 | self.stop_spin_thread() |
|
1011 | self.stop_spin_thread() | |
1012 | self._stop_spinning.clear() |
|
1012 | self._stop_spinning.clear() | |
1013 | self._spin_thread = Thread(target=self._spin_every, args=(interval,)) |
|
1013 | self._spin_thread = Thread(target=self._spin_every, args=(interval,)) | |
1014 | self._spin_thread.daemon = True |
|
1014 | self._spin_thread.daemon = True | |
1015 | self._spin_thread.start() |
|
1015 | self._spin_thread.start() | |
1016 |
|
1016 | |||
1017 | def stop_spin_thread(self): |
|
1017 | def stop_spin_thread(self): | |
1018 | """stop background spin_thread, if any""" |
|
1018 | """stop background spin_thread, if any""" | |
1019 | if self._spin_thread is not None: |
|
1019 | if self._spin_thread is not None: | |
1020 | self._stop_spinning.set() |
|
1020 | self._stop_spinning.set() | |
1021 | self._spin_thread.join() |
|
1021 | self._spin_thread.join() | |
1022 | self._spin_thread = None |
|
1022 | self._spin_thread = None | |
1023 |
|
1023 | |||
1024 | def spin(self): |
|
1024 | def spin(self): | |
1025 | """Flush any registration notifications and execution results |
|
1025 | """Flush any registration notifications and execution results | |
1026 | waiting in the ZMQ queue. |
|
1026 | waiting in the ZMQ queue. | |
1027 | """ |
|
1027 | """ | |
1028 | if self._notification_socket: |
|
1028 | if self._notification_socket: | |
1029 | self._flush_notifications() |
|
1029 | self._flush_notifications() | |
1030 | if self._iopub_socket: |
|
1030 | if self._iopub_socket: | |
1031 | self._flush_iopub(self._iopub_socket) |
|
1031 | self._flush_iopub(self._iopub_socket) | |
1032 | if self._mux_socket: |
|
1032 | if self._mux_socket: | |
1033 | self._flush_results(self._mux_socket) |
|
1033 | self._flush_results(self._mux_socket) | |
1034 | if self._task_socket: |
|
1034 | if self._task_socket: | |
1035 | self._flush_results(self._task_socket) |
|
1035 | self._flush_results(self._task_socket) | |
1036 | if self._control_socket: |
|
1036 | if self._control_socket: | |
1037 | self._flush_control(self._control_socket) |
|
1037 | self._flush_control(self._control_socket) | |
1038 | if self._query_socket: |
|
1038 | if self._query_socket: | |
1039 | self._flush_ignored_hub_replies() |
|
1039 | self._flush_ignored_hub_replies() | |
1040 |
|
1040 | |||
1041 | def wait(self, jobs=None, timeout=-1): |
|
1041 | def wait(self, jobs=None, timeout=-1): | |
1042 | """waits on one or more `jobs`, for up to `timeout` seconds. |
|
1042 | """waits on one or more `jobs`, for up to `timeout` seconds. | |
1043 |
|
1043 | |||
1044 | Parameters |
|
1044 | Parameters | |
1045 | ---------- |
|
1045 | ---------- | |
1046 |
|
1046 | |||
1047 | jobs : int, str, or list of ints and/or strs, or one or more AsyncResult objects |
|
1047 | jobs : int, str, or list of ints and/or strs, or one or more AsyncResult objects | |
1048 | ints are indices to self.history |
|
1048 | ints are indices to self.history | |
1049 | strs are msg_ids |
|
1049 | strs are msg_ids | |
1050 | default: wait on all outstanding messages |
|
1050 | default: wait on all outstanding messages | |
1051 | timeout : float |
|
1051 | timeout : float | |
1052 | a time in seconds, after which to give up. |
|
1052 | a time in seconds, after which to give up. | |
1053 | default is -1, which means no timeout |
|
1053 | default is -1, which means no timeout | |
1054 |
|
1054 | |||
1055 | Returns |
|
1055 | Returns | |
1056 | ------- |
|
1056 | ------- | |
1057 |
|
1057 | |||
1058 | True : when all msg_ids are done |
|
1058 | True : when all msg_ids are done | |
1059 | False : timeout reached, some msg_ids still outstanding |
|
1059 | False : timeout reached, some msg_ids still outstanding | |
1060 | """ |
|
1060 | """ | |
1061 | tic = time.time() |
|
1061 | tic = time.time() | |
1062 | if jobs is None: |
|
1062 | if jobs is None: | |
1063 | theids = self.outstanding |
|
1063 | theids = self.outstanding | |
1064 | else: |
|
1064 | else: | |
1065 | if isinstance(jobs, string_types + (int, AsyncResult)): |
|
1065 | if isinstance(jobs, string_types + (int, AsyncResult)): | |
1066 | jobs = [jobs] |
|
1066 | jobs = [jobs] | |
1067 | theids = set() |
|
1067 | theids = set() | |
1068 | for job in jobs: |
|
1068 | for job in jobs: | |
1069 | if isinstance(job, int): |
|
1069 | if isinstance(job, int): | |
1070 | # index access |
|
1070 | # index access | |
1071 | job = self.history[job] |
|
1071 | job = self.history[job] | |
1072 | elif isinstance(job, AsyncResult): |
|
1072 | elif isinstance(job, AsyncResult): | |
1073 | map(theids.add, job.msg_ids) |
|
1073 | map(theids.add, job.msg_ids) | |
1074 | continue |
|
1074 | continue | |
1075 | theids.add(job) |
|
1075 | theids.add(job) | |
1076 | if not theids.intersection(self.outstanding): |
|
1076 | if not theids.intersection(self.outstanding): | |
1077 | return True |
|
1077 | return True | |
1078 | self.spin() |
|
1078 | self.spin() | |
1079 | while theids.intersection(self.outstanding): |
|
1079 | while theids.intersection(self.outstanding): | |
1080 | if timeout >= 0 and ( time.time()-tic ) > timeout: |
|
1080 | if timeout >= 0 and ( time.time()-tic ) > timeout: | |
1081 | break |
|
1081 | break | |
1082 | time.sleep(1e-3) |
|
1082 | time.sleep(1e-3) | |
1083 | self.spin() |
|
1083 | self.spin() | |
1084 | return len(theids.intersection(self.outstanding)) == 0 |
|
1084 | return len(theids.intersection(self.outstanding)) == 0 | |
1085 |
|
1085 | |||
1086 | #-------------------------------------------------------------------------- |
|
1086 | #-------------------------------------------------------------------------- | |
1087 | # Control methods |
|
1087 | # Control methods | |
1088 | #-------------------------------------------------------------------------- |
|
1088 | #-------------------------------------------------------------------------- | |
1089 |
|
1089 | |||
1090 | @spin_first |
|
1090 | @spin_first | |
1091 | def clear(self, targets=None, block=None): |
|
1091 | def clear(self, targets=None, block=None): | |
1092 | """Clear the namespace in target(s).""" |
|
1092 | """Clear the namespace in target(s).""" | |
1093 | block = self.block if block is None else block |
|
1093 | block = self.block if block is None else block | |
1094 | targets = self._build_targets(targets)[0] |
|
1094 | targets = self._build_targets(targets)[0] | |
1095 | for t in targets: |
|
1095 | for t in targets: | |
1096 | self.session.send(self._control_socket, 'clear_request', content={}, ident=t) |
|
1096 | self.session.send(self._control_socket, 'clear_request', content={}, ident=t) | |
1097 | error = False |
|
1097 | error = False | |
1098 | if block: |
|
1098 | if block: | |
1099 | self._flush_ignored_control() |
|
1099 | self._flush_ignored_control() | |
1100 | for i in range(len(targets)): |
|
1100 | for i in range(len(targets)): | |
1101 | idents,msg = self.session.recv(self._control_socket,0) |
|
1101 | idents,msg = self.session.recv(self._control_socket,0) | |
1102 | if self.debug: |
|
1102 | if self.debug: | |
1103 | pprint(msg) |
|
1103 | pprint(msg) | |
1104 | if msg['content']['status'] != 'ok': |
|
1104 | if msg['content']['status'] != 'ok': | |
1105 | error = self._unwrap_exception(msg['content']) |
|
1105 | error = self._unwrap_exception(msg['content']) | |
1106 | else: |
|
1106 | else: | |
1107 | self._ignored_control_replies += len(targets) |
|
1107 | self._ignored_control_replies += len(targets) | |
1108 | if error: |
|
1108 | if error: | |
1109 | raise error |
|
1109 | raise error | |
1110 |
|
1110 | |||
1111 |
|
1111 | |||
1112 | @spin_first |
|
1112 | @spin_first | |
1113 | def abort(self, jobs=None, targets=None, block=None): |
|
1113 | def abort(self, jobs=None, targets=None, block=None): | |
1114 | """Abort specific jobs from the execution queues of target(s). |
|
1114 | """Abort specific jobs from the execution queues of target(s). | |
1115 |
|
1115 | |||
1116 | This is a mechanism to prevent jobs that have already been submitted |
|
1116 | This is a mechanism to prevent jobs that have already been submitted | |
1117 | from executing. |
|
1117 | from executing. | |
1118 |
|
1118 | |||
1119 | Parameters |
|
1119 | Parameters | |
1120 | ---------- |
|
1120 | ---------- | |
1121 |
|
1121 | |||
1122 | jobs : msg_id, list of msg_ids, or AsyncResult |
|
1122 | jobs : msg_id, list of msg_ids, or AsyncResult | |
1123 | The jobs to be aborted |
|
1123 | The jobs to be aborted | |
1124 |
|
1124 | |||
1125 | If unspecified/None: abort all outstanding jobs. |
|
1125 | If unspecified/None: abort all outstanding jobs. | |
1126 |
|
1126 | |||
1127 | """ |
|
1127 | """ | |
1128 | block = self.block if block is None else block |
|
1128 | block = self.block if block is None else block | |
1129 | jobs = jobs if jobs is not None else list(self.outstanding) |
|
1129 | jobs = jobs if jobs is not None else list(self.outstanding) | |
1130 | targets = self._build_targets(targets)[0] |
|
1130 | targets = self._build_targets(targets)[0] | |
1131 |
|
1131 | |||
1132 | msg_ids = [] |
|
1132 | msg_ids = [] | |
1133 | if isinstance(jobs, string_types + (AsyncResult,)): |
|
1133 | if isinstance(jobs, string_types + (AsyncResult,)): | |
1134 | jobs = [jobs] |
|
1134 | jobs = [jobs] | |
1135 | bad_ids = filter(lambda obj: not isinstance(obj, string_types + (AsyncResult,)), jobs) |
|
1135 | bad_ids = filter(lambda obj: not isinstance(obj, string_types + (AsyncResult,)), jobs) | |
1136 | if bad_ids: |
|
1136 | if bad_ids: | |
1137 | raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0]) |
|
1137 | raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0]) | |
1138 | for j in jobs: |
|
1138 | for j in jobs: | |
1139 | if isinstance(j, AsyncResult): |
|
1139 | if isinstance(j, AsyncResult): | |
1140 | msg_ids.extend(j.msg_ids) |
|
1140 | msg_ids.extend(j.msg_ids) | |
1141 | else: |
|
1141 | else: | |
1142 | msg_ids.append(j) |
|
1142 | msg_ids.append(j) | |
1143 | content = dict(msg_ids=msg_ids) |
|
1143 | content = dict(msg_ids=msg_ids) | |
1144 | for t in targets: |
|
1144 | for t in targets: | |
1145 | self.session.send(self._control_socket, 'abort_request', |
|
1145 | self.session.send(self._control_socket, 'abort_request', | |
1146 | content=content, ident=t) |
|
1146 | content=content, ident=t) | |
1147 | error = False |
|
1147 | error = False | |
1148 | if block: |
|
1148 | if block: | |
1149 | self._flush_ignored_control() |
|
1149 | self._flush_ignored_control() | |
1150 | for i in range(len(targets)): |
|
1150 | for i in range(len(targets)): | |
1151 | idents,msg = self.session.recv(self._control_socket,0) |
|
1151 | idents,msg = self.session.recv(self._control_socket,0) | |
1152 | if self.debug: |
|
1152 | if self.debug: | |
1153 | pprint(msg) |
|
1153 | pprint(msg) | |
1154 | if msg['content']['status'] != 'ok': |
|
1154 | if msg['content']['status'] != 'ok': | |
1155 | error = self._unwrap_exception(msg['content']) |
|
1155 | error = self._unwrap_exception(msg['content']) | |
1156 | else: |
|
1156 | else: | |
1157 | self._ignored_control_replies += len(targets) |
|
1157 | self._ignored_control_replies += len(targets) | |
1158 | if error: |
|
1158 | if error: | |
1159 | raise error |
|
1159 | raise error | |
1160 |
|
1160 | |||
1161 | @spin_first |
|
1161 | @spin_first | |
1162 | def shutdown(self, targets='all', restart=False, hub=False, block=None): |
|
1162 | def shutdown(self, targets='all', restart=False, hub=False, block=None): | |
1163 | """Terminates one or more engine processes, optionally including the hub. |
|
1163 | """Terminates one or more engine processes, optionally including the hub. | |
1164 |
|
1164 | |||
1165 | Parameters |
|
1165 | Parameters | |
1166 | ---------- |
|
1166 | ---------- | |
1167 |
|
1167 | |||
1168 | targets: list of ints or 'all' [default: all] |
|
1168 | targets: list of ints or 'all' [default: all] | |
1169 | Which engines to shutdown. |
|
1169 | Which engines to shutdown. | |
1170 | hub: bool [default: False] |
|
1170 | hub: bool [default: False] | |
1171 | Whether to include the Hub. hub=True implies targets='all'. |
|
1171 | Whether to include the Hub. hub=True implies targets='all'. | |
1172 | block: bool [default: self.block] |
|
1172 | block: bool [default: self.block] | |
1173 | Whether to wait for clean shutdown replies or not. |
|
1173 | Whether to wait for clean shutdown replies or not. | |
1174 | restart: bool [default: False] |
|
1174 | restart: bool [default: False] | |
1175 | NOT IMPLEMENTED |
|
1175 | NOT IMPLEMENTED | |
1176 | whether to restart engines after shutting them down. |
|
1176 | whether to restart engines after shutting them down. | |
1177 | """ |
|
1177 | """ | |
1178 | from IPython.parallel.error import NoEnginesRegistered |
|
1178 | from IPython.parallel.error import NoEnginesRegistered | |
1179 | if restart: |
|
1179 | if restart: | |
1180 | raise NotImplementedError("Engine restart is not yet implemented") |
|
1180 | raise NotImplementedError("Engine restart is not yet implemented") | |
1181 |
|
1181 | |||
1182 | block = self.block if block is None else block |
|
1182 | block = self.block if block is None else block | |
1183 | if hub: |
|
1183 | if hub: | |
1184 | targets = 'all' |
|
1184 | targets = 'all' | |
1185 | try: |
|
1185 | try: | |
1186 | targets = self._build_targets(targets)[0] |
|
1186 | targets = self._build_targets(targets)[0] | |
1187 | except NoEnginesRegistered: |
|
1187 | except NoEnginesRegistered: | |
1188 | targets = [] |
|
1188 | targets = [] | |
1189 | for t in targets: |
|
1189 | for t in targets: | |
1190 | self.session.send(self._control_socket, 'shutdown_request', |
|
1190 | self.session.send(self._control_socket, 'shutdown_request', | |
1191 | content={'restart':restart},ident=t) |
|
1191 | content={'restart':restart},ident=t) | |
1192 | error = False |
|
1192 | error = False | |
1193 | if block or hub: |
|
1193 | if block or hub: | |
1194 | self._flush_ignored_control() |
|
1194 | self._flush_ignored_control() | |
1195 | for i in range(len(targets)): |
|
1195 | for i in range(len(targets)): | |
1196 | idents,msg = self.session.recv(self._control_socket, 0) |
|
1196 | idents,msg = self.session.recv(self._control_socket, 0) | |
1197 | if self.debug: |
|
1197 | if self.debug: | |
1198 | pprint(msg) |
|
1198 | pprint(msg) | |
1199 | if msg['content']['status'] != 'ok': |
|
1199 | if msg['content']['status'] != 'ok': | |
1200 | error = self._unwrap_exception(msg['content']) |
|
1200 | error = self._unwrap_exception(msg['content']) | |
1201 | else: |
|
1201 | else: | |
1202 | self._ignored_control_replies += len(targets) |
|
1202 | self._ignored_control_replies += len(targets) | |
1203 |
|
1203 | |||
1204 | if hub: |
|
1204 | if hub: | |
1205 | time.sleep(0.25) |
|
1205 | time.sleep(0.25) | |
1206 | self.session.send(self._query_socket, 'shutdown_request') |
|
1206 | self.session.send(self._query_socket, 'shutdown_request') | |
1207 | idents,msg = self.session.recv(self._query_socket, 0) |
|
1207 | idents,msg = self.session.recv(self._query_socket, 0) | |
1208 | if self.debug: |
|
1208 | if self.debug: | |
1209 | pprint(msg) |
|
1209 | pprint(msg) | |
1210 | if msg['content']['status'] != 'ok': |
|
1210 | if msg['content']['status'] != 'ok': | |
1211 | error = self._unwrap_exception(msg['content']) |
|
1211 | error = self._unwrap_exception(msg['content']) | |
1212 |
|
1212 | |||
1213 | if error: |
|
1213 | if error: | |
1214 | raise error |
|
1214 | raise error | |
1215 |
|
1215 | |||
1216 | #-------------------------------------------------------------------------- |
|
1216 | #-------------------------------------------------------------------------- | |
1217 | # Execution related methods |
|
1217 | # Execution related methods | |
1218 | #-------------------------------------------------------------------------- |
|
1218 | #-------------------------------------------------------------------------- | |
1219 |
|
1219 | |||
1220 | def _maybe_raise(self, result): |
|
1220 | def _maybe_raise(self, result): | |
1221 | """wrapper for maybe raising an exception if apply failed.""" |
|
1221 | """wrapper for maybe raising an exception if apply failed.""" | |
1222 | if isinstance(result, error.RemoteError): |
|
1222 | if isinstance(result, error.RemoteError): | |
1223 | raise result |
|
1223 | raise result | |
1224 |
|
1224 | |||
1225 | return result |
|
1225 | return result | |
1226 |
|
1226 | |||
1227 | def send_apply_request(self, socket, f, args=None, kwargs=None, metadata=None, track=False, |
|
1227 | def send_apply_request(self, socket, f, args=None, kwargs=None, metadata=None, track=False, | |
1228 | ident=None): |
|
1228 | ident=None): | |
1229 | """construct and send an apply message via a socket. |
|
1229 | """construct and send an apply message via a socket. | |
1230 |
|
1230 | |||
1231 | This is the principal method with which all engine execution is performed by views. |
|
1231 | This is the principal method with which all engine execution is performed by views. | |
1232 | """ |
|
1232 | """ | |
1233 |
|
1233 | |||
1234 | if self._closed: |
|
1234 | if self._closed: | |
1235 | raise RuntimeError("Client cannot be used after its sockets have been closed") |
|
1235 | raise RuntimeError("Client cannot be used after its sockets have been closed") | |
1236 |
|
1236 | |||
1237 | # defaults: |
|
1237 | # defaults: | |
1238 | args = args if args is not None else [] |
|
1238 | args = args if args is not None else [] | |
1239 | kwargs = kwargs if kwargs is not None else {} |
|
1239 | kwargs = kwargs if kwargs is not None else {} | |
1240 | metadata = metadata if metadata is not None else {} |
|
1240 | metadata = metadata if metadata is not None else {} | |
1241 |
|
1241 | |||
1242 | # validate arguments |
|
1242 | # validate arguments | |
1243 | if not callable(f) and not isinstance(f, Reference): |
|
1243 | if not callable(f) and not isinstance(f, Reference): | |
1244 | raise TypeError("f must be callable, not %s"%type(f)) |
|
1244 | raise TypeError("f must be callable, not %s"%type(f)) | |
1245 | if not isinstance(args, (tuple, list)): |
|
1245 | if not isinstance(args, (tuple, list)): | |
1246 | raise TypeError("args must be tuple or list, not %s"%type(args)) |
|
1246 | raise TypeError("args must be tuple or list, not %s"%type(args)) | |
1247 | if not isinstance(kwargs, dict): |
|
1247 | if not isinstance(kwargs, dict): | |
1248 | raise TypeError("kwargs must be dict, not %s"%type(kwargs)) |
|
1248 | raise TypeError("kwargs must be dict, not %s"%type(kwargs)) | |
1249 | if not isinstance(metadata, dict): |
|
1249 | if not isinstance(metadata, dict): | |
1250 | raise TypeError("metadata must be dict, not %s"%type(metadata)) |
|
1250 | raise TypeError("metadata must be dict, not %s"%type(metadata)) | |
1251 |
|
1251 | |||
1252 | bufs = serialize.pack_apply_message(f, args, kwargs, |
|
1252 | bufs = serialize.pack_apply_message(f, args, kwargs, | |
1253 | buffer_threshold=self.session.buffer_threshold, |
|
1253 | buffer_threshold=self.session.buffer_threshold, | |
1254 | item_threshold=self.session.item_threshold, |
|
1254 | item_threshold=self.session.item_threshold, | |
1255 | ) |
|
1255 | ) | |
1256 |
|
1256 | |||
1257 | msg = self.session.send(socket, "apply_request", buffers=bufs, ident=ident, |
|
1257 | msg = self.session.send(socket, "apply_request", buffers=bufs, ident=ident, | |
1258 | metadata=metadata, track=track) |
|
1258 | metadata=metadata, track=track) | |
1259 |
|
1259 | |||
1260 | msg_id = msg['header']['msg_id'] |
|
1260 | msg_id = msg['header']['msg_id'] | |
1261 | self.outstanding.add(msg_id) |
|
1261 | self.outstanding.add(msg_id) | |
1262 | if ident: |
|
1262 | if ident: | |
1263 | # possibly routed to a specific engine |
|
1263 | # possibly routed to a specific engine | |
1264 | if isinstance(ident, list): |
|
1264 | if isinstance(ident, list): | |
1265 | ident = ident[-1] |
|
1265 | ident = ident[-1] | |
1266 | if ident in self._engines.values(): |
|
1266 | if ident in self._engines.values(): | |
1267 | # save for later, in case of engine death |
|
1267 | # save for later, in case of engine death | |
1268 | self._outstanding_dict[ident].add(msg_id) |
|
1268 | self._outstanding_dict[ident].add(msg_id) | |
1269 | self.history.append(msg_id) |
|
1269 | self.history.append(msg_id) | |
1270 | self.metadata[msg_id]['submitted'] = datetime.now() |
|
1270 | self.metadata[msg_id]['submitted'] = datetime.now() | |
1271 |
|
1271 | |||
1272 | return msg |
|
1272 | return msg | |
1273 |
|
1273 | |||
1274 | def send_execute_request(self, socket, code, silent=True, metadata=None, ident=None): |
|
1274 | def send_execute_request(self, socket, code, silent=True, metadata=None, ident=None): | |
1275 | """construct and send an execute request via a socket. |
|
1275 | """construct and send an execute request via a socket. | |
1276 |
|
1276 | |||
1277 | """ |
|
1277 | """ | |
1278 |
|
1278 | |||
1279 | if self._closed: |
|
1279 | if self._closed: | |
1280 | raise RuntimeError("Client cannot be used after its sockets have been closed") |
|
1280 | raise RuntimeError("Client cannot be used after its sockets have been closed") | |
1281 |
|
1281 | |||
1282 | # defaults: |
|
1282 | # defaults: | |
1283 | metadata = metadata if metadata is not None else {} |
|
1283 | metadata = metadata if metadata is not None else {} | |
1284 |
|
1284 | |||
1285 | # validate arguments |
|
1285 | # validate arguments | |
1286 | if not isinstance(code, string_types): |
|
1286 | if not isinstance(code, string_types): | |
1287 | raise TypeError("code must be text, not %s" % type(code)) |
|
1287 | raise TypeError("code must be text, not %s" % type(code)) | |
1288 | if not isinstance(metadata, dict): |
|
1288 | if not isinstance(metadata, dict): | |
1289 | raise TypeError("metadata must be dict, not %s" % type(metadata)) |
|
1289 | raise TypeError("metadata must be dict, not %s" % type(metadata)) | |
1290 |
|
1290 | |||
1291 | content = dict(code=code, silent=bool(silent), user_variables=[], user_expressions={}) |
|
1291 | content = dict(code=code, silent=bool(silent), user_variables=[], user_expressions={}) | |
1292 |
|
1292 | |||
1293 |
|
1293 | |||
1294 | msg = self.session.send(socket, "execute_request", content=content, ident=ident, |
|
1294 | msg = self.session.send(socket, "execute_request", content=content, ident=ident, | |
1295 | metadata=metadata) |
|
1295 | metadata=metadata) | |
1296 |
|
1296 | |||
1297 | msg_id = msg['header']['msg_id'] |
|
1297 | msg_id = msg['header']['msg_id'] | |
1298 | self.outstanding.add(msg_id) |
|
1298 | self.outstanding.add(msg_id) | |
1299 | if ident: |
|
1299 | if ident: | |
1300 | # possibly routed to a specific engine |
|
1300 | # possibly routed to a specific engine | |
1301 | if isinstance(ident, list): |
|
1301 | if isinstance(ident, list): | |
1302 | ident = ident[-1] |
|
1302 | ident = ident[-1] | |
1303 | if ident in self._engines.values(): |
|
1303 | if ident in self._engines.values(): | |
1304 | # save for later, in case of engine death |
|
1304 | # save for later, in case of engine death | |
1305 | self._outstanding_dict[ident].add(msg_id) |
|
1305 | self._outstanding_dict[ident].add(msg_id) | |
1306 | self.history.append(msg_id) |
|
1306 | self.history.append(msg_id) | |
1307 | self.metadata[msg_id]['submitted'] = datetime.now() |
|
1307 | self.metadata[msg_id]['submitted'] = datetime.now() | |
1308 |
|
1308 | |||
1309 | return msg |
|
1309 | return msg | |
1310 |
|
1310 | |||
1311 | #-------------------------------------------------------------------------- |
|
1311 | #-------------------------------------------------------------------------- | |
1312 | # construct a View object |
|
1312 | # construct a View object | |
1313 | #-------------------------------------------------------------------------- |
|
1313 | #-------------------------------------------------------------------------- | |
1314 |
|
1314 | |||
1315 | def load_balanced_view(self, targets=None): |
|
1315 | def load_balanced_view(self, targets=None): | |
1316 | """construct a DirectView object. |
|
1316 | """construct a DirectView object. | |
1317 |
|
1317 | |||
1318 | If no arguments are specified, create a LoadBalancedView |
|
1318 | If no arguments are specified, create a LoadBalancedView | |
1319 | using all engines. |
|
1319 | using all engines. | |
1320 |
|
1320 | |||
1321 | Parameters |
|
1321 | Parameters | |
1322 | ---------- |
|
1322 | ---------- | |
1323 |
|
1323 | |||
1324 | targets: list,slice,int,etc. [default: use all engines] |
|
1324 | targets: list,slice,int,etc. [default: use all engines] | |
1325 | The subset of engines across which to load-balance |
|
1325 | The subset of engines across which to load-balance | |
1326 | """ |
|
1326 | """ | |
1327 | if targets == 'all': |
|
1327 | if targets == 'all': | |
1328 | targets = None |
|
1328 | targets = None | |
1329 | if targets is not None: |
|
1329 | if targets is not None: | |
1330 | targets = self._build_targets(targets)[1] |
|
1330 | targets = self._build_targets(targets)[1] | |
1331 | return LoadBalancedView(client=self, socket=self._task_socket, targets=targets) |
|
1331 | return LoadBalancedView(client=self, socket=self._task_socket, targets=targets) | |
1332 |
|
1332 | |||
1333 | def direct_view(self, targets='all'): |
|
1333 | def direct_view(self, targets='all'): | |
1334 | """construct a DirectView object. |
|
1334 | """construct a DirectView object. | |
1335 |
|
1335 | |||
1336 | If no targets are specified, create a DirectView using all engines. |
|
1336 | If no targets are specified, create a DirectView using all engines. | |
1337 |
|
1337 | |||
1338 | rc.direct_view('all') is distinguished from rc[:] in that 'all' will |
|
1338 | rc.direct_view('all') is distinguished from rc[:] in that 'all' will | |
1339 | evaluate the target engines at each execution, whereas rc[:] will connect to |
|
1339 | evaluate the target engines at each execution, whereas rc[:] will connect to | |
1340 | all *current* engines, and that list will not change. |
|
1340 | all *current* engines, and that list will not change. | |
1341 |
|
1341 | |||
1342 | That is, 'all' will always use all engines, whereas rc[:] will not use |
|
1342 | That is, 'all' will always use all engines, whereas rc[:] will not use | |
1343 | engines added after the DirectView is constructed. |
|
1343 | engines added after the DirectView is constructed. | |
1344 |
|
1344 | |||
1345 | Parameters |
|
1345 | Parameters | |
1346 | ---------- |
|
1346 | ---------- | |
1347 |
|
1347 | |||
1348 | targets: list,slice,int,etc. [default: use all engines] |
|
1348 | targets: list,slice,int,etc. [default: use all engines] | |
1349 | The engines to use for the View |
|
1349 | The engines to use for the View | |
1350 | """ |
|
1350 | """ | |
1351 | single = isinstance(targets, int) |
|
1351 | single = isinstance(targets, int) | |
1352 | # allow 'all' to be lazily evaluated at each execution |
|
1352 | # allow 'all' to be lazily evaluated at each execution | |
1353 | if targets != 'all': |
|
1353 | if targets != 'all': | |
1354 | targets = self._build_targets(targets)[1] |
|
1354 | targets = self._build_targets(targets)[1] | |
1355 | if single: |
|
1355 | if single: | |
1356 | targets = targets[0] |
|
1356 | targets = targets[0] | |
1357 | return DirectView(client=self, socket=self._mux_socket, targets=targets) |
|
1357 | return DirectView(client=self, socket=self._mux_socket, targets=targets) | |
1358 |
|
1358 | |||
1359 | #-------------------------------------------------------------------------- |
|
1359 | #-------------------------------------------------------------------------- | |
1360 | # Query methods |
|
1360 | # Query methods | |
1361 | #-------------------------------------------------------------------------- |
|
1361 | #-------------------------------------------------------------------------- | |
1362 |
|
1362 | |||
1363 | @spin_first |
|
1363 | @spin_first | |
1364 | def get_result(self, indices_or_msg_ids=None, block=None): |
|
1364 | def get_result(self, indices_or_msg_ids=None, block=None): | |
1365 | """Retrieve a result by msg_id or history index, wrapped in an AsyncResult object. |
|
1365 | """Retrieve a result by msg_id or history index, wrapped in an AsyncResult object. | |
1366 |
|
1366 | |||
1367 | If the client already has the results, no request to the Hub will be made. |
|
1367 | If the client already has the results, no request to the Hub will be made. | |
1368 |
|
1368 | |||
1369 | This is a convenient way to construct AsyncResult objects, which are wrappers |
|
1369 | This is a convenient way to construct AsyncResult objects, which are wrappers | |
1370 | that include metadata about execution, and allow for awaiting results that |
|
1370 | that include metadata about execution, and allow for awaiting results that | |
1371 | were not submitted by this Client. |
|
1371 | were not submitted by this Client. | |
1372 |
|
1372 | |||
1373 | It can also be a convenient way to retrieve the metadata associated with |
|
1373 | It can also be a convenient way to retrieve the metadata associated with | |
1374 | blocking execution, since it always retrieves |
|
1374 | blocking execution, since it always retrieves | |
1375 |
|
1375 | |||
1376 | Examples |
|
1376 | Examples | |
1377 | -------- |
|
1377 | -------- | |
1378 | :: |
|
1378 | :: | |
1379 |
|
1379 | |||
1380 | In [10]: r = client.apply() |
|
1380 | In [10]: r = client.apply() | |
1381 |
|
1381 | |||
1382 | Parameters |
|
1382 | Parameters | |
1383 | ---------- |
|
1383 | ---------- | |
1384 |
|
1384 | |||
1385 | indices_or_msg_ids : integer history index, str msg_id, or list of either |
|
1385 | indices_or_msg_ids : integer history index, str msg_id, or list of either | |
1386 | The indices or msg_ids of indices to be retrieved |
|
1386 | The indices or msg_ids of indices to be retrieved | |
1387 |
|
1387 | |||
1388 | block : bool |
|
1388 | block : bool | |
1389 | Whether to wait for the result to be done |
|
1389 | Whether to wait for the result to be done | |
1390 |
|
1390 | |||
1391 | Returns |
|
1391 | Returns | |
1392 | ------- |
|
1392 | ------- | |
1393 |
|
1393 | |||
1394 | AsyncResult |
|
1394 | AsyncResult | |
1395 | A single AsyncResult object will always be returned. |
|
1395 | A single AsyncResult object will always be returned. | |
1396 |
|
1396 | |||
1397 | AsyncHubResult |
|
1397 | AsyncHubResult | |
1398 | A subclass of AsyncResult that retrieves results from the Hub |
|
1398 | A subclass of AsyncResult that retrieves results from the Hub | |
1399 |
|
1399 | |||
1400 | """ |
|
1400 | """ | |
1401 | block = self.block if block is None else block |
|
1401 | block = self.block if block is None else block | |
1402 | if indices_or_msg_ids is None: |
|
1402 | if indices_or_msg_ids is None: | |
1403 | indices_or_msg_ids = -1 |
|
1403 | indices_or_msg_ids = -1 | |
1404 |
|
1404 | |||
1405 | single_result = False |
|
1405 | single_result = False | |
1406 | if not isinstance(indices_or_msg_ids, (list,tuple)): |
|
1406 | if not isinstance(indices_or_msg_ids, (list,tuple)): | |
1407 | indices_or_msg_ids = [indices_or_msg_ids] |
|
1407 | indices_or_msg_ids = [indices_or_msg_ids] | |
1408 | single_result = True |
|
1408 | single_result = True | |
1409 |
|
1409 | |||
1410 | theids = [] |
|
1410 | theids = [] | |
1411 | for id in indices_or_msg_ids: |
|
1411 | for id in indices_or_msg_ids: | |
1412 | if isinstance(id, int): |
|
1412 | if isinstance(id, int): | |
1413 | id = self.history[id] |
|
1413 | id = self.history[id] | |
1414 | if not isinstance(id, string_types): |
|
1414 | if not isinstance(id, string_types): | |
1415 | raise TypeError("indices must be str or int, not %r"%id) |
|
1415 | raise TypeError("indices must be str or int, not %r"%id) | |
1416 | theids.append(id) |
|
1416 | theids.append(id) | |
1417 |
|
1417 | |||
1418 | local_ids = filter(lambda msg_id: msg_id in self.outstanding or msg_id in self.results, theids) |
|
1418 | local_ids = filter(lambda msg_id: msg_id in self.outstanding or msg_id in self.results, theids) | |
1419 | remote_ids = filter(lambda msg_id: msg_id not in local_ids, theids) |
|
1419 | remote_ids = filter(lambda msg_id: msg_id not in local_ids, theids) | |
1420 |
|
1420 | |||
1421 | # given single msg_id initially, get_result shot get the result itself, |
|
1421 | # given single msg_id initially, get_result shot get the result itself, | |
1422 | # not a length-one list |
|
1422 | # not a length-one list | |
1423 | if single_result: |
|
1423 | if single_result: | |
1424 | theids = theids[0] |
|
1424 | theids = theids[0] | |
1425 |
|
1425 | |||
1426 | if remote_ids: |
|
1426 | if remote_ids: | |
1427 | ar = AsyncHubResult(self, msg_ids=theids) |
|
1427 | ar = AsyncHubResult(self, msg_ids=theids) | |
1428 | else: |
|
1428 | else: | |
1429 | ar = AsyncResult(self, msg_ids=theids) |
|
1429 | ar = AsyncResult(self, msg_ids=theids) | |
1430 |
|
1430 | |||
1431 | if block: |
|
1431 | if block: | |
1432 | ar.wait() |
|
1432 | ar.wait() | |
1433 |
|
1433 | |||
1434 | return ar |
|
1434 | return ar | |
1435 |
|
1435 | |||
1436 | @spin_first |
|
1436 | @spin_first | |
1437 | def resubmit(self, indices_or_msg_ids=None, metadata=None, block=None): |
|
1437 | def resubmit(self, indices_or_msg_ids=None, metadata=None, block=None): | |
1438 | """Resubmit one or more tasks. |
|
1438 | """Resubmit one or more tasks. | |
1439 |
|
1439 | |||
1440 | in-flight tasks may not be resubmitted. |
|
1440 | in-flight tasks may not be resubmitted. | |
1441 |
|
1441 | |||
1442 | Parameters |
|
1442 | Parameters | |
1443 | ---------- |
|
1443 | ---------- | |
1444 |
|
1444 | |||
1445 | indices_or_msg_ids : integer history index, str msg_id, or list of either |
|
1445 | indices_or_msg_ids : integer history index, str msg_id, or list of either | |
1446 | The indices or msg_ids of indices to be retrieved |
|
1446 | The indices or msg_ids of indices to be retrieved | |
1447 |
|
1447 | |||
1448 | block : bool |
|
1448 | block : bool | |
1449 | Whether to wait for the result to be done |
|
1449 | Whether to wait for the result to be done | |
1450 |
|
1450 | |||
1451 | Returns |
|
1451 | Returns | |
1452 | ------- |
|
1452 | ------- | |
1453 |
|
1453 | |||
1454 | AsyncHubResult |
|
1454 | AsyncHubResult | |
1455 | A subclass of AsyncResult that retrieves results from the Hub |
|
1455 | A subclass of AsyncResult that retrieves results from the Hub | |
1456 |
|
1456 | |||
1457 | """ |
|
1457 | """ | |
1458 | block = self.block if block is None else block |
|
1458 | block = self.block if block is None else block | |
1459 | if indices_or_msg_ids is None: |
|
1459 | if indices_or_msg_ids is None: | |
1460 | indices_or_msg_ids = -1 |
|
1460 | indices_or_msg_ids = -1 | |
1461 |
|
1461 | |||
1462 | if not isinstance(indices_or_msg_ids, (list,tuple)): |
|
1462 | if not isinstance(indices_or_msg_ids, (list,tuple)): | |
1463 | indices_or_msg_ids = [indices_or_msg_ids] |
|
1463 | indices_or_msg_ids = [indices_or_msg_ids] | |
1464 |
|
1464 | |||
1465 | theids = [] |
|
1465 | theids = [] | |
1466 | for id in indices_or_msg_ids: |
|
1466 | for id in indices_or_msg_ids: | |
1467 | if isinstance(id, int): |
|
1467 | if isinstance(id, int): | |
1468 | id = self.history[id] |
|
1468 | id = self.history[id] | |
1469 | if not isinstance(id, string_types): |
|
1469 | if not isinstance(id, string_types): | |
1470 | raise TypeError("indices must be str or int, not %r"%id) |
|
1470 | raise TypeError("indices must be str or int, not %r"%id) | |
1471 | theids.append(id) |
|
1471 | theids.append(id) | |
1472 |
|
1472 | |||
1473 | content = dict(msg_ids = theids) |
|
1473 | content = dict(msg_ids = theids) | |
1474 |
|
1474 | |||
1475 | self.session.send(self._query_socket, 'resubmit_request', content) |
|
1475 | self.session.send(self._query_socket, 'resubmit_request', content) | |
1476 |
|
1476 | |||
1477 | zmq.select([self._query_socket], [], []) |
|
1477 | zmq.select([self._query_socket], [], []) | |
1478 | idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK) |
|
1478 | idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK) | |
1479 | if self.debug: |
|
1479 | if self.debug: | |
1480 | pprint(msg) |
|
1480 | pprint(msg) | |
1481 | content = msg['content'] |
|
1481 | content = msg['content'] | |
1482 | if content['status'] != 'ok': |
|
1482 | if content['status'] != 'ok': | |
1483 | raise self._unwrap_exception(content) |
|
1483 | raise self._unwrap_exception(content) | |
1484 | mapping = content['resubmitted'] |
|
1484 | mapping = content['resubmitted'] | |
1485 | new_ids = [ mapping[msg_id] for msg_id in theids ] |
|
1485 | new_ids = [ mapping[msg_id] for msg_id in theids ] | |
1486 |
|
1486 | |||
1487 | ar = AsyncHubResult(self, msg_ids=new_ids) |
|
1487 | ar = AsyncHubResult(self, msg_ids=new_ids) | |
1488 |
|
1488 | |||
1489 | if block: |
|
1489 | if block: | |
1490 | ar.wait() |
|
1490 | ar.wait() | |
1491 |
|
1491 | |||
1492 | return ar |
|
1492 | return ar | |
1493 |
|
1493 | |||
1494 | @spin_first |
|
1494 | @spin_first | |
1495 | def result_status(self, msg_ids, status_only=True): |
|
1495 | def result_status(self, msg_ids, status_only=True): | |
1496 | """Check on the status of the result(s) of the apply request with `msg_ids`. |
|
1496 | """Check on the status of the result(s) of the apply request with `msg_ids`. | |
1497 |
|
1497 | |||
1498 | If status_only is False, then the actual results will be retrieved, else |
|
1498 | If status_only is False, then the actual results will be retrieved, else | |
1499 | only the status of the results will be checked. |
|
1499 | only the status of the results will be checked. | |
1500 |
|
1500 | |||
1501 | Parameters |
|
1501 | Parameters | |
1502 | ---------- |
|
1502 | ---------- | |
1503 |
|
1503 | |||
1504 | msg_ids : list of msg_ids |
|
1504 | msg_ids : list of msg_ids | |
1505 | if int: |
|
1505 | if int: | |
1506 | Passed as index to self.history for convenience. |
|
1506 | Passed as index to self.history for convenience. | |
1507 | status_only : bool (default: True) |
|
1507 | status_only : bool (default: True) | |
1508 | if False: |
|
1508 | if False: | |
1509 | Retrieve the actual results of completed tasks. |
|
1509 | Retrieve the actual results of completed tasks. | |
1510 |
|
1510 | |||
1511 | Returns |
|
1511 | Returns | |
1512 | ------- |
|
1512 | ------- | |
1513 |
|
1513 | |||
1514 | results : dict |
|
1514 | results : dict | |
1515 | There will always be the keys 'pending' and 'completed', which will |
|
1515 | There will always be the keys 'pending' and 'completed', which will | |
1516 | be lists of msg_ids that are incomplete or complete. If `status_only` |
|
1516 | be lists of msg_ids that are incomplete or complete. If `status_only` | |
1517 | is False, then completed results will be keyed by their `msg_id`. |
|
1517 | is False, then completed results will be keyed by their `msg_id`. | |
1518 | """ |
|
1518 | """ | |
1519 | if not isinstance(msg_ids, (list,tuple)): |
|
1519 | if not isinstance(msg_ids, (list,tuple)): | |
1520 | msg_ids = [msg_ids] |
|
1520 | msg_ids = [msg_ids] | |
1521 |
|
1521 | |||
1522 | theids = [] |
|
1522 | theids = [] | |
1523 | for msg_id in msg_ids: |
|
1523 | for msg_id in msg_ids: | |
1524 | if isinstance(msg_id, int): |
|
1524 | if isinstance(msg_id, int): | |
1525 | msg_id = self.history[msg_id] |
|
1525 | msg_id = self.history[msg_id] | |
1526 | if not isinstance(msg_id, string_types): |
|
1526 | if not isinstance(msg_id, string_types): | |
1527 | raise TypeError("msg_ids must be str, not %r"%msg_id) |
|
1527 | raise TypeError("msg_ids must be str, not %r"%msg_id) | |
1528 | theids.append(msg_id) |
|
1528 | theids.append(msg_id) | |
1529 |
|
1529 | |||
1530 | completed = [] |
|
1530 | completed = [] | |
1531 | local_results = {} |
|
1531 | local_results = {} | |
1532 |
|
1532 | |||
1533 | # comment this block out to temporarily disable local shortcut: |
|
1533 | # comment this block out to temporarily disable local shortcut: | |
1534 | for msg_id in theids: |
|
1534 | for msg_id in theids: | |
1535 | if msg_id in self.results: |
|
1535 | if msg_id in self.results: | |
1536 | completed.append(msg_id) |
|
1536 | completed.append(msg_id) | |
1537 | local_results[msg_id] = self.results[msg_id] |
|
1537 | local_results[msg_id] = self.results[msg_id] | |
1538 | theids.remove(msg_id) |
|
1538 | theids.remove(msg_id) | |
1539 |
|
1539 | |||
1540 | if theids: # some not locally cached |
|
1540 | if theids: # some not locally cached | |
1541 | content = dict(msg_ids=theids, status_only=status_only) |
|
1541 | content = dict(msg_ids=theids, status_only=status_only) | |
1542 | msg = self.session.send(self._query_socket, "result_request", content=content) |
|
1542 | msg = self.session.send(self._query_socket, "result_request", content=content) | |
1543 | zmq.select([self._query_socket], [], []) |
|
1543 | zmq.select([self._query_socket], [], []) | |
1544 | idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK) |
|
1544 | idents,msg = self.session.recv(self._query_socket, zmq.NOBLOCK) | |
1545 | if self.debug: |
|
1545 | if self.debug: | |
1546 | pprint(msg) |
|
1546 | pprint(msg) | |
1547 | content = msg['content'] |
|
1547 | content = msg['content'] | |
1548 | if content['status'] != 'ok': |
|
1548 | if content['status'] != 'ok': | |
1549 | raise self._unwrap_exception(content) |
|
1549 | raise self._unwrap_exception(content) | |
1550 | buffers = msg['buffers'] |
|
1550 | buffers = msg['buffers'] | |
1551 | else: |
|
1551 | else: | |
1552 | content = dict(completed=[],pending=[]) |
|
1552 | content = dict(completed=[],pending=[]) | |
1553 |
|
1553 | |||
1554 | content['completed'].extend(completed) |
|
1554 | content['completed'].extend(completed) | |
1555 |
|
1555 | |||
1556 | if status_only: |
|
1556 | if status_only: | |
1557 | return content |
|
1557 | return content | |
1558 |
|
1558 | |||
1559 | failures = [] |
|
1559 | failures = [] | |
1560 | # load cached results into result: |
|
1560 | # load cached results into result: | |
1561 | content.update(local_results) |
|
1561 | content.update(local_results) | |
1562 |
|
1562 | |||
1563 | # update cache with results: |
|
1563 | # update cache with results: | |
1564 | for msg_id in sorted(theids): |
|
1564 | for msg_id in sorted(theids): | |
1565 | if msg_id in content['completed']: |
|
1565 | if msg_id in content['completed']: | |
1566 | rec = content[msg_id] |
|
1566 | rec = content[msg_id] | |
1567 | parent = rec['header'] |
|
1567 | parent = rec['header'] | |
1568 | header = rec['result_header'] |
|
1568 | header = rec['result_header'] | |
1569 | rcontent = rec['result_content'] |
|
1569 | rcontent = rec['result_content'] | |
1570 | iodict = rec['io'] |
|
1570 | iodict = rec['io'] | |
1571 | if isinstance(rcontent, str): |
|
1571 | if isinstance(rcontent, str): | |
1572 | rcontent = self.session.unpack(rcontent) |
|
1572 | rcontent = self.session.unpack(rcontent) | |
1573 |
|
1573 | |||
1574 | md = self.metadata[msg_id] |
|
1574 | md = self.metadata[msg_id] | |
1575 | md_msg = dict( |
|
1575 | md_msg = dict( | |
1576 | content=rcontent, |
|
1576 | content=rcontent, | |
1577 | parent_header=parent, |
|
1577 | parent_header=parent, | |
1578 | header=header, |
|
1578 | header=header, | |
1579 | metadata=rec['result_metadata'], |
|
1579 | metadata=rec['result_metadata'], | |
1580 | ) |
|
1580 | ) | |
1581 | md.update(self._extract_metadata(md_msg)) |
|
1581 | md.update(self._extract_metadata(md_msg)) | |
1582 | if rec.get('received'): |
|
1582 | if rec.get('received'): | |
1583 | md['received'] = rec['received'] |
|
1583 | md['received'] = rec['received'] | |
1584 | md.update(iodict) |
|
1584 | md.update(iodict) | |
1585 |
|
1585 | |||
1586 | if rcontent['status'] == 'ok': |
|
1586 | if rcontent['status'] == 'ok': | |
1587 | if header['msg_type'] == 'apply_reply': |
|
1587 | if header['msg_type'] == 'apply_reply': | |
1588 | res,buffers = serialize.unserialize_object(buffers) |
|
1588 | res,buffers = serialize.unserialize_object(buffers) | |
1589 | elif header['msg_type'] == 'execute_reply': |
|
1589 | elif header['msg_type'] == 'execute_reply': | |
1590 | res = ExecuteReply(msg_id, rcontent, md) |
|
1590 | res = ExecuteReply(msg_id, rcontent, md) | |
1591 | else: |
|
1591 | else: | |
1592 | raise KeyError("unhandled msg type: %r" % header['msg_type']) |
|
1592 | raise KeyError("unhandled msg type: %r" % header['msg_type']) | |
1593 | else: |
|
1593 | else: | |
1594 | res = self._unwrap_exception(rcontent) |
|
1594 | res = self._unwrap_exception(rcontent) | |
1595 | failures.append(res) |
|
1595 | failures.append(res) | |
1596 |
|
1596 | |||
1597 | self.results[msg_id] = res |
|
1597 | self.results[msg_id] = res | |
1598 | content[msg_id] = res |
|
1598 | content[msg_id] = res | |
1599 |
|
1599 | |||
1600 | if len(theids) == 1 and failures: |
|
1600 | if len(theids) == 1 and failures: | |
1601 | raise failures[0] |
|
1601 | raise failures[0] | |
1602 |
|
1602 | |||
1603 | error.collect_exceptions(failures, "result_status") |
|
1603 | error.collect_exceptions(failures, "result_status") | |
1604 | return content |
|
1604 | return content | |
1605 |
|
1605 | |||
1606 | @spin_first |
|
1606 | @spin_first | |
1607 | def queue_status(self, targets='all', verbose=False): |
|
1607 | def queue_status(self, targets='all', verbose=False): | |
1608 | """Fetch the status of engine queues. |
|
1608 | """Fetch the status of engine queues. | |
1609 |
|
1609 | |||
1610 | Parameters |
|
1610 | Parameters | |
1611 | ---------- |
|
1611 | ---------- | |
1612 |
|
1612 | |||
1613 | targets : int/str/list of ints/strs |
|
1613 | targets : int/str/list of ints/strs | |
1614 | the engines whose states are to be queried. |
|
1614 | the engines whose states are to be queried. | |
1615 | default : all |
|
1615 | default : all | |
1616 | verbose : bool |
|
1616 | verbose : bool | |
1617 | Whether to return lengths only, or lists of ids for each element |
|
1617 | Whether to return lengths only, or lists of ids for each element | |
1618 | """ |
|
1618 | """ | |
1619 | if targets == 'all': |
|
1619 | if targets == 'all': | |
1620 | # allow 'all' to be evaluated on the engine |
|
1620 | # allow 'all' to be evaluated on the engine | |
1621 | engine_ids = None |
|
1621 | engine_ids = None | |
1622 | else: |
|
1622 | else: | |
1623 | engine_ids = self._build_targets(targets)[1] |
|
1623 | engine_ids = self._build_targets(targets)[1] | |
1624 | content = dict(targets=engine_ids, verbose=verbose) |
|
1624 | content = dict(targets=engine_ids, verbose=verbose) | |
1625 | self.session.send(self._query_socket, "queue_request", content=content) |
|
1625 | self.session.send(self._query_socket, "queue_request", content=content) | |
1626 | idents,msg = self.session.recv(self._query_socket, 0) |
|
1626 | idents,msg = self.session.recv(self._query_socket, 0) | |
1627 | if self.debug: |
|
1627 | if self.debug: | |
1628 | pprint(msg) |
|
1628 | pprint(msg) | |
1629 | content = msg['content'] |
|
1629 | content = msg['content'] | |
1630 | status = content.pop('status') |
|
1630 | status = content.pop('status') | |
1631 | if status != 'ok': |
|
1631 | if status != 'ok': | |
1632 | raise self._unwrap_exception(content) |
|
1632 | raise self._unwrap_exception(content) | |
1633 | content = rekey(content) |
|
1633 | content = rekey(content) | |
1634 | if isinstance(targets, int): |
|
1634 | if isinstance(targets, int): | |
1635 | return content[targets] |
|
1635 | return content[targets] | |
1636 | else: |
|
1636 | else: | |
1637 | return content |
|
1637 | return content | |
1638 |
|
1638 | |||
1639 | def _build_msgids_from_target(self, targets=None): |
|
1639 | def _build_msgids_from_target(self, targets=None): | |
1640 | """Build a list of msg_ids from the list of engine targets""" |
|
1640 | """Build a list of msg_ids from the list of engine targets""" | |
1641 | if not targets: # needed as _build_targets otherwise uses all engines |
|
1641 | if not targets: # needed as _build_targets otherwise uses all engines | |
1642 | return [] |
|
1642 | return [] | |
1643 | target_ids = self._build_targets(targets)[0] |
|
1643 | target_ids = self._build_targets(targets)[0] | |
1644 | return filter(lambda md_id: self.metadata[md_id]["engine_uuid"] in target_ids, self.metadata) |
|
1644 | return filter(lambda md_id: self.metadata[md_id]["engine_uuid"] in target_ids, self.metadata) | |
1645 |
|
1645 | |||
1646 | def _build_msgids_from_jobs(self, jobs=None): |
|
1646 | def _build_msgids_from_jobs(self, jobs=None): | |
1647 | """Build a list of msg_ids from "jobs" """ |
|
1647 | """Build a list of msg_ids from "jobs" """ | |
1648 | if not jobs: |
|
1648 | if not jobs: | |
1649 | return [] |
|
1649 | return [] | |
1650 | msg_ids = [] |
|
1650 | msg_ids = [] | |
1651 | if isinstance(jobs, string_types + (AsyncResult,)): |
|
1651 | if isinstance(jobs, string_types + (AsyncResult,)): | |
1652 | jobs = [jobs] |
|
1652 | jobs = [jobs] | |
1653 | bad_ids = filter(lambda obj: not isinstance(obj, string_types + (AsyncResult)), jobs) |
|
1653 | bad_ids = filter(lambda obj: not isinstance(obj, string_types + (AsyncResult)), jobs) | |
1654 | if bad_ids: |
|
1654 | if bad_ids: | |
1655 | raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0]) |
|
1655 | raise TypeError("Invalid msg_id type %r, expected str or AsyncResult"%bad_ids[0]) | |
1656 | for j in jobs: |
|
1656 | for j in jobs: | |
1657 | if isinstance(j, AsyncResult): |
|
1657 | if isinstance(j, AsyncResult): | |
1658 | msg_ids.extend(j.msg_ids) |
|
1658 | msg_ids.extend(j.msg_ids) | |
1659 | else: |
|
1659 | else: | |
1660 | msg_ids.append(j) |
|
1660 | msg_ids.append(j) | |
1661 | return msg_ids |
|
1661 | return msg_ids | |
1662 |
|
1662 | |||
1663 | def purge_local_results(self, jobs=[], targets=[]): |
|
1663 | def purge_local_results(self, jobs=[], targets=[]): | |
1664 | """Clears the client caches of results and frees such memory. |
|
1664 | """Clears the client caches of results and frees such memory. | |
1665 |
|
1665 | |||
1666 | Individual results can be purged by msg_id, or the entire |
|
1666 | Individual results can be purged by msg_id, or the entire | |
1667 | history of specific targets can be purged. |
|
1667 | history of specific targets can be purged. | |
1668 |
|
1668 | |||
1669 | Use `purge_local_results('all')` to scrub everything from the Clients's db. |
|
1669 | Use `purge_local_results('all')` to scrub everything from the Clients's db. | |
1670 |
|
1670 | |||
1671 | The client must have no outstanding tasks before purging the caches. |
|
1671 | The client must have no outstanding tasks before purging the caches. | |
1672 | Raises `AssertionError` if there are still outstanding tasks. |
|
1672 | Raises `AssertionError` if there are still outstanding tasks. | |
1673 |
|
1673 | |||
1674 | After this call all `AsyncResults` are invalid and should be discarded. |
|
1674 | After this call all `AsyncResults` are invalid and should be discarded. | |
1675 |
|
1675 | |||
1676 | If you must "reget" the results, you can still do so by using |
|
1676 | If you must "reget" the results, you can still do so by using | |
1677 | `client.get_result(msg_id)` or `client.get_result(asyncresult)`. This will |
|
1677 | `client.get_result(msg_id)` or `client.get_result(asyncresult)`. This will | |
1678 | redownload the results from the hub if they are still available |
|
1678 | redownload the results from the hub if they are still available | |
1679 | (i.e `client.purge_hub_results(...)` has not been called. |
|
1679 | (i.e `client.purge_hub_results(...)` has not been called. | |
1680 |
|
1680 | |||
1681 | Parameters |
|
1681 | Parameters | |
1682 | ---------- |
|
1682 | ---------- | |
1683 |
|
1683 | |||
1684 | jobs : str or list of str or AsyncResult objects |
|
1684 | jobs : str or list of str or AsyncResult objects | |
1685 | the msg_ids whose results should be purged. |
|
1685 | the msg_ids whose results should be purged. | |
1686 | targets : int/str/list of ints/strs |
|
1686 | targets : int/str/list of ints/strs | |
1687 | The targets, by int_id, whose entire results are to be purged. |
|
1687 | The targets, by int_id, whose entire results are to be purged. | |
1688 |
|
1688 | |||
1689 | default : None |
|
1689 | default : None | |
1690 | """ |
|
1690 | """ | |
1691 | assert not self.outstanding, "Can't purge a client with outstanding tasks!" |
|
1691 | assert not self.outstanding, "Can't purge a client with outstanding tasks!" | |
1692 |
|
1692 | |||
1693 | if not targets and not jobs: |
|
1693 | if not targets and not jobs: | |
1694 | raise ValueError("Must specify at least one of `targets` and `jobs`") |
|
1694 | raise ValueError("Must specify at least one of `targets` and `jobs`") | |
1695 |
|
1695 | |||
1696 | if jobs == 'all': |
|
1696 | if jobs == 'all': | |
1697 | self.results.clear() |
|
1697 | self.results.clear() | |
1698 | self.metadata.clear() |
|
1698 | self.metadata.clear() | |
1699 | return |
|
1699 | return | |
1700 | else: |
|
1700 | else: | |
1701 | msg_ids = [] |
|
1701 | msg_ids = [] | |
1702 | msg_ids.extend(self._build_msgids_from_target(targets)) |
|
1702 | msg_ids.extend(self._build_msgids_from_target(targets)) | |
1703 | msg_ids.extend(self._build_msgids_from_jobs(jobs)) |
|
1703 | msg_ids.extend(self._build_msgids_from_jobs(jobs)) | |
1704 | map(self.results.pop, msg_ids) |
|
1704 | map(self.results.pop, msg_ids) | |
1705 | map(self.metadata.pop, msg_ids) |
|
1705 | map(self.metadata.pop, msg_ids) | |
1706 |
|
1706 | |||
1707 |
|
1707 | |||
1708 | @spin_first |
|
1708 | @spin_first | |
1709 | def purge_hub_results(self, jobs=[], targets=[]): |
|
1709 | def purge_hub_results(self, jobs=[], targets=[]): | |
1710 | """Tell the Hub to forget results. |
|
1710 | """Tell the Hub to forget results. | |
1711 |
|
1711 | |||
1712 | Individual results can be purged by msg_id, or the entire |
|
1712 | Individual results can be purged by msg_id, or the entire | |
1713 | history of specific targets can be purged. |
|
1713 | history of specific targets can be purged. | |
1714 |
|
1714 | |||
1715 | Use `purge_results('all')` to scrub everything from the Hub's db. |
|
1715 | Use `purge_results('all')` to scrub everything from the Hub's db. | |
1716 |
|
1716 | |||
1717 | Parameters |
|
1717 | Parameters | |
1718 | ---------- |
|
1718 | ---------- | |
1719 |
|
1719 | |||
1720 | jobs : str or list of str or AsyncResult objects |
|
1720 | jobs : str or list of str or AsyncResult objects | |
1721 | the msg_ids whose results should be forgotten. |
|
1721 | the msg_ids whose results should be forgotten. | |
1722 | targets : int/str/list of ints/strs |
|
1722 | targets : int/str/list of ints/strs | |
1723 | The targets, by int_id, whose entire history is to be purged. |
|
1723 | The targets, by int_id, whose entire history is to be purged. | |
1724 |
|
1724 | |||
1725 | default : None |
|
1725 | default : None | |
1726 | """ |
|
1726 | """ | |
1727 | if not targets and not jobs: |
|
1727 | if not targets and not jobs: | |
1728 | raise ValueError("Must specify at least one of `targets` and `jobs`") |
|
1728 | raise ValueError("Must specify at least one of `targets` and `jobs`") | |
1729 | if targets: |
|
1729 | if targets: | |
1730 | targets = self._build_targets(targets)[1] |
|
1730 | targets = self._build_targets(targets)[1] | |
1731 |
|
1731 | |||
1732 | # construct msg_ids from jobs |
|
1732 | # construct msg_ids from jobs | |
1733 | if jobs == 'all': |
|
1733 | if jobs == 'all': | |
1734 | msg_ids = jobs |
|
1734 | msg_ids = jobs | |
1735 | else: |
|
1735 | else: | |
1736 | msg_ids = self._build_msgids_from_jobs(jobs) |
|
1736 | msg_ids = self._build_msgids_from_jobs(jobs) | |
1737 |
|
1737 | |||
1738 | content = dict(engine_ids=targets, msg_ids=msg_ids) |
|
1738 | content = dict(engine_ids=targets, msg_ids=msg_ids) | |
1739 | self.session.send(self._query_socket, "purge_request", content=content) |
|
1739 | self.session.send(self._query_socket, "purge_request", content=content) | |
1740 | idents, msg = self.session.recv(self._query_socket, 0) |
|
1740 | idents, msg = self.session.recv(self._query_socket, 0) | |
1741 | if self.debug: |
|
1741 | if self.debug: | |
1742 | pprint(msg) |
|
1742 | pprint(msg) | |
1743 | content = msg['content'] |
|
1743 | content = msg['content'] | |
1744 | if content['status'] != 'ok': |
|
1744 | if content['status'] != 'ok': | |
1745 | raise self._unwrap_exception(content) |
|
1745 | raise self._unwrap_exception(content) | |
1746 |
|
1746 | |||
1747 | def purge_results(self, jobs=[], targets=[]): |
|
1747 | def purge_results(self, jobs=[], targets=[]): | |
1748 | """Clears the cached results from both the hub and the local client |
|
1748 | """Clears the cached results from both the hub and the local client | |
1749 |
|
1749 | |||
1750 | Individual results can be purged by msg_id, or the entire |
|
1750 | Individual results can be purged by msg_id, or the entire | |
1751 | history of specific targets can be purged. |
|
1751 | history of specific targets can be purged. | |
1752 |
|
1752 | |||
1753 | Use `purge_results('all')` to scrub every cached result from both the Hub's and |
|
1753 | Use `purge_results('all')` to scrub every cached result from both the Hub's and | |
1754 | the Client's db. |
|
1754 | the Client's db. | |
1755 |
|
1755 | |||
1756 | Equivalent to calling both `purge_hub_results()` and `purge_client_results()` with |
|
1756 | Equivalent to calling both `purge_hub_results()` and `purge_client_results()` with | |
1757 | the same arguments. |
|
1757 | the same arguments. | |
1758 |
|
1758 | |||
1759 | Parameters |
|
1759 | Parameters | |
1760 | ---------- |
|
1760 | ---------- | |
1761 |
|
1761 | |||
1762 | jobs : str or list of str or AsyncResult objects |
|
1762 | jobs : str or list of str or AsyncResult objects | |
1763 | the msg_ids whose results should be forgotten. |
|
1763 | the msg_ids whose results should be forgotten. | |
1764 | targets : int/str/list of ints/strs |
|
1764 | targets : int/str/list of ints/strs | |
1765 | The targets, by int_id, whose entire history is to be purged. |
|
1765 | The targets, by int_id, whose entire history is to be purged. | |
1766 |
|
1766 | |||
1767 | default : None |
|
1767 | default : None | |
1768 | """ |
|
1768 | """ | |
1769 | self.purge_local_results(jobs=jobs, targets=targets) |
|
1769 | self.purge_local_results(jobs=jobs, targets=targets) | |
1770 | self.purge_hub_results(jobs=jobs, targets=targets) |
|
1770 | self.purge_hub_results(jobs=jobs, targets=targets) | |
1771 |
|
1771 | |||
1772 | def purge_everything(self): |
|
1772 | def purge_everything(self): | |
1773 | """Clears all content from previous Tasks from both the hub and the local client |
|
1773 | """Clears all content from previous Tasks from both the hub and the local client | |
1774 |
|
1774 | |||
1775 | In addition to calling `purge_results("all")` it also deletes the history and |
|
1775 | In addition to calling `purge_results("all")` it also deletes the history and | |
1776 | other bookkeeping lists. |
|
1776 | other bookkeeping lists. | |
1777 | """ |
|
1777 | """ | |
1778 | self.purge_results("all") |
|
1778 | self.purge_results("all") | |
1779 | self.history = [] |
|
1779 | self.history = [] | |
1780 | self.session.digest_history.clear() |
|
1780 | self.session.digest_history.clear() | |
1781 |
|
1781 | |||
1782 | @spin_first |
|
1782 | @spin_first | |
1783 | def hub_history(self): |
|
1783 | def hub_history(self): | |
1784 | """Get the Hub's history |
|
1784 | """Get the Hub's history | |
1785 |
|
1785 | |||
1786 | Just like the Client, the Hub has a history, which is a list of msg_ids. |
|
1786 | Just like the Client, the Hub has a history, which is a list of msg_ids. | |
1787 | This will contain the history of all clients, and, depending on configuration, |
|
1787 | This will contain the history of all clients, and, depending on configuration, | |
1788 | may contain history across multiple cluster sessions. |
|
1788 | may contain history across multiple cluster sessions. | |
1789 |
|
1789 | |||
1790 | Any msg_id returned here is a valid argument to `get_result`. |
|
1790 | Any msg_id returned here is a valid argument to `get_result`. | |
1791 |
|
1791 | |||
1792 | Returns |
|
1792 | Returns | |
1793 | ------- |
|
1793 | ------- | |
1794 |
|
1794 | |||
1795 | msg_ids : list of strs |
|
1795 | msg_ids : list of strs | |
1796 | list of all msg_ids, ordered by task submission time. |
|
1796 | list of all msg_ids, ordered by task submission time. | |
1797 | """ |
|
1797 | """ | |
1798 |
|
1798 | |||
1799 | self.session.send(self._query_socket, "history_request", content={}) |
|
1799 | self.session.send(self._query_socket, "history_request", content={}) | |
1800 | idents, msg = self.session.recv(self._query_socket, 0) |
|
1800 | idents, msg = self.session.recv(self._query_socket, 0) | |
1801 |
|
1801 | |||
1802 | if self.debug: |
|
1802 | if self.debug: | |
1803 | pprint(msg) |
|
1803 | pprint(msg) | |
1804 | content = msg['content'] |
|
1804 | content = msg['content'] | |
1805 | if content['status'] != 'ok': |
|
1805 | if content['status'] != 'ok': | |
1806 | raise self._unwrap_exception(content) |
|
1806 | raise self._unwrap_exception(content) | |
1807 | else: |
|
1807 | else: | |
1808 | return content['history'] |
|
1808 | return content['history'] | |
1809 |
|
1809 | |||
1810 | @spin_first |
|
1810 | @spin_first | |
1811 | def db_query(self, query, keys=None): |
|
1811 | def db_query(self, query, keys=None): | |
1812 | """Query the Hub's TaskRecord database |
|
1812 | """Query the Hub's TaskRecord database | |
1813 |
|
1813 | |||
1814 | This will return a list of task record dicts that match `query` |
|
1814 | This will return a list of task record dicts that match `query` | |
1815 |
|
1815 | |||
1816 | Parameters |
|
1816 | Parameters | |
1817 | ---------- |
|
1817 | ---------- | |
1818 |
|
1818 | |||
1819 | query : mongodb query dict |
|
1819 | query : mongodb query dict | |
1820 | The search dict. See mongodb query docs for details. |
|
1820 | The search dict. See mongodb query docs for details. | |
1821 | keys : list of strs [optional] |
|
1821 | keys : list of strs [optional] | |
1822 | The subset of keys to be returned. The default is to fetch everything but buffers. |
|
1822 | The subset of keys to be returned. The default is to fetch everything but buffers. | |
1823 | 'msg_id' will *always* be included. |
|
1823 | 'msg_id' will *always* be included. | |
1824 | """ |
|
1824 | """ | |
1825 | if isinstance(keys, string_types): |
|
1825 | if isinstance(keys, string_types): | |
1826 | keys = [keys] |
|
1826 | keys = [keys] | |
1827 | content = dict(query=query, keys=keys) |
|
1827 | content = dict(query=query, keys=keys) | |
1828 | self.session.send(self._query_socket, "db_request", content=content) |
|
1828 | self.session.send(self._query_socket, "db_request", content=content) | |
1829 | idents, msg = self.session.recv(self._query_socket, 0) |
|
1829 | idents, msg = self.session.recv(self._query_socket, 0) | |
1830 | if self.debug: |
|
1830 | if self.debug: | |
1831 | pprint(msg) |
|
1831 | pprint(msg) | |
1832 | content = msg['content'] |
|
1832 | content = msg['content'] | |
1833 | if content['status'] != 'ok': |
|
1833 | if content['status'] != 'ok': | |
1834 | raise self._unwrap_exception(content) |
|
1834 | raise self._unwrap_exception(content) | |
1835 |
|
1835 | |||
1836 | records = content['records'] |
|
1836 | records = content['records'] | |
1837 |
|
1837 | |||
1838 | buffer_lens = content['buffer_lens'] |
|
1838 | buffer_lens = content['buffer_lens'] | |
1839 | result_buffer_lens = content['result_buffer_lens'] |
|
1839 | result_buffer_lens = content['result_buffer_lens'] | |
1840 | buffers = msg['buffers'] |
|
1840 | buffers = msg['buffers'] | |
1841 | has_bufs = buffer_lens is not None |
|
1841 | has_bufs = buffer_lens is not None | |
1842 | has_rbufs = result_buffer_lens is not None |
|
1842 | has_rbufs = result_buffer_lens is not None | |
1843 | for i,rec in enumerate(records): |
|
1843 | for i,rec in enumerate(records): | |
1844 | # relink buffers |
|
1844 | # relink buffers | |
1845 | if has_bufs: |
|
1845 | if has_bufs: | |
1846 | blen = buffer_lens[i] |
|
1846 | blen = buffer_lens[i] | |
1847 | rec['buffers'], buffers = buffers[:blen],buffers[blen:] |
|
1847 | rec['buffers'], buffers = buffers[:blen],buffers[blen:] | |
1848 | if has_rbufs: |
|
1848 | if has_rbufs: | |
1849 | blen = result_buffer_lens[i] |
|
1849 | blen = result_buffer_lens[i] | |
1850 | rec['result_buffers'], buffers = buffers[:blen],buffers[blen:] |
|
1850 | rec['result_buffers'], buffers = buffers[:blen],buffers[blen:] | |
1851 |
|
1851 | |||
1852 | return records |
|
1852 | return records | |
1853 |
|
1853 | |||
1854 | __all__ = [ 'Client' ] |
|
1854 | __all__ = [ 'Client' ] |
@@ -1,369 +1,369 b'' | |||||
1 | """some generic utilities for dealing with classes, urls, and serialization |
|
1 | """some generic utilities for dealing with classes, urls, and serialization | |
2 |
|
2 | |||
3 | Authors: |
|
3 | Authors: | |
4 |
|
4 | |||
5 | * Min RK |
|
5 | * Min RK | |
6 | """ |
|
6 | """ | |
7 | #----------------------------------------------------------------------------- |
|
7 | #----------------------------------------------------------------------------- | |
8 | # Copyright (C) 2010-2011 The IPython Development Team |
|
8 | # Copyright (C) 2010-2011 The IPython Development Team | |
9 | # |
|
9 | # | |
10 | # Distributed under the terms of the BSD License. The full license is in |
|
10 | # Distributed under the terms of the BSD License. The full license is in | |
11 | # the file COPYING, distributed as part of this software. |
|
11 | # the file COPYING, distributed as part of this software. | |
12 | #----------------------------------------------------------------------------- |
|
12 | #----------------------------------------------------------------------------- | |
13 |
|
13 | |||
14 | #----------------------------------------------------------------------------- |
|
14 | #----------------------------------------------------------------------------- | |
15 | # Imports |
|
15 | # Imports | |
16 | #----------------------------------------------------------------------------- |
|
16 | #----------------------------------------------------------------------------- | |
17 |
|
17 | |||
18 | # Standard library imports. |
|
18 | # Standard library imports. | |
19 | import logging |
|
19 | import logging | |
20 | import os |
|
20 | import os | |
21 | import re |
|
21 | import re | |
22 | import stat |
|
22 | import stat | |
23 | import socket |
|
23 | import socket | |
24 | import sys |
|
24 | import sys | |
25 | from signal import signal, SIGINT, SIGABRT, SIGTERM |
|
25 | from signal import signal, SIGINT, SIGABRT, SIGTERM | |
26 | try: |
|
26 | try: | |
27 | from signal import SIGKILL |
|
27 | from signal import SIGKILL | |
28 | except ImportError: |
|
28 | except ImportError: | |
29 | SIGKILL=None |
|
29 | SIGKILL=None | |
30 |
|
30 | |||
31 | try: |
|
31 | try: | |
32 | import cPickle |
|
32 | import cPickle | |
33 | pickle = cPickle |
|
33 | pickle = cPickle | |
34 | except: |
|
34 | except: | |
35 | cPickle = None |
|
35 | cPickle = None | |
36 | import pickle |
|
36 | import pickle | |
37 |
|
37 | |||
38 | # System library imports |
|
38 | # System library imports | |
39 | import zmq |
|
39 | import zmq | |
40 | from zmq.log import handlers |
|
40 | from zmq.log import handlers | |
41 |
|
41 | |||
42 | from IPython.external.decorator import decorator |
|
42 | from IPython.external.decorator import decorator | |
43 |
|
43 | |||
44 | # IPython imports |
|
44 | # IPython imports | |
45 | from IPython.config.application import Application |
|
45 | from IPython.config.application import Application | |
46 | from IPython.utils.localinterfaces import localhost, is_public_ip, public_ips |
|
46 | from IPython.utils.localinterfaces import localhost, is_public_ip, public_ips | |
47 | from IPython.utils.py3compat import string_types |
|
47 | from IPython.utils.py3compat import string_types | |
48 | from IPython.kernel.zmq.log import EnginePUBHandler |
|
48 | from IPython.kernel.zmq.log import EnginePUBHandler | |
49 | from IPython.kernel.zmq.serialize import ( |
|
49 | from IPython.kernel.zmq.serialize import ( | |
50 | unserialize_object, serialize_object, pack_apply_message, unpack_apply_message |
|
50 | unserialize_object, serialize_object, pack_apply_message, unpack_apply_message | |
51 | ) |
|
51 | ) | |
52 |
|
52 | |||
53 | #----------------------------------------------------------------------------- |
|
53 | #----------------------------------------------------------------------------- | |
54 | # Classes |
|
54 | # Classes | |
55 | #----------------------------------------------------------------------------- |
|
55 | #----------------------------------------------------------------------------- | |
56 |
|
56 | |||
57 | class Namespace(dict): |
|
57 | class Namespace(dict): | |
58 | """Subclass of dict for attribute access to keys.""" |
|
58 | """Subclass of dict for attribute access to keys.""" | |
59 |
|
59 | |||
60 | def __getattr__(self, key): |
|
60 | def __getattr__(self, key): | |
61 | """getattr aliased to getitem""" |
|
61 | """getattr aliased to getitem""" | |
62 |
if key in self |
|
62 | if key in self: | |
63 | return self[key] |
|
63 | return self[key] | |
64 | else: |
|
64 | else: | |
65 | raise NameError(key) |
|
65 | raise NameError(key) | |
66 |
|
66 | |||
67 | def __setattr__(self, key, value): |
|
67 | def __setattr__(self, key, value): | |
68 | """setattr aliased to setitem, with strict""" |
|
68 | """setattr aliased to setitem, with strict""" | |
69 | if hasattr(dict, key): |
|
69 | if hasattr(dict, key): | |
70 | raise KeyError("Cannot override dict keys %r"%key) |
|
70 | raise KeyError("Cannot override dict keys %r"%key) | |
71 | self[key] = value |
|
71 | self[key] = value | |
72 |
|
72 | |||
73 |
|
73 | |||
74 | class ReverseDict(dict): |
|
74 | class ReverseDict(dict): | |
75 | """simple double-keyed subset of dict methods.""" |
|
75 | """simple double-keyed subset of dict methods.""" | |
76 |
|
76 | |||
77 | def __init__(self, *args, **kwargs): |
|
77 | def __init__(self, *args, **kwargs): | |
78 | dict.__init__(self, *args, **kwargs) |
|
78 | dict.__init__(self, *args, **kwargs) | |
79 | self._reverse = dict() |
|
79 | self._reverse = dict() | |
80 | for key, value in self.iteritems(): |
|
80 | for key, value in self.iteritems(): | |
81 | self._reverse[value] = key |
|
81 | self._reverse[value] = key | |
82 |
|
82 | |||
83 | def __getitem__(self, key): |
|
83 | def __getitem__(self, key): | |
84 | try: |
|
84 | try: | |
85 | return dict.__getitem__(self, key) |
|
85 | return dict.__getitem__(self, key) | |
86 | except KeyError: |
|
86 | except KeyError: | |
87 | return self._reverse[key] |
|
87 | return self._reverse[key] | |
88 |
|
88 | |||
89 | def __setitem__(self, key, value): |
|
89 | def __setitem__(self, key, value): | |
90 | if key in self._reverse: |
|
90 | if key in self._reverse: | |
91 | raise KeyError("Can't have key %r on both sides!"%key) |
|
91 | raise KeyError("Can't have key %r on both sides!"%key) | |
92 | dict.__setitem__(self, key, value) |
|
92 | dict.__setitem__(self, key, value) | |
93 | self._reverse[value] = key |
|
93 | self._reverse[value] = key | |
94 |
|
94 | |||
95 | def pop(self, key): |
|
95 | def pop(self, key): | |
96 | value = dict.pop(self, key) |
|
96 | value = dict.pop(self, key) | |
97 | self._reverse.pop(value) |
|
97 | self._reverse.pop(value) | |
98 | return value |
|
98 | return value | |
99 |
|
99 | |||
100 | def get(self, key, default=None): |
|
100 | def get(self, key, default=None): | |
101 | try: |
|
101 | try: | |
102 | return self[key] |
|
102 | return self[key] | |
103 | except KeyError: |
|
103 | except KeyError: | |
104 | return default |
|
104 | return default | |
105 |
|
105 | |||
106 | #----------------------------------------------------------------------------- |
|
106 | #----------------------------------------------------------------------------- | |
107 | # Functions |
|
107 | # Functions | |
108 | #----------------------------------------------------------------------------- |
|
108 | #----------------------------------------------------------------------------- | |
109 |
|
109 | |||
110 | @decorator |
|
110 | @decorator | |
111 | def log_errors(f, self, *args, **kwargs): |
|
111 | def log_errors(f, self, *args, **kwargs): | |
112 | """decorator to log unhandled exceptions raised in a method. |
|
112 | """decorator to log unhandled exceptions raised in a method. | |
113 |
|
113 | |||
114 | For use wrapping on_recv callbacks, so that exceptions |
|
114 | For use wrapping on_recv callbacks, so that exceptions | |
115 | do not cause the stream to be closed. |
|
115 | do not cause the stream to be closed. | |
116 | """ |
|
116 | """ | |
117 | try: |
|
117 | try: | |
118 | return f(self, *args, **kwargs) |
|
118 | return f(self, *args, **kwargs) | |
119 | except Exception: |
|
119 | except Exception: | |
120 | self.log.error("Uncaught exception in %r" % f, exc_info=True) |
|
120 | self.log.error("Uncaught exception in %r" % f, exc_info=True) | |
121 |
|
121 | |||
122 |
|
122 | |||
123 | def is_url(url): |
|
123 | def is_url(url): | |
124 | """boolean check for whether a string is a zmq url""" |
|
124 | """boolean check for whether a string is a zmq url""" | |
125 | if '://' not in url: |
|
125 | if '://' not in url: | |
126 | return False |
|
126 | return False | |
127 | proto, addr = url.split('://', 1) |
|
127 | proto, addr = url.split('://', 1) | |
128 | if proto.lower() not in ['tcp','pgm','epgm','ipc','inproc']: |
|
128 | if proto.lower() not in ['tcp','pgm','epgm','ipc','inproc']: | |
129 | return False |
|
129 | return False | |
130 | return True |
|
130 | return True | |
131 |
|
131 | |||
132 | def validate_url(url): |
|
132 | def validate_url(url): | |
133 | """validate a url for zeromq""" |
|
133 | """validate a url for zeromq""" | |
134 | if not isinstance(url, string_types): |
|
134 | if not isinstance(url, string_types): | |
135 | raise TypeError("url must be a string, not %r"%type(url)) |
|
135 | raise TypeError("url must be a string, not %r"%type(url)) | |
136 | url = url.lower() |
|
136 | url = url.lower() | |
137 |
|
137 | |||
138 | proto_addr = url.split('://') |
|
138 | proto_addr = url.split('://') | |
139 | assert len(proto_addr) == 2, 'Invalid url: %r'%url |
|
139 | assert len(proto_addr) == 2, 'Invalid url: %r'%url | |
140 | proto, addr = proto_addr |
|
140 | proto, addr = proto_addr | |
141 | assert proto in ['tcp','pgm','epgm','ipc','inproc'], "Invalid protocol: %r"%proto |
|
141 | assert proto in ['tcp','pgm','epgm','ipc','inproc'], "Invalid protocol: %r"%proto | |
142 |
|
142 | |||
143 | # domain pattern adapted from http://www.regexlib.com/REDetails.aspx?regexp_id=391 |
|
143 | # domain pattern adapted from http://www.regexlib.com/REDetails.aspx?regexp_id=391 | |
144 | # author: Remi Sabourin |
|
144 | # author: Remi Sabourin | |
145 | pat = re.compile(r'^([\w\d]([\w\d\-]{0,61}[\w\d])?\.)*[\w\d]([\w\d\-]{0,61}[\w\d])?$') |
|
145 | pat = re.compile(r'^([\w\d]([\w\d\-]{0,61}[\w\d])?\.)*[\w\d]([\w\d\-]{0,61}[\w\d])?$') | |
146 |
|
146 | |||
147 | if proto == 'tcp': |
|
147 | if proto == 'tcp': | |
148 | lis = addr.split(':') |
|
148 | lis = addr.split(':') | |
149 | assert len(lis) == 2, 'Invalid url: %r'%url |
|
149 | assert len(lis) == 2, 'Invalid url: %r'%url | |
150 | addr,s_port = lis |
|
150 | addr,s_port = lis | |
151 | try: |
|
151 | try: | |
152 | port = int(s_port) |
|
152 | port = int(s_port) | |
153 | except ValueError: |
|
153 | except ValueError: | |
154 | raise AssertionError("Invalid port %r in url: %r"%(port, url)) |
|
154 | raise AssertionError("Invalid port %r in url: %r"%(port, url)) | |
155 |
|
155 | |||
156 | assert addr == '*' or pat.match(addr) is not None, 'Invalid url: %r'%url |
|
156 | assert addr == '*' or pat.match(addr) is not None, 'Invalid url: %r'%url | |
157 |
|
157 | |||
158 | else: |
|
158 | else: | |
159 | # only validate tcp urls currently |
|
159 | # only validate tcp urls currently | |
160 | pass |
|
160 | pass | |
161 |
|
161 | |||
162 | return True |
|
162 | return True | |
163 |
|
163 | |||
164 |
|
164 | |||
165 | def validate_url_container(container): |
|
165 | def validate_url_container(container): | |
166 | """validate a potentially nested collection of urls.""" |
|
166 | """validate a potentially nested collection of urls.""" | |
167 | if isinstance(container, string_types): |
|
167 | if isinstance(container, string_types): | |
168 | url = container |
|
168 | url = container | |
169 | return validate_url(url) |
|
169 | return validate_url(url) | |
170 | elif isinstance(container, dict): |
|
170 | elif isinstance(container, dict): | |
171 | container = container.itervalues() |
|
171 | container = container.itervalues() | |
172 |
|
172 | |||
173 | for element in container: |
|
173 | for element in container: | |
174 | validate_url_container(element) |
|
174 | validate_url_container(element) | |
175 |
|
175 | |||
176 |
|
176 | |||
177 | def split_url(url): |
|
177 | def split_url(url): | |
178 | """split a zmq url (tcp://ip:port) into ('tcp','ip','port').""" |
|
178 | """split a zmq url (tcp://ip:port) into ('tcp','ip','port').""" | |
179 | proto_addr = url.split('://') |
|
179 | proto_addr = url.split('://') | |
180 | assert len(proto_addr) == 2, 'Invalid url: %r'%url |
|
180 | assert len(proto_addr) == 2, 'Invalid url: %r'%url | |
181 | proto, addr = proto_addr |
|
181 | proto, addr = proto_addr | |
182 | lis = addr.split(':') |
|
182 | lis = addr.split(':') | |
183 | assert len(lis) == 2, 'Invalid url: %r'%url |
|
183 | assert len(lis) == 2, 'Invalid url: %r'%url | |
184 | addr,s_port = lis |
|
184 | addr,s_port = lis | |
185 | return proto,addr,s_port |
|
185 | return proto,addr,s_port | |
186 |
|
186 | |||
187 | def disambiguate_ip_address(ip, location=None): |
|
187 | def disambiguate_ip_address(ip, location=None): | |
188 | """turn multi-ip interfaces '0.0.0.0' and '*' into connectable |
|
188 | """turn multi-ip interfaces '0.0.0.0' and '*' into connectable | |
189 | ones, based on the location (default interpretation of location is localhost).""" |
|
189 | ones, based on the location (default interpretation of location is localhost).""" | |
190 | if ip in ('0.0.0.0', '*'): |
|
190 | if ip in ('0.0.0.0', '*'): | |
191 | if location is None or is_public_ip(location) or not public_ips(): |
|
191 | if location is None or is_public_ip(location) or not public_ips(): | |
192 | # If location is unspecified or cannot be determined, assume local |
|
192 | # If location is unspecified or cannot be determined, assume local | |
193 | ip = localhost() |
|
193 | ip = localhost() | |
194 | elif location: |
|
194 | elif location: | |
195 | return location |
|
195 | return location | |
196 | return ip |
|
196 | return ip | |
197 |
|
197 | |||
198 | def disambiguate_url(url, location=None): |
|
198 | def disambiguate_url(url, location=None): | |
199 | """turn multi-ip interfaces '0.0.0.0' and '*' into connectable |
|
199 | """turn multi-ip interfaces '0.0.0.0' and '*' into connectable | |
200 | ones, based on the location (default interpretation is localhost). |
|
200 | ones, based on the location (default interpretation is localhost). | |
201 |
|
201 | |||
202 | This is for zeromq urls, such as tcp://*:10101.""" |
|
202 | This is for zeromq urls, such as tcp://*:10101.""" | |
203 | try: |
|
203 | try: | |
204 | proto,ip,port = split_url(url) |
|
204 | proto,ip,port = split_url(url) | |
205 | except AssertionError: |
|
205 | except AssertionError: | |
206 | # probably not tcp url; could be ipc, etc. |
|
206 | # probably not tcp url; could be ipc, etc. | |
207 | return url |
|
207 | return url | |
208 |
|
208 | |||
209 | ip = disambiguate_ip_address(ip,location) |
|
209 | ip = disambiguate_ip_address(ip,location) | |
210 |
|
210 | |||
211 | return "%s://%s:%s"%(proto,ip,port) |
|
211 | return "%s://%s:%s"%(proto,ip,port) | |
212 |
|
212 | |||
213 |
|
213 | |||
214 | #-------------------------------------------------------------------------- |
|
214 | #-------------------------------------------------------------------------- | |
215 | # helpers for implementing old MEC API via view.apply |
|
215 | # helpers for implementing old MEC API via view.apply | |
216 | #-------------------------------------------------------------------------- |
|
216 | #-------------------------------------------------------------------------- | |
217 |
|
217 | |||
218 | def interactive(f): |
|
218 | def interactive(f): | |
219 | """decorator for making functions appear as interactively defined. |
|
219 | """decorator for making functions appear as interactively defined. | |
220 | This results in the function being linked to the user_ns as globals() |
|
220 | This results in the function being linked to the user_ns as globals() | |
221 | instead of the module globals(). |
|
221 | instead of the module globals(). | |
222 | """ |
|
222 | """ | |
223 | f.__module__ = '__main__' |
|
223 | f.__module__ = '__main__' | |
224 | return f |
|
224 | return f | |
225 |
|
225 | |||
226 | @interactive |
|
226 | @interactive | |
227 | def _push(**ns): |
|
227 | def _push(**ns): | |
228 | """helper method for implementing `client.push` via `client.apply`""" |
|
228 | """helper method for implementing `client.push` via `client.apply`""" | |
229 | user_ns = globals() |
|
229 | user_ns = globals() | |
230 | tmp = '_IP_PUSH_TMP_' |
|
230 | tmp = '_IP_PUSH_TMP_' | |
231 | while tmp in user_ns: |
|
231 | while tmp in user_ns: | |
232 | tmp = tmp + '_' |
|
232 | tmp = tmp + '_' | |
233 | try: |
|
233 | try: | |
234 | for name, value in ns.iteritems(): |
|
234 | for name, value in ns.iteritems(): | |
235 | user_ns[tmp] = value |
|
235 | user_ns[tmp] = value | |
236 | exec("%s = %s" % (name, tmp), user_ns) |
|
236 | exec("%s = %s" % (name, tmp), user_ns) | |
237 | finally: |
|
237 | finally: | |
238 | user_ns.pop(tmp, None) |
|
238 | user_ns.pop(tmp, None) | |
239 |
|
239 | |||
240 | @interactive |
|
240 | @interactive | |
241 | def _pull(keys): |
|
241 | def _pull(keys): | |
242 | """helper method for implementing `client.pull` via `client.apply`""" |
|
242 | """helper method for implementing `client.pull` via `client.apply`""" | |
243 | if isinstance(keys, (list,tuple, set)): |
|
243 | if isinstance(keys, (list,tuple, set)): | |
244 | return map(lambda key: eval(key, globals()), keys) |
|
244 | return map(lambda key: eval(key, globals()), keys) | |
245 | else: |
|
245 | else: | |
246 | return eval(keys, globals()) |
|
246 | return eval(keys, globals()) | |
247 |
|
247 | |||
248 | @interactive |
|
248 | @interactive | |
249 | def _execute(code): |
|
249 | def _execute(code): | |
250 | """helper method for implementing `client.execute` via `client.apply`""" |
|
250 | """helper method for implementing `client.execute` via `client.apply`""" | |
251 | exec(code, globals()) |
|
251 | exec(code, globals()) | |
252 |
|
252 | |||
253 | #-------------------------------------------------------------------------- |
|
253 | #-------------------------------------------------------------------------- | |
254 | # extra process management utilities |
|
254 | # extra process management utilities | |
255 | #-------------------------------------------------------------------------- |
|
255 | #-------------------------------------------------------------------------- | |
256 |
|
256 | |||
257 | _random_ports = set() |
|
257 | _random_ports = set() | |
258 |
|
258 | |||
259 | def select_random_ports(n): |
|
259 | def select_random_ports(n): | |
260 | """Selects and return n random ports that are available.""" |
|
260 | """Selects and return n random ports that are available.""" | |
261 | ports = [] |
|
261 | ports = [] | |
262 | for i in range(n): |
|
262 | for i in range(n): | |
263 | sock = socket.socket() |
|
263 | sock = socket.socket() | |
264 | sock.bind(('', 0)) |
|
264 | sock.bind(('', 0)) | |
265 | while sock.getsockname()[1] in _random_ports: |
|
265 | while sock.getsockname()[1] in _random_ports: | |
266 | sock.close() |
|
266 | sock.close() | |
267 | sock = socket.socket() |
|
267 | sock = socket.socket() | |
268 | sock.bind(('', 0)) |
|
268 | sock.bind(('', 0)) | |
269 | ports.append(sock) |
|
269 | ports.append(sock) | |
270 | for i, sock in enumerate(ports): |
|
270 | for i, sock in enumerate(ports): | |
271 | port = sock.getsockname()[1] |
|
271 | port = sock.getsockname()[1] | |
272 | sock.close() |
|
272 | sock.close() | |
273 | ports[i] = port |
|
273 | ports[i] = port | |
274 | _random_ports.add(port) |
|
274 | _random_ports.add(port) | |
275 | return ports |
|
275 | return ports | |
276 |
|
276 | |||
277 | def signal_children(children): |
|
277 | def signal_children(children): | |
278 | """Relay interupt/term signals to children, for more solid process cleanup.""" |
|
278 | """Relay interupt/term signals to children, for more solid process cleanup.""" | |
279 | def terminate_children(sig, frame): |
|
279 | def terminate_children(sig, frame): | |
280 | log = Application.instance().log |
|
280 | log = Application.instance().log | |
281 | log.critical("Got signal %i, terminating children..."%sig) |
|
281 | log.critical("Got signal %i, terminating children..."%sig) | |
282 | for child in children: |
|
282 | for child in children: | |
283 | child.terminate() |
|
283 | child.terminate() | |
284 |
|
284 | |||
285 | sys.exit(sig != SIGINT) |
|
285 | sys.exit(sig != SIGINT) | |
286 | # sys.exit(sig) |
|
286 | # sys.exit(sig) | |
287 | for sig in (SIGINT, SIGABRT, SIGTERM): |
|
287 | for sig in (SIGINT, SIGABRT, SIGTERM): | |
288 | signal(sig, terminate_children) |
|
288 | signal(sig, terminate_children) | |
289 |
|
289 | |||
290 | def generate_exec_key(keyfile): |
|
290 | def generate_exec_key(keyfile): | |
291 | import uuid |
|
291 | import uuid | |
292 | newkey = str(uuid.uuid4()) |
|
292 | newkey = str(uuid.uuid4()) | |
293 | with open(keyfile, 'w') as f: |
|
293 | with open(keyfile, 'w') as f: | |
294 | # f.write('ipython-key ') |
|
294 | # f.write('ipython-key ') | |
295 | f.write(newkey+'\n') |
|
295 | f.write(newkey+'\n') | |
296 | # set user-only RW permissions (0600) |
|
296 | # set user-only RW permissions (0600) | |
297 | # this will have no effect on Windows |
|
297 | # this will have no effect on Windows | |
298 | os.chmod(keyfile, stat.S_IRUSR|stat.S_IWUSR) |
|
298 | os.chmod(keyfile, stat.S_IRUSR|stat.S_IWUSR) | |
299 |
|
299 | |||
300 |
|
300 | |||
301 | def integer_loglevel(loglevel): |
|
301 | def integer_loglevel(loglevel): | |
302 | try: |
|
302 | try: | |
303 | loglevel = int(loglevel) |
|
303 | loglevel = int(loglevel) | |
304 | except ValueError: |
|
304 | except ValueError: | |
305 | if isinstance(loglevel, str): |
|
305 | if isinstance(loglevel, str): | |
306 | loglevel = getattr(logging, loglevel) |
|
306 | loglevel = getattr(logging, loglevel) | |
307 | return loglevel |
|
307 | return loglevel | |
308 |
|
308 | |||
309 | def connect_logger(logname, context, iface, root="ip", loglevel=logging.DEBUG): |
|
309 | def connect_logger(logname, context, iface, root="ip", loglevel=logging.DEBUG): | |
310 | logger = logging.getLogger(logname) |
|
310 | logger = logging.getLogger(logname) | |
311 | if any([isinstance(h, handlers.PUBHandler) for h in logger.handlers]): |
|
311 | if any([isinstance(h, handlers.PUBHandler) for h in logger.handlers]): | |
312 | # don't add a second PUBHandler |
|
312 | # don't add a second PUBHandler | |
313 | return |
|
313 | return | |
314 | loglevel = integer_loglevel(loglevel) |
|
314 | loglevel = integer_loglevel(loglevel) | |
315 | lsock = context.socket(zmq.PUB) |
|
315 | lsock = context.socket(zmq.PUB) | |
316 | lsock.connect(iface) |
|
316 | lsock.connect(iface) | |
317 | handler = handlers.PUBHandler(lsock) |
|
317 | handler = handlers.PUBHandler(lsock) | |
318 | handler.setLevel(loglevel) |
|
318 | handler.setLevel(loglevel) | |
319 | handler.root_topic = root |
|
319 | handler.root_topic = root | |
320 | logger.addHandler(handler) |
|
320 | logger.addHandler(handler) | |
321 | logger.setLevel(loglevel) |
|
321 | logger.setLevel(loglevel) | |
322 |
|
322 | |||
323 | def connect_engine_logger(context, iface, engine, loglevel=logging.DEBUG): |
|
323 | def connect_engine_logger(context, iface, engine, loglevel=logging.DEBUG): | |
324 | logger = logging.getLogger() |
|
324 | logger = logging.getLogger() | |
325 | if any([isinstance(h, handlers.PUBHandler) for h in logger.handlers]): |
|
325 | if any([isinstance(h, handlers.PUBHandler) for h in logger.handlers]): | |
326 | # don't add a second PUBHandler |
|
326 | # don't add a second PUBHandler | |
327 | return |
|
327 | return | |
328 | loglevel = integer_loglevel(loglevel) |
|
328 | loglevel = integer_loglevel(loglevel) | |
329 | lsock = context.socket(zmq.PUB) |
|
329 | lsock = context.socket(zmq.PUB) | |
330 | lsock.connect(iface) |
|
330 | lsock.connect(iface) | |
331 | handler = EnginePUBHandler(engine, lsock) |
|
331 | handler = EnginePUBHandler(engine, lsock) | |
332 | handler.setLevel(loglevel) |
|
332 | handler.setLevel(loglevel) | |
333 | logger.addHandler(handler) |
|
333 | logger.addHandler(handler) | |
334 | logger.setLevel(loglevel) |
|
334 | logger.setLevel(loglevel) | |
335 | return logger |
|
335 | return logger | |
336 |
|
336 | |||
337 | def local_logger(logname, loglevel=logging.DEBUG): |
|
337 | def local_logger(logname, loglevel=logging.DEBUG): | |
338 | loglevel = integer_loglevel(loglevel) |
|
338 | loglevel = integer_loglevel(loglevel) | |
339 | logger = logging.getLogger(logname) |
|
339 | logger = logging.getLogger(logname) | |
340 | if any([isinstance(h, logging.StreamHandler) for h in logger.handlers]): |
|
340 | if any([isinstance(h, logging.StreamHandler) for h in logger.handlers]): | |
341 | # don't add a second StreamHandler |
|
341 | # don't add a second StreamHandler | |
342 | return |
|
342 | return | |
343 | handler = logging.StreamHandler() |
|
343 | handler = logging.StreamHandler() | |
344 | handler.setLevel(loglevel) |
|
344 | handler.setLevel(loglevel) | |
345 | formatter = logging.Formatter("%(asctime)s.%(msecs).03d [%(name)s] %(message)s", |
|
345 | formatter = logging.Formatter("%(asctime)s.%(msecs).03d [%(name)s] %(message)s", | |
346 | datefmt="%Y-%m-%d %H:%M:%S") |
|
346 | datefmt="%Y-%m-%d %H:%M:%S") | |
347 | handler.setFormatter(formatter) |
|
347 | handler.setFormatter(formatter) | |
348 |
|
348 | |||
349 | logger.addHandler(handler) |
|
349 | logger.addHandler(handler) | |
350 | logger.setLevel(loglevel) |
|
350 | logger.setLevel(loglevel) | |
351 | return logger |
|
351 | return logger | |
352 |
|
352 | |||
353 | def set_hwm(sock, hwm=0): |
|
353 | def set_hwm(sock, hwm=0): | |
354 | """set zmq High Water Mark on a socket |
|
354 | """set zmq High Water Mark on a socket | |
355 |
|
355 | |||
356 | in a way that always works for various pyzmq / libzmq versions. |
|
356 | in a way that always works for various pyzmq / libzmq versions. | |
357 | """ |
|
357 | """ | |
358 | import zmq |
|
358 | import zmq | |
359 |
|
359 | |||
360 | for key in ('HWM', 'SNDHWM', 'RCVHWM'): |
|
360 | for key in ('HWM', 'SNDHWM', 'RCVHWM'): | |
361 | opt = getattr(zmq, key, None) |
|
361 | opt = getattr(zmq, key, None) | |
362 | if opt is None: |
|
362 | if opt is None: | |
363 | continue |
|
363 | continue | |
364 | try: |
|
364 | try: | |
365 | sock.setsockopt(opt, hwm) |
|
365 | sock.setsockopt(opt, hwm) | |
366 | except zmq.ZMQError: |
|
366 | except zmq.ZMQError: | |
367 | pass |
|
367 | pass | |
368 |
|
368 | |||
369 | No newline at end of file |
|
369 |
@@ -1,229 +1,229 b'' | |||||
1 | """Utilities to manipulate JSON objects. |
|
1 | """Utilities to manipulate JSON objects. | |
2 | """ |
|
2 | """ | |
3 | #----------------------------------------------------------------------------- |
|
3 | #----------------------------------------------------------------------------- | |
4 | # Copyright (C) 2010-2011 The IPython Development Team |
|
4 | # Copyright (C) 2010-2011 The IPython Development Team | |
5 | # |
|
5 | # | |
6 | # Distributed under the terms of the BSD License. The full license is in |
|
6 | # Distributed under the terms of the BSD License. The full license is in | |
7 | # the file COPYING.txt, distributed as part of this software. |
|
7 | # the file COPYING.txt, distributed as part of this software. | |
8 | #----------------------------------------------------------------------------- |
|
8 | #----------------------------------------------------------------------------- | |
9 |
|
9 | |||
10 | #----------------------------------------------------------------------------- |
|
10 | #----------------------------------------------------------------------------- | |
11 | # Imports |
|
11 | # Imports | |
12 | #----------------------------------------------------------------------------- |
|
12 | #----------------------------------------------------------------------------- | |
13 | # stdlib |
|
13 | # stdlib | |
14 | import math |
|
14 | import math | |
15 | import re |
|
15 | import re | |
16 | import types |
|
16 | import types | |
17 | from datetime import datetime |
|
17 | from datetime import datetime | |
18 |
|
18 | |||
19 | try: |
|
19 | try: | |
20 | # base64.encodestring is deprecated in Python 3.x |
|
20 | # base64.encodestring is deprecated in Python 3.x | |
21 | from base64 import encodebytes |
|
21 | from base64 import encodebytes | |
22 | except ImportError: |
|
22 | except ImportError: | |
23 | # Python 2.x |
|
23 | # Python 2.x | |
24 | from base64 import encodestring as encodebytes |
|
24 | from base64 import encodestring as encodebytes | |
25 |
|
25 | |||
26 | from IPython.utils import py3compat |
|
26 | from IPython.utils import py3compat | |
27 | from IPython.utils.py3compat import string_types, unicode_type |
|
27 | from IPython.utils.py3compat import string_types, unicode_type | |
28 | from IPython.utils.encoding import DEFAULT_ENCODING |
|
28 | from IPython.utils.encoding import DEFAULT_ENCODING | |
29 | next_attr_name = '__next__' if py3compat.PY3 else 'next' |
|
29 | next_attr_name = '__next__' if py3compat.PY3 else 'next' | |
30 |
|
30 | |||
31 | #----------------------------------------------------------------------------- |
|
31 | #----------------------------------------------------------------------------- | |
32 | # Globals and constants |
|
32 | # Globals and constants | |
33 | #----------------------------------------------------------------------------- |
|
33 | #----------------------------------------------------------------------------- | |
34 |
|
34 | |||
35 | # timestamp formats |
|
35 | # timestamp formats | |
36 | ISO8601="%Y-%m-%dT%H:%M:%S.%f" |
|
36 | ISO8601="%Y-%m-%dT%H:%M:%S.%f" | |
37 | ISO8601_PAT=re.compile(r"^(\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d+)Z?([\+\-]\d{2}:?\d{2})?$") |
|
37 | ISO8601_PAT=re.compile(r"^(\d{4}-\d{2}-\d{2}T\d{2}:\d{2}:\d{2}\.\d+)Z?([\+\-]\d{2}:?\d{2})?$") | |
38 |
|
38 | |||
39 | #----------------------------------------------------------------------------- |
|
39 | #----------------------------------------------------------------------------- | |
40 | # Classes and functions |
|
40 | # Classes and functions | |
41 | #----------------------------------------------------------------------------- |
|
41 | #----------------------------------------------------------------------------- | |
42 |
|
42 | |||
43 | def rekey(dikt): |
|
43 | def rekey(dikt): | |
44 | """Rekey a dict that has been forced to use str keys where there should be |
|
44 | """Rekey a dict that has been forced to use str keys where there should be | |
45 | ints by json.""" |
|
45 | ints by json.""" | |
46 |
for k in dikt |
|
46 | for k in dikt: | |
47 | if isinstance(k, string_types): |
|
47 | if isinstance(k, string_types): | |
48 | ik=fk=None |
|
48 | ik=fk=None | |
49 | try: |
|
49 | try: | |
50 | ik = int(k) |
|
50 | ik = int(k) | |
51 | except ValueError: |
|
51 | except ValueError: | |
52 | try: |
|
52 | try: | |
53 | fk = float(k) |
|
53 | fk = float(k) | |
54 | except ValueError: |
|
54 | except ValueError: | |
55 | continue |
|
55 | continue | |
56 | if ik is not None: |
|
56 | if ik is not None: | |
57 | nk = ik |
|
57 | nk = ik | |
58 | else: |
|
58 | else: | |
59 | nk = fk |
|
59 | nk = fk | |
60 | if nk in dikt: |
|
60 | if nk in dikt: | |
61 | raise KeyError("already have key %r"%nk) |
|
61 | raise KeyError("already have key %r"%nk) | |
62 | dikt[nk] = dikt.pop(k) |
|
62 | dikt[nk] = dikt.pop(k) | |
63 | return dikt |
|
63 | return dikt | |
64 |
|
64 | |||
65 |
|
65 | |||
66 | def extract_dates(obj): |
|
66 | def extract_dates(obj): | |
67 | """extract ISO8601 dates from unpacked JSON""" |
|
67 | """extract ISO8601 dates from unpacked JSON""" | |
68 | if isinstance(obj, dict): |
|
68 | if isinstance(obj, dict): | |
69 | obj = dict(obj) # don't clobber |
|
69 | obj = dict(obj) # don't clobber | |
70 | for k,v in obj.iteritems(): |
|
70 | for k,v in obj.iteritems(): | |
71 | obj[k] = extract_dates(v) |
|
71 | obj[k] = extract_dates(v) | |
72 | elif isinstance(obj, (list, tuple)): |
|
72 | elif isinstance(obj, (list, tuple)): | |
73 | obj = [ extract_dates(o) for o in obj ] |
|
73 | obj = [ extract_dates(o) for o in obj ] | |
74 | elif isinstance(obj, string_types): |
|
74 | elif isinstance(obj, string_types): | |
75 | m = ISO8601_PAT.match(obj) |
|
75 | m = ISO8601_PAT.match(obj) | |
76 | if m: |
|
76 | if m: | |
77 | # FIXME: add actual timezone support |
|
77 | # FIXME: add actual timezone support | |
78 | # this just drops the timezone info |
|
78 | # this just drops the timezone info | |
79 | notz = m.groups()[0] |
|
79 | notz = m.groups()[0] | |
80 | obj = datetime.strptime(notz, ISO8601) |
|
80 | obj = datetime.strptime(notz, ISO8601) | |
81 | return obj |
|
81 | return obj | |
82 |
|
82 | |||
83 | def squash_dates(obj): |
|
83 | def squash_dates(obj): | |
84 | """squash datetime objects into ISO8601 strings""" |
|
84 | """squash datetime objects into ISO8601 strings""" | |
85 | if isinstance(obj, dict): |
|
85 | if isinstance(obj, dict): | |
86 | obj = dict(obj) # don't clobber |
|
86 | obj = dict(obj) # don't clobber | |
87 | for k,v in obj.iteritems(): |
|
87 | for k,v in obj.iteritems(): | |
88 | obj[k] = squash_dates(v) |
|
88 | obj[k] = squash_dates(v) | |
89 | elif isinstance(obj, (list, tuple)): |
|
89 | elif isinstance(obj, (list, tuple)): | |
90 | obj = [ squash_dates(o) for o in obj ] |
|
90 | obj = [ squash_dates(o) for o in obj ] | |
91 | elif isinstance(obj, datetime): |
|
91 | elif isinstance(obj, datetime): | |
92 | obj = obj.isoformat() |
|
92 | obj = obj.isoformat() | |
93 | return obj |
|
93 | return obj | |
94 |
|
94 | |||
95 | def date_default(obj): |
|
95 | def date_default(obj): | |
96 | """default function for packing datetime objects in JSON.""" |
|
96 | """default function for packing datetime objects in JSON.""" | |
97 | if isinstance(obj, datetime): |
|
97 | if isinstance(obj, datetime): | |
98 | return obj.isoformat() |
|
98 | return obj.isoformat() | |
99 | else: |
|
99 | else: | |
100 | raise TypeError("%r is not JSON serializable"%obj) |
|
100 | raise TypeError("%r is not JSON serializable"%obj) | |
101 |
|
101 | |||
102 |
|
102 | |||
103 | # constants for identifying png/jpeg data |
|
103 | # constants for identifying png/jpeg data | |
104 | PNG = b'\x89PNG\r\n\x1a\n' |
|
104 | PNG = b'\x89PNG\r\n\x1a\n' | |
105 | # front of PNG base64-encoded |
|
105 | # front of PNG base64-encoded | |
106 | PNG64 = b'iVBORw0KG' |
|
106 | PNG64 = b'iVBORw0KG' | |
107 | JPEG = b'\xff\xd8' |
|
107 | JPEG = b'\xff\xd8' | |
108 | # front of JPEG base64-encoded |
|
108 | # front of JPEG base64-encoded | |
109 | JPEG64 = b'/9' |
|
109 | JPEG64 = b'/9' | |
110 |
|
110 | |||
111 | def encode_images(format_dict): |
|
111 | def encode_images(format_dict): | |
112 | """b64-encodes images in a displaypub format dict |
|
112 | """b64-encodes images in a displaypub format dict | |
113 |
|
113 | |||
114 | Perhaps this should be handled in json_clean itself? |
|
114 | Perhaps this should be handled in json_clean itself? | |
115 |
|
115 | |||
116 | Parameters |
|
116 | Parameters | |
117 | ---------- |
|
117 | ---------- | |
118 |
|
118 | |||
119 | format_dict : dict |
|
119 | format_dict : dict | |
120 | A dictionary of display data keyed by mime-type |
|
120 | A dictionary of display data keyed by mime-type | |
121 |
|
121 | |||
122 | Returns |
|
122 | Returns | |
123 | ------- |
|
123 | ------- | |
124 |
|
124 | |||
125 | format_dict : dict |
|
125 | format_dict : dict | |
126 | A copy of the same dictionary, |
|
126 | A copy of the same dictionary, | |
127 | but binary image data ('image/png' or 'image/jpeg') |
|
127 | but binary image data ('image/png' or 'image/jpeg') | |
128 | is base64-encoded. |
|
128 | is base64-encoded. | |
129 |
|
129 | |||
130 | """ |
|
130 | """ | |
131 | encoded = format_dict.copy() |
|
131 | encoded = format_dict.copy() | |
132 |
|
132 | |||
133 | pngdata = format_dict.get('image/png') |
|
133 | pngdata = format_dict.get('image/png') | |
134 | if isinstance(pngdata, bytes): |
|
134 | if isinstance(pngdata, bytes): | |
135 | # make sure we don't double-encode |
|
135 | # make sure we don't double-encode | |
136 | if not pngdata.startswith(PNG64): |
|
136 | if not pngdata.startswith(PNG64): | |
137 | pngdata = encodebytes(pngdata) |
|
137 | pngdata = encodebytes(pngdata) | |
138 | encoded['image/png'] = pngdata.decode('ascii') |
|
138 | encoded['image/png'] = pngdata.decode('ascii') | |
139 |
|
139 | |||
140 | jpegdata = format_dict.get('image/jpeg') |
|
140 | jpegdata = format_dict.get('image/jpeg') | |
141 | if isinstance(jpegdata, bytes): |
|
141 | if isinstance(jpegdata, bytes): | |
142 | # make sure we don't double-encode |
|
142 | # make sure we don't double-encode | |
143 | if not jpegdata.startswith(JPEG64): |
|
143 | if not jpegdata.startswith(JPEG64): | |
144 | jpegdata = encodebytes(jpegdata) |
|
144 | jpegdata = encodebytes(jpegdata) | |
145 | encoded['image/jpeg'] = jpegdata.decode('ascii') |
|
145 | encoded['image/jpeg'] = jpegdata.decode('ascii') | |
146 |
|
146 | |||
147 | return encoded |
|
147 | return encoded | |
148 |
|
148 | |||
149 |
|
149 | |||
150 | def json_clean(obj): |
|
150 | def json_clean(obj): | |
151 | """Clean an object to ensure it's safe to encode in JSON. |
|
151 | """Clean an object to ensure it's safe to encode in JSON. | |
152 |
|
152 | |||
153 | Atomic, immutable objects are returned unmodified. Sets and tuples are |
|
153 | Atomic, immutable objects are returned unmodified. Sets and tuples are | |
154 | converted to lists, lists are copied and dicts are also copied. |
|
154 | converted to lists, lists are copied and dicts are also copied. | |
155 |
|
155 | |||
156 | Note: dicts whose keys could cause collisions upon encoding (such as a dict |
|
156 | Note: dicts whose keys could cause collisions upon encoding (such as a dict | |
157 | with both the number 1 and the string '1' as keys) will cause a ValueError |
|
157 | with both the number 1 and the string '1' as keys) will cause a ValueError | |
158 | to be raised. |
|
158 | to be raised. | |
159 |
|
159 | |||
160 | Parameters |
|
160 | Parameters | |
161 | ---------- |
|
161 | ---------- | |
162 | obj : any python object |
|
162 | obj : any python object | |
163 |
|
163 | |||
164 | Returns |
|
164 | Returns | |
165 | ------- |
|
165 | ------- | |
166 | out : object |
|
166 | out : object | |
167 |
|
167 | |||
168 | A version of the input which will not cause an encoding error when |
|
168 | A version of the input which will not cause an encoding error when | |
169 | encoded as JSON. Note that this function does not *encode* its inputs, |
|
169 | encoded as JSON. Note that this function does not *encode* its inputs, | |
170 | it simply sanitizes it so that there will be no encoding errors later. |
|
170 | it simply sanitizes it so that there will be no encoding errors later. | |
171 |
|
171 | |||
172 | Examples |
|
172 | Examples | |
173 | -------- |
|
173 | -------- | |
174 | >>> json_clean(4) |
|
174 | >>> json_clean(4) | |
175 | 4 |
|
175 | 4 | |
176 | >>> json_clean(range(10)) |
|
176 | >>> json_clean(range(10)) | |
177 | [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] |
|
177 | [0, 1, 2, 3, 4, 5, 6, 7, 8, 9] | |
178 | >>> sorted(json_clean(dict(x=1, y=2)).items()) |
|
178 | >>> sorted(json_clean(dict(x=1, y=2)).items()) | |
179 | [('x', 1), ('y', 2)] |
|
179 | [('x', 1), ('y', 2)] | |
180 | >>> sorted(json_clean(dict(x=1, y=2, z=[1,2,3])).items()) |
|
180 | >>> sorted(json_clean(dict(x=1, y=2, z=[1,2,3])).items()) | |
181 | [('x', 1), ('y', 2), ('z', [1, 2, 3])] |
|
181 | [('x', 1), ('y', 2), ('z', [1, 2, 3])] | |
182 | >>> json_clean(True) |
|
182 | >>> json_clean(True) | |
183 | True |
|
183 | True | |
184 | """ |
|
184 | """ | |
185 | # types that are 'atomic' and ok in json as-is. bool doesn't need to be |
|
185 | # types that are 'atomic' and ok in json as-is. bool doesn't need to be | |
186 | # listed explicitly because bools pass as int instances |
|
186 | # listed explicitly because bools pass as int instances | |
187 | atomic_ok = (unicode_type, int, types.NoneType) |
|
187 | atomic_ok = (unicode_type, int, types.NoneType) | |
188 |
|
188 | |||
189 | # containers that we need to convert into lists |
|
189 | # containers that we need to convert into lists | |
190 | container_to_list = (tuple, set, types.GeneratorType) |
|
190 | container_to_list = (tuple, set, types.GeneratorType) | |
191 |
|
191 | |||
192 | if isinstance(obj, float): |
|
192 | if isinstance(obj, float): | |
193 | # cast out-of-range floats to their reprs |
|
193 | # cast out-of-range floats to their reprs | |
194 | if math.isnan(obj) or math.isinf(obj): |
|
194 | if math.isnan(obj) or math.isinf(obj): | |
195 | return repr(obj) |
|
195 | return repr(obj) | |
196 | return obj |
|
196 | return obj | |
197 |
|
197 | |||
198 | if isinstance(obj, atomic_ok): |
|
198 | if isinstance(obj, atomic_ok): | |
199 | return obj |
|
199 | return obj | |
200 |
|
200 | |||
201 | if isinstance(obj, bytes): |
|
201 | if isinstance(obj, bytes): | |
202 | return obj.decode(DEFAULT_ENCODING, 'replace') |
|
202 | return obj.decode(DEFAULT_ENCODING, 'replace') | |
203 |
|
203 | |||
204 | if isinstance(obj, container_to_list) or ( |
|
204 | if isinstance(obj, container_to_list) or ( | |
205 | hasattr(obj, '__iter__') and hasattr(obj, next_attr_name)): |
|
205 | hasattr(obj, '__iter__') and hasattr(obj, next_attr_name)): | |
206 | obj = list(obj) |
|
206 | obj = list(obj) | |
207 |
|
207 | |||
208 | if isinstance(obj, list): |
|
208 | if isinstance(obj, list): | |
209 | return [json_clean(x) for x in obj] |
|
209 | return [json_clean(x) for x in obj] | |
210 |
|
210 | |||
211 | if isinstance(obj, dict): |
|
211 | if isinstance(obj, dict): | |
212 | # First, validate that the dict won't lose data in conversion due to |
|
212 | # First, validate that the dict won't lose data in conversion due to | |
213 | # key collisions after stringification. This can happen with keys like |
|
213 | # key collisions after stringification. This can happen with keys like | |
214 | # True and 'true' or 1 and '1', which collide in JSON. |
|
214 | # True and 'true' or 1 and '1', which collide in JSON. | |
215 | nkeys = len(obj) |
|
215 | nkeys = len(obj) | |
216 | nkeys_collapsed = len(set(map(str, obj))) |
|
216 | nkeys_collapsed = len(set(map(str, obj))) | |
217 | if nkeys != nkeys_collapsed: |
|
217 | if nkeys != nkeys_collapsed: | |
218 | raise ValueError('dict can not be safely converted to JSON: ' |
|
218 | raise ValueError('dict can not be safely converted to JSON: ' | |
219 | 'key collision would lead to dropped values') |
|
219 | 'key collision would lead to dropped values') | |
220 | # If all OK, proceed by making the new dict that will be json-safe |
|
220 | # If all OK, proceed by making the new dict that will be json-safe | |
221 | out = {} |
|
221 | out = {} | |
222 | for k,v in obj.iteritems(): |
|
222 | for k,v in obj.iteritems(): | |
223 | out[str(k)] = json_clean(v) |
|
223 | out[str(k)] = json_clean(v) | |
224 | return out |
|
224 | return out | |
225 |
|
225 | |||
226 | # If we get here, we don't know how to handle the object, so we just get |
|
226 | # If we get here, we don't know how to handle the object, so we just get | |
227 | # its repr and return that. This will catch lambdas, open sockets, class |
|
227 | # its repr and return that. This will catch lambdas, open sockets, class | |
228 | # objects, and any other complicated contraption that json can't encode |
|
228 | # objects, and any other complicated contraption that json can't encode | |
229 | return repr(obj) |
|
229 | return repr(obj) |
General Comments 0
You need to be logged in to leave comments.
Login now