Show More
@@ -1,215 +1,216 b'' | |||
|
1 | 1 | """The basic dict based notebook format. |
|
2 | 2 | |
|
3 | 3 | The Python representation of a notebook is a nested structure of |
|
4 | 4 | dictionary subclasses that support attribute access |
|
5 | 5 | (IPython.utils.ipstruct.Struct). The functions in this module are merely |
|
6 | 6 | helpers to build the structs in the right form. |
|
7 | 7 | |
|
8 | 8 | Authors: |
|
9 | 9 | |
|
10 | 10 | * Brian Granger |
|
11 | 11 | """ |
|
12 | 12 | |
|
13 | 13 | #----------------------------------------------------------------------------- |
|
14 | 14 | # Copyright (C) 2008-2011 The IPython Development Team |
|
15 | 15 | # |
|
16 | 16 | # Distributed under the terms of the BSD License. The full license is in |
|
17 | 17 | # the file COPYING, distributed as part of this software. |
|
18 | 18 | #----------------------------------------------------------------------------- |
|
19 | 19 | |
|
20 | 20 | #----------------------------------------------------------------------------- |
|
21 | 21 | # Imports |
|
22 | 22 | #----------------------------------------------------------------------------- |
|
23 | 23 | |
|
24 | 24 | import pprint |
|
25 | 25 | import uuid |
|
26 | 26 | |
|
27 | 27 | from IPython.utils.ipstruct import Struct |
|
28 | from IPython.utils.py3compat import cast_unicode | |
|
28 | 29 | |
|
29 | 30 | #----------------------------------------------------------------------------- |
|
30 | 31 | # Code |
|
31 | 32 | #----------------------------------------------------------------------------- |
|
32 | 33 | |
|
33 | 34 | # Change this when incrementing the nbformat version |
|
34 | 35 | nbformat = 3 |
|
35 | 36 | nbformat_minor = 0 |
|
36 | 37 | |
|
37 | 38 | class NotebookNode(Struct): |
|
38 | 39 | pass |
|
39 | 40 | |
|
40 | 41 | |
|
41 | 42 | def from_dict(d): |
|
42 | 43 | if isinstance(d, dict): |
|
43 | 44 | newd = NotebookNode() |
|
44 | 45 | for k,v in d.items(): |
|
45 | 46 | newd[k] = from_dict(v) |
|
46 | 47 | return newd |
|
47 | 48 | elif isinstance(d, (tuple, list)): |
|
48 | 49 | return [from_dict(i) for i in d] |
|
49 | 50 | else: |
|
50 | 51 | return d |
|
51 | 52 | |
|
52 | 53 | |
|
53 | 54 | def new_output(output_type=None, output_text=None, output_png=None, |
|
54 | 55 | output_html=None, output_svg=None, output_latex=None, output_json=None, |
|
55 | 56 | output_javascript=None, output_jpeg=None, prompt_number=None, |
|
56 | 57 | ename=None, evalue=None, traceback=None, stream=None, metadata=None): |
|
57 | 58 | """Create a new code cell with input and output""" |
|
58 | 59 | output = NotebookNode() |
|
59 | 60 | if output_type is not None: |
|
60 | 61 | output.output_type = unicode(output_type) |
|
61 | 62 | |
|
62 | 63 | if metadata is None: |
|
63 | 64 | metadata = {} |
|
64 | 65 | if not isinstance(metadata, dict): |
|
65 | 66 | raise TypeError("metadata must be dict") |
|
66 | 67 | output.metadata = metadata |
|
67 | 68 | |
|
68 | 69 | if output_type != 'pyerr': |
|
69 | 70 | if output_text is not None: |
|
70 | output.text = unicode(output_text) | |
|
71 | output.text = cast_unicode(output_text) | |
|
71 | 72 | if output_png is not None: |
|
72 | output.png = unicode(output_png) | |
|
73 | output.png = cast_unicode(output_png) | |
|
73 | 74 | if output_jpeg is not None: |
|
74 | output.jpeg = unicode(output_jpeg) | |
|
75 | output.jpeg = cast_unicode(output_jpeg) | |
|
75 | 76 | if output_html is not None: |
|
76 | output.html = unicode(output_html) | |
|
77 | output.html = cast_unicode(output_html) | |
|
77 | 78 | if output_svg is not None: |
|
78 | output.svg = unicode(output_svg) | |
|
79 | output.svg = cast_unicode(output_svg) | |
|
79 | 80 | if output_latex is not None: |
|
80 | output.latex = unicode(output_latex) | |
|
81 | output.latex = cast_unicode(output_latex) | |
|
81 | 82 | if output_json is not None: |
|
82 | output.json = unicode(output_json) | |
|
83 | output.json = cast_unicode(output_json) | |
|
83 | 84 | if output_javascript is not None: |
|
84 | output.javascript = unicode(output_javascript) | |
|
85 | output.javascript = cast_unicode(output_javascript) | |
|
85 | 86 | |
|
86 | 87 | if output_type == u'pyout': |
|
87 | 88 | if prompt_number is not None: |
|
88 | 89 | output.prompt_number = int(prompt_number) |
|
89 | 90 | |
|
90 | 91 | if output_type == u'pyerr': |
|
91 | 92 | if ename is not None: |
|
92 | output.ename = unicode(ename) | |
|
93 | output.ename = cast_unicode(ename) | |
|
93 | 94 | if evalue is not None: |
|
94 | output.evalue = unicode(evalue) | |
|
95 | output.evalue = cast_unicode(evalue) | |
|
95 | 96 | if traceback is not None: |
|
96 | output.traceback = [unicode(frame) for frame in list(traceback)] | |
|
97 | output.traceback = [cast_unicode(frame) for frame in list(traceback)] | |
|
97 | 98 | |
|
98 | 99 | if output_type == u'stream': |
|
99 | output.stream = 'stdout' if stream is None else unicode(stream) | |
|
100 | output.stream = 'stdout' if stream is None else cast_unicode(stream) | |
|
100 | 101 | |
|
101 | 102 | return output |
|
102 | 103 | |
|
103 | 104 | |
|
104 | 105 | def new_code_cell(input=None, prompt_number=None, outputs=None, |
|
105 | 106 | language=u'python', collapsed=False, metadata=None): |
|
106 | 107 | """Create a new code cell with input and output""" |
|
107 | 108 | cell = NotebookNode() |
|
108 | 109 | cell.cell_type = u'code' |
|
109 | 110 | if language is not None: |
|
110 | cell.language = unicode(language) | |
|
111 | cell.language = cast_unicode(language) | |
|
111 | 112 | if input is not None: |
|
112 | cell.input = unicode(input) | |
|
113 | cell.input = cast_unicode(input) | |
|
113 | 114 | if prompt_number is not None: |
|
114 | 115 | cell.prompt_number = int(prompt_number) |
|
115 | 116 | if outputs is None: |
|
116 | 117 | cell.outputs = [] |
|
117 | 118 | else: |
|
118 | 119 | cell.outputs = outputs |
|
119 | 120 | if collapsed is not None: |
|
120 | 121 | cell.collapsed = bool(collapsed) |
|
121 | 122 | cell.metadata = NotebookNode(metadata or {}) |
|
122 | 123 | |
|
123 | 124 | return cell |
|
124 | 125 | |
|
125 | 126 | def new_text_cell(cell_type, source=None, rendered=None, metadata=None): |
|
126 | 127 | """Create a new text cell.""" |
|
127 | 128 | cell = NotebookNode() |
|
128 | 129 | # VERSIONHACK: plaintext -> raw |
|
129 | 130 | # handle never-released plaintext name for raw cells |
|
130 | 131 | if cell_type == 'plaintext': |
|
131 | 132 | cell_type = 'raw' |
|
132 | 133 | if source is not None: |
|
133 | cell.source = unicode(source) | |
|
134 | cell.source = cast_unicode(source) | |
|
134 | 135 | if rendered is not None: |
|
135 | cell.rendered = unicode(rendered) | |
|
136 | cell.rendered = cast_unicode(rendered) | |
|
136 | 137 | cell.metadata = NotebookNode(metadata or {}) |
|
137 | 138 | cell.cell_type = cell_type |
|
138 | 139 | return cell |
|
139 | 140 | |
|
140 | 141 | |
|
141 | 142 | def new_heading_cell(source=None, rendered=None, level=1, metadata=None): |
|
142 | 143 | """Create a new section cell with a given integer level.""" |
|
143 | 144 | cell = NotebookNode() |
|
144 | 145 | cell.cell_type = u'heading' |
|
145 | 146 | if source is not None: |
|
146 | cell.source = unicode(source) | |
|
147 | cell.source = cast_unicode(source) | |
|
147 | 148 | if rendered is not None: |
|
148 | cell.rendered = unicode(rendered) | |
|
149 | cell.rendered = cast_unicode(rendered) | |
|
149 | 150 | cell.level = int(level) |
|
150 | 151 | cell.metadata = NotebookNode(metadata or {}) |
|
151 | 152 | return cell |
|
152 | 153 | |
|
153 | 154 | |
|
154 | 155 | def new_worksheet(name=None, cells=None, metadata=None): |
|
155 | 156 | """Create a worksheet by name with with a list of cells.""" |
|
156 | 157 | ws = NotebookNode() |
|
157 | 158 | if name is not None: |
|
158 | ws.name = unicode(name) | |
|
159 | ws.name = cast_unicode(name) | |
|
159 | 160 | if cells is None: |
|
160 | 161 | ws.cells = [] |
|
161 | 162 | else: |
|
162 | 163 | ws.cells = list(cells) |
|
163 | 164 | ws.metadata = NotebookNode(metadata or {}) |
|
164 | 165 | return ws |
|
165 | 166 | |
|
166 | 167 | |
|
167 | 168 | def new_notebook(name=None, metadata=None, worksheets=None): |
|
168 | 169 | """Create a notebook by name, id and a list of worksheets.""" |
|
169 | 170 | nb = NotebookNode() |
|
170 | 171 | nb.nbformat = nbformat |
|
171 | 172 | nb.nbformat_minor = nbformat_minor |
|
172 | 173 | if worksheets is None: |
|
173 | 174 | nb.worksheets = [] |
|
174 | 175 | else: |
|
175 | 176 | nb.worksheets = list(worksheets) |
|
176 | 177 | if metadata is None: |
|
177 | 178 | nb.metadata = new_metadata() |
|
178 | 179 | else: |
|
179 | 180 | nb.metadata = NotebookNode(metadata) |
|
180 | 181 | if name is not None: |
|
181 | nb.metadata.name = unicode(name) | |
|
182 | nb.metadata.name = cast_unicode(name) | |
|
182 | 183 | return nb |
|
183 | 184 | |
|
184 | 185 | |
|
185 | 186 | def new_metadata(name=None, authors=None, license=None, created=None, |
|
186 | 187 | modified=None, gistid=None): |
|
187 | 188 | """Create a new metadata node.""" |
|
188 | 189 | metadata = NotebookNode() |
|
189 | 190 | if name is not None: |
|
190 | metadata.name = unicode(name) | |
|
191 | metadata.name = cast_unicode(name) | |
|
191 | 192 | if authors is not None: |
|
192 | 193 | metadata.authors = list(authors) |
|
193 | 194 | if created is not None: |
|
194 | metadata.created = unicode(created) | |
|
195 | metadata.created = cast_unicode(created) | |
|
195 | 196 | if modified is not None: |
|
196 | metadata.modified = unicode(modified) | |
|
197 | metadata.modified = cast_unicode(modified) | |
|
197 | 198 | if license is not None: |
|
198 | metadata.license = unicode(license) | |
|
199 | metadata.license = cast_unicode(license) | |
|
199 | 200 | if gistid is not None: |
|
200 | metadata.gistid = unicode(gistid) | |
|
201 | metadata.gistid = cast_unicode(gistid) | |
|
201 | 202 | return metadata |
|
202 | 203 | |
|
203 | 204 | def new_author(name=None, email=None, affiliation=None, url=None): |
|
204 | 205 | """Create a new author.""" |
|
205 | 206 | author = NotebookNode() |
|
206 | 207 | if name is not None: |
|
207 | author.name = unicode(name) | |
|
208 | author.name = cast_unicode(name) | |
|
208 | 209 | if email is not None: |
|
209 | author.email = unicode(email) | |
|
210 | author.email = cast_unicode(email) | |
|
210 | 211 | if affiliation is not None: |
|
211 | author.affiliation = unicode(affiliation) | |
|
212 | author.affiliation = cast_unicode(affiliation) | |
|
212 | 213 | if url is not None: |
|
213 | author.url = unicode(url) | |
|
214 | author.url = cast_unicode(url) | |
|
214 | 215 | return author |
|
215 | 216 |
@@ -1,204 +1,204 b'' | |||
|
1 | 1 | """Read and write notebooks as regular .py files. |
|
2 | 2 | |
|
3 | 3 | Authors: |
|
4 | 4 | |
|
5 | 5 | * Brian Granger |
|
6 | 6 | """ |
|
7 | 7 | |
|
8 | 8 | #----------------------------------------------------------------------------- |
|
9 | 9 | # Copyright (C) 2008-2011 The IPython Development Team |
|
10 | 10 | # |
|
11 | 11 | # Distributed under the terms of the BSD License. The full license is in |
|
12 | 12 | # the file COPYING, distributed as part of this software. |
|
13 | 13 | #----------------------------------------------------------------------------- |
|
14 | 14 | |
|
15 | 15 | #----------------------------------------------------------------------------- |
|
16 | 16 | # Imports |
|
17 | 17 | #----------------------------------------------------------------------------- |
|
18 | 18 | |
|
19 | 19 | import re |
|
20 | 20 | from .rwbase import NotebookReader, NotebookWriter |
|
21 | 21 | from .nbbase import ( |
|
22 | 22 | new_code_cell, new_text_cell, new_worksheet, |
|
23 | 23 | new_notebook, new_heading_cell, nbformat, nbformat_minor, |
|
24 | 24 | ) |
|
25 | 25 | |
|
26 | 26 | #----------------------------------------------------------------------------- |
|
27 | 27 | # Code |
|
28 | 28 | #----------------------------------------------------------------------------- |
|
29 | 29 | |
|
30 | 30 | _encoding_declaration_re = re.compile(r"^#.*coding[:=]\s*([-\w.]+)") |
|
31 | 31 | |
|
32 | 32 | class PyReaderError(Exception): |
|
33 | 33 | pass |
|
34 | 34 | |
|
35 | 35 | |
|
36 | 36 | class PyReader(NotebookReader): |
|
37 | 37 | |
|
38 | 38 | def reads(self, s, **kwargs): |
|
39 | 39 | return self.to_notebook(s,**kwargs) |
|
40 | 40 | |
|
41 | 41 | def to_notebook(self, s, **kwargs): |
|
42 | 42 | lines = s.splitlines() |
|
43 | 43 | cells = [] |
|
44 | 44 | cell_lines = [] |
|
45 | 45 | kwargs = {} |
|
46 | 46 | state = u'codecell' |
|
47 | 47 | for line in lines: |
|
48 | 48 | if line.startswith(u'# <nbformat>') or _encoding_declaration_re.match(line): |
|
49 | 49 | pass |
|
50 | 50 | elif line.startswith(u'# <codecell>'): |
|
51 | 51 | cell = self.new_cell(state, cell_lines, **kwargs) |
|
52 | 52 | if cell is not None: |
|
53 | 53 | cells.append(cell) |
|
54 | 54 | state = u'codecell' |
|
55 | 55 | cell_lines = [] |
|
56 | 56 | kwargs = {} |
|
57 | 57 | elif line.startswith(u'# <htmlcell>'): |
|
58 | 58 | cell = self.new_cell(state, cell_lines, **kwargs) |
|
59 | 59 | if cell is not None: |
|
60 | 60 | cells.append(cell) |
|
61 | 61 | state = u'htmlcell' |
|
62 | 62 | cell_lines = [] |
|
63 | 63 | kwargs = {} |
|
64 | 64 | elif line.startswith(u'# <markdowncell>'): |
|
65 | 65 | cell = self.new_cell(state, cell_lines, **kwargs) |
|
66 | 66 | if cell is not None: |
|
67 | 67 | cells.append(cell) |
|
68 | 68 | state = u'markdowncell' |
|
69 | 69 | cell_lines = [] |
|
70 | 70 | kwargs = {} |
|
71 | 71 | # VERSIONHACK: plaintext -> raw |
|
72 | 72 | elif line.startswith(u'# <rawcell>') or line.startswith(u'# <plaintextcell>'): |
|
73 | 73 | cell = self.new_cell(state, cell_lines, **kwargs) |
|
74 | 74 | if cell is not None: |
|
75 | 75 | cells.append(cell) |
|
76 | 76 | state = u'rawcell' |
|
77 | 77 | cell_lines = [] |
|
78 | 78 | kwargs = {} |
|
79 | 79 | elif line.startswith(u'# <headingcell'): |
|
80 | 80 | cell = self.new_cell(state, cell_lines, **kwargs) |
|
81 | 81 | if cell is not None: |
|
82 | 82 | cells.append(cell) |
|
83 | 83 | cell_lines = [] |
|
84 | 84 | m = re.match(r'# <headingcell level=(?P<level>\d)>',line) |
|
85 | 85 | if m is not None: |
|
86 | 86 | state = u'headingcell' |
|
87 | 87 | kwargs = {} |
|
88 | 88 | kwargs['level'] = int(m.group('level')) |
|
89 | 89 | else: |
|
90 | 90 | state = u'codecell' |
|
91 | 91 | kwargs = {} |
|
92 | 92 | cell_lines = [] |
|
93 | 93 | else: |
|
94 | 94 | cell_lines.append(line) |
|
95 | 95 | if cell_lines and state == u'codecell': |
|
96 | 96 | cell = self.new_cell(state, cell_lines) |
|
97 | 97 | if cell is not None: |
|
98 | 98 | cells.append(cell) |
|
99 | 99 | ws = new_worksheet(cells=cells) |
|
100 | 100 | nb = new_notebook(worksheets=[ws]) |
|
101 | 101 | return nb |
|
102 | 102 | |
|
103 | 103 | def new_cell(self, state, lines, **kwargs): |
|
104 | 104 | if state == u'codecell': |
|
105 | 105 | input = u'\n'.join(lines) |
|
106 | 106 | input = input.strip(u'\n') |
|
107 | 107 | if input: |
|
108 | 108 | return new_code_cell(input=input) |
|
109 | 109 | elif state == u'htmlcell': |
|
110 | 110 | text = self._remove_comments(lines) |
|
111 | 111 | if text: |
|
112 | 112 | return new_text_cell(u'html',source=text) |
|
113 | 113 | elif state == u'markdowncell': |
|
114 | 114 | text = self._remove_comments(lines) |
|
115 | 115 | if text: |
|
116 | 116 | return new_text_cell(u'markdown',source=text) |
|
117 | 117 | elif state == u'rawcell': |
|
118 | 118 | text = self._remove_comments(lines) |
|
119 | 119 | if text: |
|
120 | 120 | return new_text_cell(u'raw',source=text) |
|
121 | 121 | elif state == u'headingcell': |
|
122 | 122 | text = self._remove_comments(lines) |
|
123 | 123 | level = kwargs.get('level',1) |
|
124 | 124 | if text: |
|
125 | 125 | return new_heading_cell(source=text,level=level) |
|
126 | 126 | |
|
127 | 127 | def _remove_comments(self, lines): |
|
128 | 128 | new_lines = [] |
|
129 | 129 | for line in lines: |
|
130 | 130 | if line.startswith(u'#'): |
|
131 | 131 | new_lines.append(line[2:]) |
|
132 | 132 | else: |
|
133 | 133 | new_lines.append(line) |
|
134 | 134 | text = u'\n'.join(new_lines) |
|
135 | 135 | text = text.strip(u'\n') |
|
136 | 136 | return text |
|
137 | 137 | |
|
138 | 138 | def split_lines_into_blocks(self, lines): |
|
139 | 139 | if len(lines) == 1: |
|
140 | 140 | yield lines[0] |
|
141 | 141 | raise StopIteration() |
|
142 | 142 | import ast |
|
143 | 143 | source = '\n'.join(lines) |
|
144 | 144 | code = ast.parse(source) |
|
145 | 145 | starts = [x.lineno-1 for x in code.body] |
|
146 | 146 | for i in range(len(starts)-1): |
|
147 | 147 | yield '\n'.join(lines[starts[i]:starts[i+1]]).strip('\n') |
|
148 | 148 | yield '\n'.join(lines[starts[-1]:]).strip('\n') |
|
149 | 149 | |
|
150 | 150 | |
|
151 | 151 | class PyWriter(NotebookWriter): |
|
152 | 152 | |
|
153 | 153 | def writes(self, nb, **kwargs): |
|
154 | 154 | lines = [u'# -*- coding: utf-8 -*-'] |
|
155 | 155 | lines.extend([ |
|
156 | 156 | u'# <nbformat>%i.%i</nbformat>' % (nbformat, nbformat_minor), |
|
157 | 157 | u'', |
|
158 | 158 | ]) |
|
159 | 159 | for ws in nb.worksheets: |
|
160 | 160 | for cell in ws.cells: |
|
161 | 161 | if cell.cell_type == u'code': |
|
162 | 162 | input = cell.get(u'input') |
|
163 | 163 | if input is not None: |
|
164 | 164 | lines.extend([u'# <codecell>',u'']) |
|
165 | 165 | lines.extend(input.splitlines()) |
|
166 | 166 | lines.append(u'') |
|
167 | 167 | elif cell.cell_type == u'html': |
|
168 | 168 | input = cell.get(u'source') |
|
169 | 169 | if input is not None: |
|
170 | 170 | lines.extend([u'# <htmlcell>',u'']) |
|
171 | 171 | lines.extend([u'# ' + line for line in input.splitlines()]) |
|
172 | 172 | lines.append(u'') |
|
173 | 173 | elif cell.cell_type == u'markdown': |
|
174 | 174 | input = cell.get(u'source') |
|
175 | 175 | if input is not None: |
|
176 | 176 | lines.extend([u'# <markdowncell>',u'']) |
|
177 | 177 | lines.extend([u'# ' + line for line in input.splitlines()]) |
|
178 | 178 | lines.append(u'') |
|
179 | 179 | elif cell.cell_type == u'raw': |
|
180 | 180 | input = cell.get(u'source') |
|
181 | 181 | if input is not None: |
|
182 | 182 | lines.extend([u'# <rawcell>',u'']) |
|
183 | 183 | lines.extend([u'# ' + line for line in input.splitlines()]) |
|
184 | 184 | lines.append(u'') |
|
185 | 185 | elif cell.cell_type == u'heading': |
|
186 | 186 | input = cell.get(u'source') |
|
187 | 187 | level = cell.get(u'level',1) |
|
188 | 188 | if input is not None: |
|
189 | 189 | lines.extend([u'# <headingcell level=%s>' % level,u'']) |
|
190 | 190 | lines.extend([u'# ' + line for line in input.splitlines()]) |
|
191 | 191 | lines.append(u'') |
|
192 | 192 | lines.append('') |
|
193 |
return u |
|
|
193 | return u'\n'.join(lines) | |
|
194 | 194 | |
|
195 | 195 | |
|
196 | 196 | _reader = PyReader() |
|
197 | 197 | _writer = PyWriter() |
|
198 | 198 | |
|
199 | 199 | reads = _reader.reads |
|
200 | 200 | read = _reader.read |
|
201 | 201 | to_notebook = _reader.to_notebook |
|
202 | 202 | write = _writer.write |
|
203 | 203 | writes = _writer.writes |
|
204 | 204 |
General Comments 0
You need to be logged in to leave comments.
Login now