Show More
@@ -1,260 +1,375 b'' | |||||
1 | """Analysis of text input into executable blocks. |
|
1 | """Analysis of text input into executable blocks. | |
2 |
|
2 | |||
3 | This is a simple example of how an interactive terminal-based client can use |
|
3 | This is a simple example of how an interactive terminal-based client can use | |
4 | this tool:: |
|
4 | this tool:: | |
5 |
|
5 | |||
6 | bb = BlockBreaker() |
|
6 | bb = BlockBreaker() | |
7 | while not bb.interactive_block_ready(): |
|
7 | while not bb.interactive_block_ready(): | |
8 | bb.push(raw_input('>>> ')) |
|
8 | bb.push(raw_input('>>> ')) | |
9 | print 'Input source was:\n', bb.source, |
|
9 | print 'Input source was:\n', bb.source, | |
10 | """ |
|
10 | """ | |
11 | #----------------------------------------------------------------------------- |
|
11 | #----------------------------------------------------------------------------- | |
12 | # Copyright (C) 2010 The IPython Development Team |
|
12 | # Copyright (C) 2010 The IPython Development Team | |
13 | # |
|
13 | # | |
14 | # Distributed under the terms of the BSD License. The full license is in |
|
14 | # Distributed under the terms of the BSD License. The full license is in | |
15 | # the file COPYING, distributed as part of this software. |
|
15 | # the file COPYING, distributed as part of this software. | |
16 | #----------------------------------------------------------------------------- |
|
16 | #----------------------------------------------------------------------------- | |
17 |
|
17 | |||
18 | #----------------------------------------------------------------------------- |
|
18 | #----------------------------------------------------------------------------- | |
19 | # Imports |
|
19 | # Imports | |
20 | #----------------------------------------------------------------------------- |
|
20 | #----------------------------------------------------------------------------- | |
21 | # stdlib |
|
21 | # stdlib | |
22 | import codeop |
|
22 | import codeop | |
23 | import re |
|
23 | import re | |
24 | import sys |
|
24 | import sys | |
25 |
|
25 | |||
26 | #----------------------------------------------------------------------------- |
|
26 | #----------------------------------------------------------------------------- | |
27 | # Utilities |
|
27 | # Utilities | |
28 | #----------------------------------------------------------------------------- |
|
28 | #----------------------------------------------------------------------------- | |
29 |
|
29 | |||
30 | # FIXME: move these utilities to the general ward... |
|
30 | # FIXME: move these utilities to the general ward... | |
31 |
|
31 | |||
32 | # compiled regexps for autoindent management |
|
32 | # compiled regexps for autoindent management | |
33 | dedent_re = re.compile(r'^\s+raise|^\s+return|^\s+pass') |
|
33 | dedent_re = re.compile(r'^\s+raise|^\s+return|^\s+pass') | |
34 | ini_spaces_re = re.compile(r'^([ \t\r\f\v]+)') |
|
34 | ini_spaces_re = re.compile(r'^([ \t\r\f\v]+)') | |
35 |
|
35 | |||
36 |
|
36 | |||
37 | def num_ini_spaces(s): |
|
37 | def num_ini_spaces(s): | |
38 | """Return the number of initial spaces in a string. |
|
38 | """Return the number of initial spaces in a string. | |
39 |
|
39 | |||
40 | Note that tabs are counted as a single space. For now, we do *not* support |
|
40 | Note that tabs are counted as a single space. For now, we do *not* support | |
41 | mixing of tabs and spaces in the user's input. |
|
41 | mixing of tabs and spaces in the user's input. | |
42 |
|
42 | |||
43 | Parameters |
|
43 | Parameters | |
44 | ---------- |
|
44 | ---------- | |
45 | s : string |
|
45 | s : string | |
46 | """ |
|
46 | """ | |
47 |
|
47 | |||
48 | ini_spaces = ini_spaces_re.match(s) |
|
48 | ini_spaces = ini_spaces_re.match(s) | |
49 | if ini_spaces: |
|
49 | if ini_spaces: | |
50 | return ini_spaces.end() |
|
50 | return ini_spaces.end() | |
51 | else: |
|
51 | else: | |
52 | return 0 |
|
52 | return 0 | |
53 |
|
53 | |||
54 |
|
54 | |||
55 | def remove_comments(src): |
|
55 | def remove_comments(src): | |
56 | """Remove all comments from input source. |
|
56 | """Remove all comments from input source. | |
57 |
|
57 | |||
58 | Note: comments are NOT recognized inside of strings! |
|
58 | Note: comments are NOT recognized inside of strings! | |
59 |
|
59 | |||
60 | Parameters |
|
60 | Parameters | |
61 | ---------- |
|
61 | ---------- | |
62 | src : string |
|
62 | src : string | |
63 | A single or multiline input string. |
|
63 | A single or multiline input string. | |
64 |
|
64 | |||
65 | Returns |
|
65 | Returns | |
66 | ------- |
|
66 | ------- | |
67 | String with all Python comments removed. |
|
67 | String with all Python comments removed. | |
68 | """ |
|
68 | """ | |
69 |
|
69 | |||
70 | return re.sub('#.*', '', src) |
|
70 | return re.sub('#.*', '', src) | |
71 |
|
71 | |||
72 |
|
72 | |||
73 | def get_input_encoding(): |
|
73 | def get_input_encoding(): | |
74 | """Return the default standard input encoding.""" |
|
74 | """Return the default standard input encoding.""" | |
75 | return getattr(sys.stdin, 'encoding', 'ascii') |
|
75 | return getattr(sys.stdin, 'encoding', 'ascii') | |
76 |
|
76 | |||
77 | #----------------------------------------------------------------------------- |
|
77 | #----------------------------------------------------------------------------- | |
78 | # Classes and functions |
|
78 | # Classes and functions | |
79 | #----------------------------------------------------------------------------- |
|
79 | #----------------------------------------------------------------------------- | |
80 |
|
80 | |||
81 | class BlockBreaker(object): |
|
81 | class BlockBreaker(object): | |
82 | # Command compiler |
|
82 | # Command compiler | |
83 | compile = None |
|
83 | compile = None | |
84 | # Number of spaces of indentation |
|
84 | # Number of spaces of indentation | |
85 | indent_spaces = 0 |
|
85 | indent_spaces = 0 | |
|
86 | # Mark when input has changed indentation all the way back to flush-left | |||
|
87 | full_dedent = False | |||
86 | # String, indicating the default input encoding |
|
88 | # String, indicating the default input encoding | |
87 | encoding = '' |
|
89 | encoding = '' | |
88 | # String where the current full source input is stored, properly encoded |
|
90 | # String where the current full source input is stored, properly encoded | |
89 | source = '' |
|
91 | source = '' | |
90 | # Code object corresponding to the current source |
|
92 | # Code object corresponding to the current source | |
91 | code = None |
|
93 | code = None | |
92 | # Boolean indicating whether the current block is complete |
|
94 | # Boolean indicating whether the current block is complete | |
93 | is_complete = None |
|
95 | is_complete = None | |
94 | # Input mode |
|
96 | # Input mode | |
95 | input_mode = 'append' |
|
97 | input_mode = 'append' | |
96 |
|
98 | |||
97 | # Private attributes |
|
99 | # Private attributes | |
98 |
|
100 | |||
99 | # List |
|
101 | # List | |
100 | _buffer = None |
|
102 | _buffer = None | |
101 |
|
103 | |||
102 | def __init__(self, input_mode=None): |
|
104 | def __init__(self, input_mode=None): | |
103 | """Create a new BlockBreaker instance. |
|
105 | """Create a new BlockBreaker instance. | |
104 |
|
106 | |||
105 | Parameters |
|
107 | Parameters | |
106 | ---------- |
|
108 | ---------- | |
107 | input_mode : str |
|
109 | input_mode : str | |
108 |
|
110 | |||
109 | One of 'append', 'replace', default is 'append'. This controls how |
|
111 | One of 'append', 'replace', default is 'append'. This controls how | |
110 | new inputs are used: in 'append' mode, they are appended to the |
|
112 | new inputs are used: in 'append' mode, they are appended to the | |
111 | existing buffer and the whole buffer is compiled; in 'replace' mode, |
|
113 | existing buffer and the whole buffer is compiled; in 'replace' mode, | |
112 | each new input completely replaces all prior inputs. Replace mode is |
|
114 | each new input completely replaces all prior inputs. Replace mode is | |
113 | thus equivalent to prepending a full reset() to every push() call. |
|
115 | thus equivalent to prepending a full reset() to every push() call. | |
114 |
|
116 | |||
115 | In practice, line-oriented clients likely want to use 'append' mode |
|
117 | In practice, line-oriented clients likely want to use 'append' mode | |
116 | while block-oriented ones will want to use 'replace'. |
|
118 | while block-oriented ones will want to use 'replace'. | |
117 | """ |
|
119 | """ | |
118 | self._buffer = [] |
|
120 | self._buffer = [] | |
119 | self.compile = codeop.CommandCompiler() |
|
121 | self.compile = codeop.CommandCompiler() | |
120 | self.encoding = get_input_encoding() |
|
122 | self.encoding = get_input_encoding() | |
121 | self.input_mode = BlockBreaker.input_mode if input_mode is None \ |
|
123 | self.input_mode = BlockBreaker.input_mode if input_mode is None \ | |
122 | else input_mode |
|
124 | else input_mode | |
123 |
|
125 | |||
124 | def reset(self): |
|
126 | def reset(self): | |
125 | """Reset the input buffer and associated state.""" |
|
127 | """Reset the input buffer and associated state.""" | |
126 | self.indent_spaces = 0 |
|
128 | self.indent_spaces = 0 | |
127 | self._buffer[:] = [] |
|
129 | self._buffer[:] = [] | |
128 | self.source = '' |
|
130 | self.source = '' | |
129 | self.code = None |
|
131 | self.code = None | |
|
132 | self.is_complete = False | |||
|
133 | self.full_dedent = False | |||
130 |
|
134 | |||
131 | def source_reset(self): |
|
135 | def source_reset(self): | |
132 | """Return the input source and perform a full reset. |
|
136 | """Return the input source and perform a full reset. | |
133 | """ |
|
137 | """ | |
134 | out = self.source |
|
138 | out = self.source | |
135 | self.reset() |
|
139 | self.reset() | |
136 | return out |
|
140 | return out | |
137 |
|
141 | |||
138 | def push(self, lines): |
|
142 | def push(self, lines): | |
139 | """Push one ore more lines of input. |
|
143 | """Push one ore more lines of input. | |
140 |
|
144 | |||
141 | This stores the given lines and returns a status code indicating |
|
145 | This stores the given lines and returns a status code indicating | |
142 | whether the code forms a complete Python block or not. |
|
146 | whether the code forms a complete Python block or not. | |
143 |
|
147 | |||
144 | Any exceptions generated in compilation are allowed to propagate. |
|
148 | Any exceptions generated in compilation are allowed to propagate. | |
145 |
|
149 | |||
146 | Parameters |
|
150 | Parameters | |
147 | ---------- |
|
151 | ---------- | |
148 | lines : string |
|
152 | lines : string | |
149 | One or more lines of Python input. |
|
153 | One or more lines of Python input. | |
150 |
|
154 | |||
151 | Returns |
|
155 | Returns | |
152 | ------- |
|
156 | ------- | |
153 | is_complete : boolean |
|
157 | is_complete : boolean | |
154 | True if the current input source (the result of the current input |
|
158 | True if the current input source (the result of the current input | |
155 | plus prior inputs) forms a complete Python execution block. Note that |
|
159 | plus prior inputs) forms a complete Python execution block. Note that | |
156 | this value is also stored as an attribute so it can be queried at any |
|
160 | this value is also stored as an attribute so it can be queried at any | |
157 | time. |
|
161 | time. | |
158 | """ |
|
162 | """ | |
159 | if self.input_mode == 'replace': |
|
163 | if self.input_mode == 'replace': | |
160 | self.reset() |
|
164 | self.reset() | |
161 |
|
165 | |||
162 | # If the source code has leading blanks, add 'if 1:\n' to it |
|
166 | # If the source code has leading blanks, add 'if 1:\n' to it | |
163 | # this allows execution of indented pasted code. It is tempting |
|
167 | # this allows execution of indented pasted code. It is tempting | |
164 | # to add '\n' at the end of source to run commands like ' a=1' |
|
168 | # to add '\n' at the end of source to run commands like ' a=1' | |
165 | # directly, but this fails for more complicated scenarios |
|
169 | # directly, but this fails for more complicated scenarios | |
166 | if not self._buffer and lines[:1] in [' ', '\t']: |
|
170 | if not self._buffer and lines[:1] in [' ', '\t']: | |
167 | lines = 'if 1:\n%s' % lines |
|
171 | lines = 'if 1:\n%s' % lines | |
168 |
|
172 | |||
169 | self._store(lines) |
|
173 | self._store(lines) | |
170 | source = self.source |
|
174 | source = self.source | |
171 |
|
175 | |||
172 | # Before calling compile(), reset the code object to None so that if an |
|
176 | # Before calling compile(), reset the code object to None so that if an | |
173 | # exception is raised in compilation, we don't mislead by having |
|
177 | # exception is raised in compilation, we don't mislead by having | |
174 | # inconsistent code/source attributes. |
|
178 | # inconsistent code/source attributes. | |
175 | self.code, self.is_complete = None, None |
|
179 | self.code, self.is_complete = None, None | |
|
180 | ||||
|
181 | self._update_indent(lines) | |||
176 | try: |
|
182 | try: | |
177 | self.code = self.compile(source) |
|
183 | self.code = self.compile(source) | |
178 | # Invalid syntax can produce any of a number of different errors from |
|
184 | # Invalid syntax can produce any of a number of different errors from | |
179 | # inside the compiler, so we have to catch them all. Syntax errors |
|
185 | # inside the compiler, so we have to catch them all. Syntax errors | |
180 | # immediately produce a 'ready' block, so the invalid Python can be |
|
186 | # immediately produce a 'ready' block, so the invalid Python can be | |
181 | # sent to the kernel for evaluation with possible ipython |
|
187 | # sent to the kernel for evaluation with possible ipython | |
182 | # special-syntax conversion. |
|
188 | # special-syntax conversion. | |
183 |
except (SyntaxError, OverflowError, ValueError, TypeError, |
|
189 | except (SyntaxError, OverflowError, ValueError, TypeError, | |
|
190 | MemoryError): | |||
184 | self.is_complete = True |
|
191 | self.is_complete = True | |
185 | else: |
|
192 | else: | |
186 | # Compilation didn't produce any exceptions (though it may not have |
|
193 | # Compilation didn't produce any exceptions (though it may not have | |
187 | # given a complete code object) |
|
194 | # given a complete code object) | |
188 | self.is_complete = self.code is not None |
|
195 | self.is_complete = self.code is not None | |
189 | self._update_indent(lines) |
|
|||
190 |
|
196 | |||
191 | return self.is_complete |
|
197 | return self.is_complete | |
192 |
|
198 | |||
193 | def interactive_block_ready(self): |
|
199 | def interactive_block_ready(self): | |
194 | """Return whether a block of interactive input is ready for execution. |
|
200 | """Return whether a block of interactive input is ready for execution. | |
195 |
|
201 | |||
196 | This method is meant to be used by line-oriented frontends, who need to |
|
202 | This method is meant to be used by line-oriented frontends, who need to | |
197 | guess whether a block is complete or not based solely on prior and |
|
203 | guess whether a block is complete or not based solely on prior and | |
198 | current input lines. The BlockBreaker considers it has a complete |
|
204 | current input lines. The BlockBreaker considers it has a complete | |
199 | interactive block when *all* of the following are true: |
|
205 | interactive block when *all* of the following are true: | |
200 |
|
206 | |||
201 | 1. The input compiles to a complete statement. |
|
207 | 1. The input compiles to a complete statement. | |
202 |
|
208 | |||
203 | 2. The indentation level is flush-left (because if we are indented, |
|
209 | 2. The indentation level is flush-left (because if we are indented, | |
204 | like inside a function definition or for loop, we need to keep |
|
210 | like inside a function definition or for loop, we need to keep | |
205 | reading new input). |
|
211 | reading new input). | |
206 |
|
212 | |||
207 | 3. There is one extra line consisting only of whitespace. |
|
213 | 3. There is one extra line consisting only of whitespace. | |
208 |
|
214 | |||
209 | Because of condition #3, this method should be used only by |
|
215 | Because of condition #3, this method should be used only by | |
210 | *line-oriented* frontends, since it means that intermediate blank lines |
|
216 | *line-oriented* frontends, since it means that intermediate blank lines | |
211 | are not allowed in function definitions (or any other indented block). |
|
217 | are not allowed in function definitions (or any other indented block). | |
212 |
|
218 | |||
213 | Block-oriented frontends that have a separate keyboard event to |
|
219 | Block-oriented frontends that have a separate keyboard event to | |
214 | indicate execution should use the :meth:`split_blocks` method instead. |
|
220 | indicate execution should use the :meth:`split_blocks` method instead. | |
215 | """ |
|
221 | """ | |
|
222 | #print 'complete?', self.source # dbg | |||
|
223 | #if self.full_dedent: | |||
|
224 | # True | |||
|
225 | ||||
216 | if not self.is_complete: |
|
226 | if not self.is_complete: | |
217 | return False |
|
227 | return False | |
218 | if self.indent_spaces==0: |
|
228 | if self.indent_spaces==0: | |
219 | return True |
|
229 | return True | |
220 | last_line = self.source.splitlines()[-1] |
|
230 | last_line = self.source.splitlines()[-1] | |
221 | if not last_line or last_line.isspace(): |
|
231 | if not last_line or last_line.isspace(): | |
222 | return True |
|
232 | return True | |
223 | else: |
|
233 | else: | |
224 | return False |
|
234 | return False | |
225 |
|
235 | |||
226 | def split_blocks(self, lines): |
|
236 | def split_blocks(self, lines): | |
227 |
"""Split a multiline string into multiple input blocks |
|
237 | """Split a multiline string into multiple input blocks. | |
228 | raise NotImplementedError |
|
238 | ||
|
239 | Note: this method starts by performing a full reset(). | |||
|
240 | ||||
|
241 | Parameters | |||
|
242 | ---------- | |||
|
243 | lines : str | |||
|
244 | A possibly multiline string. | |||
|
245 | ||||
|
246 | Returns | |||
|
247 | ------- | |||
|
248 | blocks : list | |||
|
249 | A list of strings, each possibly multiline. Each string corresponds | |||
|
250 | to a single block that can be compiled in 'single' mode (unless it | |||
|
251 | has a syntax error).""" | |||
|
252 | ||||
|
253 | # This code is fairly delicate. If you make any changes here, make | |||
|
254 | # absolutely sure that you do run the full test suite and ALL tests | |||
|
255 | # pass. | |||
|
256 | ||||
|
257 | self.reset() | |||
|
258 | blocks = [] | |||
|
259 | ||||
|
260 | # Reversed copy so we can use pop() efficiently and consume the input | |||
|
261 | # as a stack | |||
|
262 | lines = lines.splitlines()[::-1] | |||
|
263 | # Outer loop over all input | |||
|
264 | while lines: | |||
|
265 | # Inner loop to build each block | |||
|
266 | while True: | |||
|
267 | # Safety exit from inner loop | |||
|
268 | if not lines: | |||
|
269 | break | |||
|
270 | # Grab next line but don't push it yet | |||
|
271 | next_line = lines.pop() | |||
|
272 | # Blank/empty lines are pushed as-is | |||
|
273 | if not next_line or next_line.isspace(): | |||
|
274 | self.push(next_line) | |||
|
275 | continue | |||
|
276 | ||||
|
277 | # Check indentation changes caused by the *next* line | |||
|
278 | indent_spaces, full_dedent = self._find_indent(next_line) | |||
|
279 | ||||
|
280 | # If the next line causes a dedent, it can be for two differnt | |||
|
281 | # reasons: either an explicit de-dent by the user or a | |||
|
282 | # return/raise/pass statement. These MUST be handled | |||
|
283 | # separately: | |||
|
284 | # | |||
|
285 | # 1. the first case is only detected when the actual explicit | |||
|
286 | # dedent happens, and that would be the *first* line of a *new* | |||
|
287 | # block. Thus, we must put the line back into the input buffer | |||
|
288 | # so that it starts a new block on the next pass. | |||
|
289 | # | |||
|
290 | # 2. the second case is detected in the line before the actual | |||
|
291 | # dedent happens, so , we consume the line and we can break out | |||
|
292 | # to start a new block. | |||
|
293 | ||||
|
294 | # Case 1, explicit dedent causes a break | |||
|
295 | if full_dedent and not next_line.startswith(' '): | |||
|
296 | lines.append(next_line) | |||
|
297 | break | |||
|
298 | ||||
|
299 | # Otherwise any line is pushed | |||
|
300 | self.push(next_line) | |||
|
301 | ||||
|
302 | # Case 2, full dedent with full block ready: | |||
|
303 | if full_dedent or \ | |||
|
304 | self.indent_spaces==0 and self.interactive_block_ready(): | |||
|
305 | break | |||
|
306 | # Form the new block with the current source input | |||
|
307 | blocks.append(self.source_reset()) | |||
|
308 | ||||
|
309 | return blocks | |||
229 |
|
310 | |||
230 | #------------------------------------------------------------------------ |
|
311 | #------------------------------------------------------------------------ | |
231 | # Private interface |
|
312 | # Private interface | |
232 | #------------------------------------------------------------------------ |
|
313 | #------------------------------------------------------------------------ | |
233 |
|
||||
234 | def _update_indent(self, lines): |
|
|||
235 | """Keep track of the indent level.""" |
|
|||
236 |
|
314 | |||
237 | for line in remove_comments(lines).splitlines(): |
|
315 | def _find_indent(self, line): | |
|
316 | """Compute the new indentation level for a single line. | |||
|
317 | ||||
|
318 | Parameters | |||
|
319 | ---------- | |||
|
320 | line : str | |||
|
321 | A single new line of non-whitespace, non-comment Python input. | |||
|
322 | ||||
|
323 | Returns | |||
|
324 | ------- | |||
|
325 | indent_spaces : int | |||
|
326 | New value for the indent level (it may be equal to self.indent_spaces | |||
|
327 | if indentation doesn't change. | |||
|
328 | ||||
|
329 | full_dedent : boolean | |||
|
330 | Whether the new line causes a full flush-left dedent. | |||
|
331 | """ | |||
|
332 | indent_spaces = self.indent_spaces | |||
|
333 | full_dedent = self.full_dedent | |||
|
334 | ||||
|
335 | inisp = num_ini_spaces(line) | |||
|
336 | if inisp < indent_spaces: | |||
|
337 | indent_spaces = inisp | |||
|
338 | if indent_spaces <= 0: | |||
|
339 | #print 'Full dedent in text',self.source # dbg | |||
|
340 | full_dedent = True | |||
|
341 | ||||
|
342 | if line[-1] == ':': | |||
|
343 | indent_spaces += 4 | |||
|
344 | elif dedent_re.match(line): | |||
|
345 | indent_spaces -= 4 | |||
|
346 | if indent_spaces <= 0: | |||
|
347 | full_dedent = True | |||
|
348 | ||||
|
349 | # Safety | |||
|
350 | if indent_spaces < 0: | |||
|
351 | indent_spaces = 0 | |||
|
352 | #print 'safety' # dbg | |||
238 |
|
353 | |||
|
354 | return indent_spaces, full_dedent | |||
|
355 | ||||
|
356 | def _update_indent(self, lines): | |||
|
357 | for line in remove_comments(lines).splitlines(): | |||
239 | if line and not line.isspace(): |
|
358 | if line and not line.isspace(): | |
240 | if self.code is not None: |
|
359 | self.indent_spaces, self.full_dedent = self._find_indent(line) | |
241 | inisp = num_ini_spaces(line) |
|
|||
242 | if inisp < self.indent_spaces: |
|
|||
243 | self.indent_spaces = inisp |
|
|||
244 |
|
||||
245 | if line[-1] == ':': |
|
|||
246 | self.indent_spaces += 4 |
|
|||
247 | elif dedent_re.match(line): |
|
|||
248 | self.indent_spaces -= 4 |
|
|||
249 |
|
360 | |||
250 | def _store(self, lines): |
|
361 | def _store(self, lines): | |
251 | """Store one or more lines of input. |
|
362 | """Store one or more lines of input. | |
252 |
|
363 | |||
253 | If input lines are not newline-terminated, a newline is automatically |
|
364 | If input lines are not newline-terminated, a newline is automatically | |
254 | appended.""" |
|
365 | appended.""" | |
255 |
|
366 | |||
256 | if lines.endswith('\n'): |
|
367 | if lines.endswith('\n'): | |
257 | self._buffer.append(lines) |
|
368 | self._buffer.append(lines) | |
258 | else: |
|
369 | else: | |
259 | self._buffer.append(lines+'\n') |
|
370 | self._buffer.append(lines+'\n') | |
|
371 | self._set_source() | |||
|
372 | ||||
|
373 | def _set_source(self): | |||
260 | self.source = ''.join(self._buffer).encode(self.encoding) |
|
374 | self.source = ''.join(self._buffer).encode(self.encoding) | |
|
375 |
@@ -1,189 +1,276 b'' | |||||
1 | """Tests for the blockbreaker module. |
|
1 | """Tests for the blockbreaker module. | |
2 | """ |
|
2 | """ | |
3 | #----------------------------------------------------------------------------- |
|
3 | #----------------------------------------------------------------------------- | |
4 | # Copyright (C) 2010 The IPython Development Team |
|
4 | # Copyright (C) 2010 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, distributed as part of this software. |
|
7 | # the file COPYING, distributed as part of this software. | |
8 | #----------------------------------------------------------------------------- |
|
8 | #----------------------------------------------------------------------------- | |
9 |
|
9 | |||
10 | #----------------------------------------------------------------------------- |
|
10 | #----------------------------------------------------------------------------- | |
11 | # Imports |
|
11 | # Imports | |
12 | #----------------------------------------------------------------------------- |
|
12 | #----------------------------------------------------------------------------- | |
13 | # stdlib |
|
13 | # stdlib | |
14 | import unittest |
|
14 | import unittest | |
15 |
|
15 | |||
16 | # Third party |
|
16 | # Third party | |
17 | import nose.tools as nt |
|
17 | import nose.tools as nt | |
18 |
|
18 | |||
19 | # Our own |
|
19 | # Our own | |
20 | from IPython.core import blockbreaker as BB |
|
20 | from IPython.core import blockbreaker as BB | |
21 |
|
21 | |||
22 | #----------------------------------------------------------------------------- |
|
22 | #----------------------------------------------------------------------------- | |
|
23 | # Test utilities, just for local use | |||
|
24 | #----------------------------------------------------------------------------- | |||
|
25 | ||||
|
26 | def assemble(block): | |||
|
27 | """Assemble a block into multi-line sub-blocks.""" | |||
|
28 | return ['\n'.join(sub_block)+'\n' for sub_block in block] | |||
|
29 | ||||
|
30 | #----------------------------------------------------------------------------- | |||
23 | # Tests |
|
31 | # Tests | |
24 | #----------------------------------------------------------------------------- |
|
32 | #----------------------------------------------------------------------------- | |
25 | def test_spaces(): |
|
33 | def test_spaces(): | |
26 | tests = [('', 0), |
|
34 | tests = [('', 0), | |
27 | (' ', 1), |
|
35 | (' ', 1), | |
28 | ('\n', 0), |
|
36 | ('\n', 0), | |
29 | (' \n', 1), |
|
37 | (' \n', 1), | |
30 | ('x', 0), |
|
38 | ('x', 0), | |
31 | (' x', 1), |
|
39 | (' x', 1), | |
32 | (' x',2), |
|
40 | (' x',2), | |
33 | (' x',4), |
|
41 | (' x',4), | |
34 | # Note: tabs are counted as a single whitespace! |
|
42 | # Note: tabs are counted as a single whitespace! | |
35 | ('\tx', 1), |
|
43 | ('\tx', 1), | |
36 | ('\t x', 2), |
|
44 | ('\t x', 2), | |
37 | ] |
|
45 | ] | |
38 |
|
46 | |||
39 | for s, nsp in tests: |
|
47 | for s, nsp in tests: | |
40 | nt.assert_equal(BB.num_ini_spaces(s), nsp) |
|
48 | nt.assert_equal(BB.num_ini_spaces(s), nsp) | |
41 |
|
49 | |||
42 |
|
50 | |||
43 | def test_remove_comments(): |
|
51 | def test_remove_comments(): | |
44 | tests = [('text', 'text'), |
|
52 | tests = [('text', 'text'), | |
45 | ('text # comment', 'text '), |
|
53 | ('text # comment', 'text '), | |
46 | ('text # comment\n', 'text \n'), |
|
54 | ('text # comment\n', 'text \n'), | |
47 | ('text # comment \n', 'text \n'), |
|
55 | ('text # comment \n', 'text \n'), | |
48 | ('line # c \nline\n','line \nline\n'), |
|
56 | ('line # c \nline\n','line \nline\n'), | |
49 | ('line # c \nline#c2 \nline\nline #c\n\n', |
|
57 | ('line # c \nline#c2 \nline\nline #c\n\n', | |
50 | 'line \nline\nline\nline \n\n'), |
|
58 | 'line \nline\nline\nline \n\n'), | |
51 | ] |
|
59 | ] | |
52 |
|
60 | |||
53 | for inp, out in tests: |
|
61 | for inp, out in tests: | |
54 | nt.assert_equal(BB.remove_comments(inp), out) |
|
62 | nt.assert_equal(BB.remove_comments(inp), out) | |
55 |
|
63 | |||
56 |
|
64 | |||
57 | def test_get_input_encoding(): |
|
65 | def test_get_input_encoding(): | |
58 | encoding = BB.get_input_encoding() |
|
66 | encoding = BB.get_input_encoding() | |
59 | nt.assert_true(isinstance(encoding, basestring)) |
|
67 | nt.assert_true(isinstance(encoding, basestring)) | |
60 | # simple-minded check that at least encoding a simple string works with the |
|
68 | # simple-minded check that at least encoding a simple string works with the | |
61 | # encoding we got. |
|
69 | # encoding we got. | |
62 | nt.assert_equal('test'.encode(encoding), 'test') |
|
70 | nt.assert_equal('test'.encode(encoding), 'test') | |
63 |
|
71 | |||
64 |
|
72 | |||
65 | class BlockBreakerTestCase(unittest.TestCase): |
|
73 | class BlockBreakerTestCase(unittest.TestCase): | |
66 | def setUp(self): |
|
74 | def setUp(self): | |
67 | self.bb = BB.BlockBreaker() |
|
75 | self.bb = BB.BlockBreaker() | |
68 |
|
76 | |||
69 | def test_reset(self): |
|
77 | def test_reset(self): | |
70 | bb = self.bb |
|
78 | bb = self.bb | |
71 | bb.push('x=1') |
|
79 | bb.push('x=1') | |
72 | bb.reset() |
|
80 | bb.reset() | |
73 | self.assertEqual(bb._buffer, []) |
|
81 | self.assertEqual(bb._buffer, []) | |
74 | self.assertEqual(bb.indent_spaces, 0) |
|
82 | self.assertEqual(bb.indent_spaces, 0) | |
75 | self.assertEqual(bb.source, '') |
|
83 | self.assertEqual(bb.source, '') | |
76 | self.assertEqual(bb.code, None) |
|
84 | self.assertEqual(bb.code, None) | |
|
85 | self.assertEqual(bb.is_complete, False) | |||
77 |
|
86 | |||
78 | def test_source(self): |
|
87 | def test_source(self): | |
79 | self.bb._store('1') |
|
88 | self.bb._store('1') | |
80 | self.bb._store('2') |
|
89 | self.bb._store('2') | |
81 | self.assertEqual(self.bb.source, '1\n2\n') |
|
90 | self.assertEqual(self.bb.source, '1\n2\n') | |
82 | self.assertTrue(len(self.bb._buffer)>0) |
|
91 | self.assertTrue(len(self.bb._buffer)>0) | |
83 | self.assertEqual(self.bb.source_reset(), '1\n2\n') |
|
92 | self.assertEqual(self.bb.source_reset(), '1\n2\n') | |
84 | self.assertEqual(self.bb._buffer, []) |
|
93 | self.assertEqual(self.bb._buffer, []) | |
85 | self.assertEqual(self.bb.source, '') |
|
94 | self.assertEqual(self.bb.source, '') | |
86 |
|
95 | |||
87 | def test_indent(self): |
|
96 | def test_indent(self): | |
88 | bb = self.bb # shorthand |
|
97 | bb = self.bb # shorthand | |
89 | bb.push('x=1') |
|
98 | bb.push('x=1') | |
90 | self.assertEqual(bb.indent_spaces, 0) |
|
99 | self.assertEqual(bb.indent_spaces, 0) | |
91 | bb.push('if 1:\n x=1') |
|
100 | bb.push('if 1:\n x=1') | |
92 | self.assertEqual(bb.indent_spaces, 4) |
|
101 | self.assertEqual(bb.indent_spaces, 4) | |
93 | bb.push('y=2\n') |
|
102 | bb.push('y=2\n') | |
94 | self.assertEqual(bb.indent_spaces, 0) |
|
103 | self.assertEqual(bb.indent_spaces, 0) | |
95 | bb.push('if 1:') |
|
104 | bb.push('if 1:') | |
96 | self.assertEqual(bb.indent_spaces, 4) |
|
105 | self.assertEqual(bb.indent_spaces, 4) | |
97 | bb.push(' x=1') |
|
106 | bb.push(' x=1') | |
98 | self.assertEqual(bb.indent_spaces, 4) |
|
107 | self.assertEqual(bb.indent_spaces, 4) | |
99 | # Blank lines shouldn't change the indent level |
|
108 | # Blank lines shouldn't change the indent level | |
100 | bb.push(' '*2) |
|
109 | bb.push(' '*2) | |
101 | self.assertEqual(bb.indent_spaces, 4) |
|
110 | self.assertEqual(bb.indent_spaces, 4) | |
102 |
|
111 | |||
103 | def test_indent2(self): |
|
112 | def test_indent2(self): | |
104 | bb = self.bb |
|
113 | bb = self.bb | |
105 | # When a multiline statement contains parens or multiline strings, we |
|
114 | # When a multiline statement contains parens or multiline strings, we | |
106 | # shouldn't get confused. |
|
115 | # shouldn't get confused. | |
107 | bb.push("if 1:") |
|
116 | bb.push("if 1:") | |
108 | bb.push(" x = (1+\n 2)") |
|
117 | bb.push(" x = (1+\n 2)") | |
109 | self.assertEqual(bb.indent_spaces, 4) |
|
118 | self.assertEqual(bb.indent_spaces, 4) | |
110 |
|
119 | |||
111 | def test_dedent(self): |
|
120 | def test_dedent(self): | |
112 | bb = self.bb # shorthand |
|
121 | bb = self.bb # shorthand | |
113 | bb.push('if 1:') |
|
122 | bb.push('if 1:') | |
114 | self.assertEqual(bb.indent_spaces, 4) |
|
123 | self.assertEqual(bb.indent_spaces, 4) | |
115 | bb.push(' pass') |
|
124 | bb.push(' pass') | |
116 | self.assertEqual(bb.indent_spaces, 0) |
|
125 | self.assertEqual(bb.indent_spaces, 0) | |
117 |
|
126 | |||
118 | def test_push(self): |
|
127 | def test_push(self): | |
119 | bb = self.bb |
|
128 | bb = self.bb | |
120 | bb.push('x=1') |
|
129 | bb.push('x=1') | |
121 | self.assertTrue(bb.is_complete) |
|
130 | self.assertTrue(bb.is_complete) | |
122 |
|
131 | |||
123 | def test_push2(self): |
|
132 | def test_push2(self): | |
124 | bb = self.bb |
|
133 | bb = self.bb | |
125 | bb.push('if 1:') |
|
134 | bb.push('if 1:') | |
126 | self.assertFalse(bb.is_complete) |
|
135 | self.assertFalse(bb.is_complete) | |
127 | for line in [' x=1', '# a comment', ' y=2']: |
|
136 | for line in [' x=1', '# a comment', ' y=2']: | |
128 | bb.push(line) |
|
137 | bb.push(line) | |
129 | self.assertTrue(bb.is_complete) |
|
138 | self.assertTrue(bb.is_complete) | |
130 |
|
139 | |||
131 | def test_push3(self): |
|
140 | def test_push3(self): | |
132 | """Test input with leading whitespace""" |
|
141 | """Test input with leading whitespace""" | |
133 | bb = self.bb |
|
142 | bb = self.bb | |
134 | bb.push(' x=1') |
|
143 | bb.push(' x=1') | |
135 | bb.push(' y=2') |
|
144 | bb.push(' y=2') | |
136 | self.assertEqual(bb.source, 'if 1:\n x=1\n y=2\n') |
|
145 | self.assertEqual(bb.source, 'if 1:\n x=1\n y=2\n') | |
137 |
|
146 | |||
138 | def test_replace_mode(self): |
|
147 | def test_replace_mode(self): | |
139 | bb = self.bb |
|
148 | bb = self.bb | |
140 | bb.input_mode = 'replace' |
|
149 | bb.input_mode = 'replace' | |
141 | bb.push('x=1') |
|
150 | bb.push('x=1') | |
142 | self.assertEqual(bb.source, 'x=1\n') |
|
151 | self.assertEqual(bb.source, 'x=1\n') | |
143 | bb.push('x=2') |
|
152 | bb.push('x=2') | |
144 | self.assertEqual(bb.source, 'x=2\n') |
|
153 | self.assertEqual(bb.source, 'x=2\n') | |
145 |
|
154 | |||
146 | def test_interactive_block_ready(self): |
|
155 | def test_interactive_block_ready(self): | |
147 | bb = self.bb |
|
156 | bb = self.bb | |
148 | bb.push('x=1') |
|
157 | bb.push('x=1') | |
149 | self.assertTrue(bb.interactive_block_ready()) |
|
158 | self.assertTrue(bb.interactive_block_ready()) | |
150 |
|
159 | |||
151 | def test_interactive_block_ready2(self): |
|
160 | def test_interactive_block_ready2(self): | |
152 | bb = self.bb |
|
161 | bb = self.bb | |
153 | bb.push('if 1:') |
|
162 | bb.push('if 1:') | |
154 | self.assertFalse(bb.interactive_block_ready()) |
|
163 | self.assertFalse(bb.interactive_block_ready()) | |
155 | bb.push(' x=1') |
|
164 | bb.push(' x=1') | |
156 | self.assertFalse(bb.interactive_block_ready()) |
|
165 | self.assertFalse(bb.interactive_block_ready()) | |
157 | bb.push('') |
|
166 | bb.push('') | |
158 | self.assertTrue(bb.interactive_block_ready()) |
|
167 | self.assertTrue(bb.interactive_block_ready()) | |
159 |
|
168 | |||
160 | def test_interactive_block_ready3(self): |
|
169 | def test_interactive_block_ready3(self): | |
161 | bb = self.bb |
|
170 | bb = self.bb | |
162 | bb.push("x = (2+\n3)") |
|
171 | bb.push("x = (2+\n3)") | |
163 | self.assertTrue(bb.interactive_block_ready()) |
|
172 | self.assertTrue(bb.interactive_block_ready()) | |
164 |
|
173 | |||
165 | def test_interactive_block_ready4(self): |
|
174 | def test_interactive_block_ready4(self): | |
166 | bb = self.bb |
|
175 | bb = self.bb | |
167 | # When a multiline statement contains parens or multiline strings, we |
|
176 | # When a multiline statement contains parens or multiline strings, we | |
168 | # shouldn't get confused. |
|
177 | # shouldn't get confused. | |
169 | # FIXME: we should be able to better handle de-dents in statements like |
|
178 | # FIXME: we should be able to better handle de-dents in statements like | |
170 | # multiline strings and multiline expressions (continued with \ or |
|
179 | # multiline strings and multiline expressions (continued with \ or | |
171 | # parens). Right now we aren't handling the indentation tracking quite |
|
180 | # parens). Right now we aren't handling the indentation tracking quite | |
172 | # correctly with this, though in practice it may not be too much of a |
|
181 | # correctly with this, though in practice it may not be too much of a | |
173 | # problem. We'll need to see. |
|
182 | # problem. We'll need to see. | |
174 | bb.push("if 1:") |
|
183 | bb.push("if 1:") | |
175 | bb.push(" x = (2+") |
|
184 | bb.push(" x = (2+") | |
176 | bb.push(" 3)") |
|
185 | bb.push(" 3)") | |
177 | self.assertFalse(bb.interactive_block_ready()) |
|
186 | self.assertFalse(bb.interactive_block_ready()) | |
178 | bb.push(" y = 3") |
|
187 | bb.push(" y = 3") | |
179 | self.assertFalse(bb.interactive_block_ready()) |
|
188 | self.assertFalse(bb.interactive_block_ready()) | |
180 | bb.push('') |
|
189 | bb.push('') | |
181 | self.assertTrue(bb.interactive_block_ready()) |
|
190 | self.assertTrue(bb.interactive_block_ready()) | |
182 |
|
191 | |||
183 | def test_syntax_error(self): |
|
192 | def test_syntax_error(self): | |
184 | bb = self.bb |
|
193 | bb = self.bb | |
185 | # Syntax errors immediately produce a 'ready' block, so the invalid |
|
194 | # Syntax errors immediately produce a 'ready' block, so the invalid | |
186 | # Python can be sent to the kernel for evaluation with possible ipython |
|
195 | # Python can be sent to the kernel for evaluation with possible ipython | |
187 | # special-syntax conversion. |
|
196 | # special-syntax conversion. | |
188 | bb.push('run foo') |
|
197 | bb.push('run foo') | |
189 | self.assertTrue(bb.interactive_block_ready()) |
|
198 | self.assertTrue(bb.interactive_block_ready()) | |
|
199 | ||||
|
200 | def check_split(self, block_lines, compile=True): | |||
|
201 | blocks = assemble(block_lines) | |||
|
202 | lines = ''.join(blocks) | |||
|
203 | oblock = self.bb.split_blocks(lines) | |||
|
204 | self.assertEqual(oblock, blocks) | |||
|
205 | if compile: | |||
|
206 | for block in blocks: | |||
|
207 | self.bb.compile(block) | |||
|
208 | ||||
|
209 | def test_split(self): | |||
|
210 | # All blocks of input we want to test in a list. The format for each | |||
|
211 | # block is a list of lists, with each inner lists consisting of all the | |||
|
212 | # lines (as single-lines) that should make up a sub-block. | |||
|
213 | ||||
|
214 | # Note: do NOT put here sub-blocks that don't compile, as the | |||
|
215 | # check_split() routine makes a final verification pass to check that | |||
|
216 | # each sub_block, as returned by split_blocks(), does compile | |||
|
217 | # correctly. | |||
|
218 | all_blocks = [ [['x=1']], | |||
|
219 | ||||
|
220 | [['x=1'], | |||
|
221 | ['y=2']], | |||
|
222 | ||||
|
223 | [['x=1'], | |||
|
224 | ['# a comment'], | |||
|
225 | ['y=11']], | |||
|
226 | ||||
|
227 | [['if 1:', | |||
|
228 | ' x=1'], | |||
|
229 | ['y=3']], | |||
|
230 | ||||
|
231 | [['def f(x):', | |||
|
232 | ' return x'], | |||
|
233 | ['x=1']], | |||
|
234 | ||||
|
235 | [['def f(x):', | |||
|
236 | ' x+=1', | |||
|
237 | ' ', | |||
|
238 | ' return x'], | |||
|
239 | ['x=1']], | |||
|
240 | ||||
|
241 | [['def f(x):', | |||
|
242 | ' if x>0:', | |||
|
243 | ' y=1', | |||
|
244 | ' # a comment', | |||
|
245 | ' else:', | |||
|
246 | ' y=4', | |||
|
247 | ' ', | |||
|
248 | ' return y'], | |||
|
249 | ['x=1'], | |||
|
250 | ['if 1:', | |||
|
251 | ' y=11'] ], | |||
|
252 | ||||
|
253 | [['for i in range(10):' | |||
|
254 | ' x=i**2']], | |||
|
255 | ||||
|
256 | [['for i in range(10):' | |||
|
257 | ' x=i**2'], | |||
|
258 | ['z = 1']], | |||
|
259 | ] | |||
|
260 | for block_lines in all_blocks: | |||
|
261 | self.check_split(block_lines) | |||
|
262 | ||||
|
263 | def test_split_syntax_errors(self): | |||
|
264 | # Block splitting with invalid syntax | |||
|
265 | all_blocks = [ [['a syntax error']], | |||
|
266 | ||||
|
267 | [['x=1'], | |||
|
268 | ['a syntax error']], | |||
|
269 | ||||
|
270 | [['for i in range(10):' | |||
|
271 | ' an error']], | |||
|
272 | ||||
|
273 | ] | |||
|
274 | for block_lines in all_blocks: | |||
|
275 | self.check_split(block_lines, compile=False) | |||
|
276 |
General Comments 0
You need to be logged in to leave comments.
Login now