Show More
@@ -1,495 +1,493 b'' | |||
|
1 | 1 | """Various display related classes. |
|
2 | 2 | |
|
3 | 3 | Authors : MinRK, gregcaporaso, dannystaple |
|
4 | 4 | """ |
|
5 | 5 | from os.path import exists, isfile, splitext, abspath, join, isdir |
|
6 | 6 | from os import walk, sep |
|
7 | 7 | |
|
8 | 8 | from IPython.core.display import DisplayObject |
|
9 | 9 | |
|
10 | 10 | |
|
11 | 11 | class Audio(DisplayObject): |
|
12 | 12 | """Create an audio object. |
|
13 | 13 | |
|
14 | 14 | When this object is returned by an input cell or passed to the |
|
15 | 15 | display function, it will result in Audio controls being displayed |
|
16 | 16 | in the frontend (only works in the notebook). |
|
17 | 17 | |
|
18 | 18 | Parameters |
|
19 | 19 | ---------- |
|
20 | 20 | data : numpy array, list, unicode, str or bytes |
|
21 | 21 | Can be a |
|
22 | 22 | * Numpy 1d array containing the desired waveform (mono) |
|
23 | 23 | * List of float or integer representing the waveform (mono) |
|
24 | 24 | * String containing the filename |
|
25 | 25 | * Bytestring containing raw PCM data or |
|
26 | 26 | * URL pointing to a file on the web. |
|
27 | 27 | |
|
28 | 28 | If the array option is used the waveform will be normalized. |
|
29 | 29 | |
|
30 | 30 | If a filename or url is used the format support will be browser |
|
31 | 31 | dependent. |
|
32 | 32 | url : unicode |
|
33 | 33 | A URL to download the data from. |
|
34 | 34 | filename : unicode |
|
35 | 35 | Path to a local file to load the data from. |
|
36 | 36 | embed : boolean |
|
37 | 37 | Should the image data be embedded using a data URI (True) or should |
|
38 | 38 | the original source be referenced. Set this to True if you want the |
|
39 | 39 | audio to playable later with no internet connection in the notebook. |
|
40 | 40 | |
|
41 | 41 | Default is `True`, unless the keyword argument `url` is set, then |
|
42 | 42 | default value is `False`. |
|
43 | 43 | rate : integer |
|
44 | 44 | The sampling rate of the raw data. |
|
45 | 45 | Only required when data parameter is being used as an array |
|
46 | 46 | autoplay : bool |
|
47 | 47 | Set to True if the audio should immediately start playing. |
|
48 | 48 | Default is `False`. |
|
49 | 49 | |
|
50 | 50 | Examples |
|
51 | 51 | -------- |
|
52 | 52 | |
|
53 | 53 | # Generate a sound |
|
54 | 54 | import numpy as np |
|
55 | 55 | framerate = 44100 |
|
56 | 56 | t = np.linspace(0,5,framerate*5) |
|
57 | 57 | data = np.sin(2*np.pi*220*t) + np.sin(2*np.pi*224*t)) |
|
58 | 58 | Audio(data,rate=framerate) |
|
59 | 59 | |
|
60 | 60 | Audio("http://www.nch.com.au/acm/8k16bitpcm.wav") |
|
61 | 61 | Audio(url="http://www.w3schools.com/html/horse.ogg") |
|
62 | 62 | |
|
63 | 63 | Audio('/path/to/sound.wav') |
|
64 | 64 | Audio(filename='/path/to/sound.ogg') |
|
65 | 65 | |
|
66 | 66 | Audio(b'RAW_WAV_DATA..) |
|
67 | 67 | Audio(data=b'RAW_WAV_DATA..) |
|
68 | 68 | |
|
69 | 69 | """ |
|
70 | 70 | |
|
71 | 71 | def __init__(self, data=None, filename=None, url=None, embed=None, rate=None, autoplay=False): |
|
72 | 72 | if filename is None and url is None and data is None: |
|
73 | 73 | raise ValueError("No image data found. Expecting filename, url, or data.") |
|
74 | 74 | if embed is False and url is None: |
|
75 | 75 | raise ValueError("No url found. Expecting url when embed=False") |
|
76 | 76 | |
|
77 | 77 | if url is not None and embed is not True: |
|
78 | 78 | self.embed = False |
|
79 | 79 | else: |
|
80 | 80 | self.embed = True |
|
81 | 81 | self.autoplay = autoplay |
|
82 | 82 | super(Audio, self).__init__(data=data, url=url, filename=filename) |
|
83 | 83 | |
|
84 | 84 | if self.data is not None and not isinstance(self.data, bytes): |
|
85 | 85 | self.data = self._make_wav(data,rate) |
|
86 | 86 | |
|
87 | 87 | def reload(self): |
|
88 | 88 | """Reload the raw data from file or URL.""" |
|
89 | 89 | import mimetypes |
|
90 | 90 | if self.embed: |
|
91 | 91 | super(Audio, self).reload() |
|
92 | 92 | |
|
93 | 93 | if self.filename is not None: |
|
94 | 94 | self.mimetype = mimetypes.guess_type(self.filename)[0] |
|
95 | 95 | elif self.url is not None: |
|
96 | 96 | self.mimetype = mimetypes.guess_type(self.url)[0] |
|
97 | 97 | else: |
|
98 | 98 | self.mimetype = "audio/wav" |
|
99 | 99 | |
|
100 | 100 | def _make_wav(self, data, rate): |
|
101 | 101 | """ Transform a numpy array to a PCM bytestring """ |
|
102 | 102 | import struct |
|
103 | 103 | from io import BytesIO |
|
104 | 104 | import wave |
|
105 | 105 | try: |
|
106 | 106 | import numpy as np |
|
107 | 107 | data = np.array(data,dtype=float) |
|
108 | 108 | if len(data.shape) > 1: |
|
109 | 109 | raise ValueError("encoding of stereo PCM signals are unsupported") |
|
110 | 110 | scaled = np.int16(data/np.max(np.abs(data))*32767).tolist() |
|
111 | 111 | except ImportError: |
|
112 | 112 | maxabsvalue = float(max([abs(x) for x in data])) |
|
113 | 113 | scaled = [int(x/maxabsvalue*32767) for x in data] |
|
114 | 114 | fp = BytesIO() |
|
115 | 115 | waveobj = wave.open(fp,mode='wb') |
|
116 | 116 | waveobj.setnchannels(1) |
|
117 | 117 | waveobj.setframerate(rate) |
|
118 | 118 | waveobj.setsampwidth(2) |
|
119 | 119 | waveobj.setcomptype('NONE','NONE') |
|
120 | 120 | waveobj.writeframes(b''.join([struct.pack('<h',x) for x in scaled])) |
|
121 | 121 | val = fp.getvalue() |
|
122 | 122 | waveobj.close() |
|
123 | 123 | return val |
|
124 | 124 | |
|
125 | 125 | def _data_and_metadata(self): |
|
126 | 126 | """shortcut for returning metadata with url information, if defined""" |
|
127 | 127 | md = {} |
|
128 | 128 | if self.url: |
|
129 | 129 | md['url'] = self.url |
|
130 | 130 | if md: |
|
131 | 131 | return self.data, md |
|
132 | 132 | else: |
|
133 | 133 | return self.data |
|
134 | 134 | |
|
135 | 135 | def _repr_html_(self): |
|
136 | 136 | src = """ |
|
137 | 137 | <audio controls="controls" {autoplay}> |
|
138 | 138 | <source src="{src}" type="{type}" /> |
|
139 | 139 | Your browser does not support the audio element. |
|
140 | 140 | </audio> |
|
141 | 141 | """ |
|
142 | 142 | return src.format(src=self.src_attr(),type=self.mimetype, autoplay=self.autoplay_attr()) |
|
143 | 143 | |
|
144 | 144 | def src_attr(self): |
|
145 | 145 | import base64 |
|
146 | 146 | if self.embed and (self.data is not None): |
|
147 | 147 | data = base64=base64.b64encode(self.data).decode('ascii') |
|
148 | 148 | return """data:{type};base64,{base64}""".format(type=self.mimetype, |
|
149 | 149 | base64=data) |
|
150 | 150 | elif self.url is not None: |
|
151 | 151 | return self.url |
|
152 | 152 | else: |
|
153 | 153 | return "" |
|
154 | 154 | |
|
155 | 155 | def autoplay_attr(self): |
|
156 | 156 | if(self.autoplay): |
|
157 | 157 | return 'autoplay="autoplay"' |
|
158 | 158 | else: |
|
159 | 159 | return '' |
|
160 | 160 | |
|
161 | 161 | class IFrame(object): |
|
162 | 162 | """ |
|
163 | 163 | Generic class to embed an iframe in an IPython notebook |
|
164 | 164 | """ |
|
165 | 165 | |
|
166 | 166 | iframe = """ |
|
167 | 167 | <iframe |
|
168 | 168 | width="{width}" |
|
169 | 169 | height={height}" |
|
170 | 170 | src="{src}{params}" |
|
171 | 171 | frameborder="0" |
|
172 | 172 | allowfullscreen |
|
173 | 173 | ></iframe> |
|
174 | 174 | """ |
|
175 | 175 | |
|
176 | 176 | def __init__(self, src, width, height, **kwargs): |
|
177 | 177 | self.src = src |
|
178 | 178 | self.width = width |
|
179 | 179 | self.height = height |
|
180 | 180 | self.params = kwargs |
|
181 | 181 | |
|
182 | 182 | def _repr_html_(self): |
|
183 | 183 | """return the embed iframe""" |
|
184 | 184 | if self.params: |
|
185 | 185 | from urllib import urlencode |
|
186 | 186 | params = "?" + urlencode(self.params) |
|
187 | 187 | else: |
|
188 | 188 | params = "" |
|
189 | 189 | return self.iframe.format(src=self.src, |
|
190 | 190 | width=self.width, |
|
191 | 191 | height=self.height, |
|
192 | 192 | params=params) |
|
193 | 193 | |
|
194 | 194 | class YouTubeVideo(IFrame): |
|
195 | 195 | """Class for embedding a YouTube Video in an IPython session, based on its video id. |
|
196 | 196 | |
|
197 | 197 | e.g. to embed the video on this page: |
|
198 | 198 | |
|
199 | 199 | http://www.youtube.com/watch?v=foo |
|
200 | 200 | |
|
201 | 201 | you would do: |
|
202 | 202 | |
|
203 | 203 | vid = YouTubeVideo("foo") |
|
204 | 204 | display(vid) |
|
205 | 205 | |
|
206 | 206 | To start from 30 seconds: |
|
207 | 207 | |
|
208 | 208 | vid = YouTubeVideo("abc", start=30) |
|
209 | 209 | display(vid) |
|
210 | 210 | |
|
211 | 211 | To calculate seconds from time as hours, minutes, seconds use: |
|
212 | 212 | start=int(timedelta(hours=1, minutes=46, seconds=40).total_seconds()) |
|
213 | 213 | |
|
214 | 214 | Other parameters can be provided as documented at |
|
215 | 215 | https://developers.google.com/youtube/player_parameters#parameter-subheader |
|
216 | 216 | """ |
|
217 | 217 | |
|
218 | 218 | def __init__(self, id, width=400, height=300, **kwargs): |
|
219 | 219 | src = "http://www.youtube.com/embed/{0}".format(id) |
|
220 | 220 | super(YouTubeVideo, self).__init__(src, width, height, **kwargs) |
|
221 | 221 | |
|
222 | 222 | class VimeoVideo(IFrame): |
|
223 | 223 | """ |
|
224 | 224 | Class for embedding a Vimeo video in an IPython session, based on its video id. |
|
225 | 225 | """ |
|
226 | 226 | |
|
227 | 227 | def __init__(self, id, width=400, height=300, **kwargs): |
|
228 | 228 | src="http://player.vimeo.com/video/{0}".format(id) |
|
229 | 229 | super(VimeoVideo, self).__init__(src, width, height, **kwargs) |
|
230 | 230 | |
|
231 | 231 | class ScribdDocument(IFrame): |
|
232 | 232 | """ |
|
233 | 233 | Class for embedding a Scribd document in an IPython session |
|
234 | 234 | |
|
235 | 235 | Use the start_page params to specify a starting point in the document |
|
236 | 236 | Use the view_mode params to specify display type one off scroll | slideshow | book |
|
237 | 237 | |
|
238 | 238 | e.g to Display Wes' foundational paper about PANDAS in book mode from page 3 |
|
239 | 239 | |
|
240 | 240 | ScribdDocument(71048089, width=800, height=400, start_page=3, view_mode="book") |
|
241 | 241 | """ |
|
242 | 242 | |
|
243 | 243 | def __init__(self, id, width=400, height=300, **kwargs): |
|
244 | 244 | src="http://www.scribd.com/embeds/{0}/content".format(id) |
|
245 | 245 | super(ScribdDocument, self).__init__(src, width, height, **kwargs) |
|
246 | 246 | |
|
247 | 247 | class FileLink(object): |
|
248 | 248 | """Class for embedding a local file link in an IPython session, based on path |
|
249 | 249 | |
|
250 | 250 | e.g. to embed a link that was generated in the IPython notebook as my/data.txt |
|
251 | 251 | |
|
252 | 252 | you would do:: |
|
253 | 253 | |
|
254 | 254 | local_file = FileLink("my/data.txt") |
|
255 | 255 | display(local_file) |
|
256 | 256 | |
|
257 | 257 | or in the HTML notebook, just:: |
|
258 | 258 | |
|
259 | 259 | FileLink("my/data.txt") |
|
260 | 260 | """ |
|
261 | 261 | |
|
262 | 262 | html_link_str = "<a href='%s' target='_blank'>%s</a>" |
|
263 | 263 | |
|
264 | 264 | def __init__(self, |
|
265 | 265 | path, |
|
266 | 266 | url_prefix='files/', |
|
267 | 267 | result_html_prefix='', |
|
268 | 268 | result_html_suffix='<br>'): |
|
269 | 269 | """ |
|
270 | 270 | Parameters |
|
271 | 271 | ---------- |
|
272 | 272 | path : str |
|
273 | 273 | path to the file or directory that should be formatted |
|
274 | 274 | directory_prefix : str |
|
275 | 275 | prefix to be prepended to all files to form a working link [default: |
|
276 | 276 | 'files'] |
|
277 | 277 | result_html_prefix : str |
|
278 | 278 | text to append to beginning to link [default: none] |
|
279 | 279 | result_html_suffix : str |
|
280 | 280 | text to append at the end of link [default: '<br>'] |
|
281 | 281 | """ |
|
282 | 282 | if isdir(path): |
|
283 | raise ValueError,\ | |
|
284 | ("Cannot display a directory using FileLink. " | |
|
283 | raise ValueError("Cannot display a directory using FileLink. " | |
|
285 | 284 | "Use FileLinks to display '%s'." % path) |
|
286 | 285 | self.path = path |
|
287 | 286 | self.url_prefix = url_prefix |
|
288 | 287 | self.result_html_prefix = result_html_prefix |
|
289 | 288 | self.result_html_suffix = result_html_suffix |
|
290 | 289 | |
|
291 | 290 | def _format_path(self): |
|
292 | 291 | fp = ''.join([self.url_prefix,self.path]) |
|
293 | 292 | return ''.join([self.result_html_prefix, |
|
294 | 293 | self.html_link_str % (fp, self.path), |
|
295 | 294 | self.result_html_suffix]) |
|
296 | 295 | |
|
297 | 296 | def _repr_html_(self): |
|
298 | 297 | """return html link to file |
|
299 | 298 | """ |
|
300 | 299 | if not exists(self.path): |
|
301 | 300 | return ("Path (<tt>%s</tt>) doesn't exist. " |
|
302 | 301 | "It may still be in the process of " |
|
303 | 302 | "being generated, or you may have the " |
|
304 | 303 | "incorrect path." % self.path) |
|
305 | 304 | |
|
306 | 305 | return self._format_path() |
|
307 | 306 | |
|
308 | 307 | def __repr__(self): |
|
309 | 308 | """return absolute path to file |
|
310 | 309 | """ |
|
311 | 310 | return abspath(self.path) |
|
312 | 311 | |
|
313 | 312 | class FileLinks(FileLink): |
|
314 | 313 | """Class for embedding local file links in an IPython session, based on path |
|
315 | 314 | |
|
316 | 315 | e.g. to embed links to files that were generated in the IPython notebook under my/data |
|
317 | 316 | |
|
318 | 317 | you would do: |
|
319 | 318 | |
|
320 | 319 | local_files = FileLinks("my/data") |
|
321 | 320 | display(local_files) |
|
322 | 321 | |
|
323 | 322 | or in the HTML notebook, just |
|
324 | 323 | |
|
325 | 324 | FileLinks("my/data") |
|
326 | 325 | |
|
327 | 326 | """ |
|
328 | 327 | def __init__(self, |
|
329 | 328 | path, |
|
330 | 329 | url_prefix='files/', |
|
331 | 330 | included_suffixes=None, |
|
332 | 331 | result_html_prefix='', |
|
333 | 332 | result_html_suffix='<br>', |
|
334 | 333 | notebook_display_formatter=None, |
|
335 | 334 | terminal_display_formatter=None): |
|
336 | 335 | """ |
|
337 | 336 | included_suffixes : list of filename suffixes to include when |
|
338 | 337 | formatting output [default: include all files] |
|
339 | 338 | |
|
340 | 339 | See the FileLink (baseclass of LocalDirectory) docstring for |
|
341 | 340 | information on additional parameters. |
|
342 | 341 | |
|
343 | 342 | notebook_display_formatter : func used to format links for display |
|
344 | 343 | in the notebook. See discussion of formatter function below. |
|
345 | 344 | |
|
346 | 345 | terminal_display_formatter : func used to format links for display |
|
347 | 346 | in the terminal. See discussion of formatter function below. |
|
348 | 347 | |
|
349 | 348 | |
|
350 | 349 | Passing custom formatter functions |
|
351 | 350 | ---------------------------------- |
|
352 | 351 | Formatter functions must be of the form: |
|
353 | 352 | f(dirname, fnames, included_suffixes) |
|
354 | 353 | dirname : the name of a directory (a string), |
|
355 | 354 | fnames : a list of the files in that directory |
|
356 | 355 | included_suffixes : a list of the file suffixes that should be |
|
357 | 356 | included in the output (passing None means |
|
358 | 357 | to include all suffixes in the output in |
|
359 | 358 | the built-in formatters) |
|
360 | 359 | |
|
361 | 360 | returns a list of lines that should will be print in the |
|
362 | 361 | notebook (if passing notebook_display_formatter) or the terminal |
|
363 | 362 | (if passing terminal_display_formatter). This function is iterated |
|
364 | 363 | over for each directory in self.path. Default formatters are in |
|
365 | 364 | place, can be passed here to support alternative formatting. |
|
366 | 365 | |
|
367 | 366 | """ |
|
368 | 367 | if isfile(path): |
|
369 | raise ValueError,\ | |
|
370 | ("Cannot display a file using FileLinks. " | |
|
368 | raise ValueError("Cannot display a file using FileLinks. " | |
|
371 | 369 | "Use FileLink to display '%s'." % path) |
|
372 | 370 | self.included_suffixes = included_suffixes |
|
373 | 371 | # remove trailing slashs for more consistent output formatting |
|
374 | 372 | path = path.rstrip('/') |
|
375 | 373 | |
|
376 | 374 | self.path = path |
|
377 | 375 | self.url_prefix = url_prefix |
|
378 | 376 | self.result_html_prefix = result_html_prefix |
|
379 | 377 | self.result_html_suffix = result_html_suffix |
|
380 | 378 | |
|
381 | 379 | self.notebook_display_formatter = \ |
|
382 | 380 | notebook_display_formatter or self._get_notebook_display_formatter() |
|
383 | 381 | self.terminal_display_formatter = \ |
|
384 | 382 | terminal_display_formatter or self._get_terminal_display_formatter() |
|
385 | 383 | |
|
386 | 384 | def _get_display_formatter(self, |
|
387 | 385 | dirname_output_format, |
|
388 | 386 | fname_output_format, |
|
389 | 387 | fp_format, |
|
390 | 388 | fp_cleaner=None): |
|
391 | 389 | """ generate built-in formatter function |
|
392 | 390 | |
|
393 | 391 | this is used to define both the notebook and terminal built-in |
|
394 | 392 | formatters as they only differ by some wrapper text for each entry |
|
395 | 393 | |
|
396 | 394 | dirname_output_format: string to use for formatting directory |
|
397 | 395 | names, dirname will be substituted for a single "%s" which |
|
398 | 396 | must appear in this string |
|
399 | 397 | fname_output_format: string to use for formatting file names, |
|
400 | 398 | if a single "%s" appears in the string, fname will be substituted |
|
401 | 399 | if two "%s" appear in the string, the path to fname will be |
|
402 | 400 | substituted for the first and fname will be substituted for the |
|
403 | 401 | second |
|
404 | 402 | fp_format: string to use for formatting filepaths, must contain |
|
405 | 403 | exactly two "%s" and the dirname will be subsituted for the first |
|
406 | 404 | and fname will be substituted for the second |
|
407 | 405 | """ |
|
408 | 406 | def f(dirname, fnames, included_suffixes=None): |
|
409 | 407 | result = [] |
|
410 | 408 | # begin by figuring out which filenames, if any, |
|
411 | 409 | # are going to be displayed |
|
412 | 410 | display_fnames = [] |
|
413 | 411 | for fname in fnames: |
|
414 | 412 | if (isfile(join(dirname,fname)) and |
|
415 | 413 | (included_suffixes == None or |
|
416 | 414 | splitext(fname)[1] in included_suffixes)): |
|
417 | 415 | display_fnames.append(fname) |
|
418 | 416 | |
|
419 | 417 | if len(display_fnames) == 0: |
|
420 | 418 | # if there are no filenames to display, don't print anything |
|
421 | 419 | # (not even the directory name) |
|
422 | 420 | pass |
|
423 | 421 | else: |
|
424 | 422 | # otherwise print the formatted directory name followed by |
|
425 | 423 | # the formatted filenames |
|
426 | 424 | dirname_output_line = dirname_output_format % dirname |
|
427 | 425 | result.append(dirname_output_line) |
|
428 | 426 | for fname in display_fnames: |
|
429 | 427 | fp = fp_format % (dirname,fname) |
|
430 | 428 | if fp_cleaner is not None: |
|
431 | 429 | fp = fp_cleaner(fp) |
|
432 | 430 | try: |
|
433 | 431 | # output can include both a filepath and a filename... |
|
434 | 432 | fname_output_line = fname_output_format % (fp, fname) |
|
435 | 433 | except TypeError: |
|
436 | 434 | # ... or just a single filepath |
|
437 | 435 | fname_output_line = fname_output_format % fname |
|
438 | 436 | result.append(fname_output_line) |
|
439 | 437 | return result |
|
440 | 438 | return f |
|
441 | 439 | |
|
442 | 440 | def _get_notebook_display_formatter(self, |
|
443 | 441 | spacer=" "): |
|
444 | 442 | """ generate function to use for notebook formatting |
|
445 | 443 | """ |
|
446 | 444 | dirname_output_format = \ |
|
447 | 445 | self.result_html_prefix + "%s/" + self.result_html_suffix |
|
448 | 446 | fname_output_format = \ |
|
449 | 447 | self.result_html_prefix + spacer + self.html_link_str + self.result_html_suffix |
|
450 | 448 | fp_format = self.url_prefix + '%s/%s' |
|
451 | 449 | if sep == "\\": |
|
452 | 450 | # Working on a platform where the path separator is "\", so |
|
453 | 451 | # must convert these to "/" for generating a URI |
|
454 | 452 | def fp_cleaner(fp): |
|
455 | 453 | # Replace all occurences of backslash ("\") with a forward |
|
456 | 454 | # slash ("/") - this is necessary on windows when a path is |
|
457 | 455 | # provided as input, but we must link to a URI |
|
458 | 456 | return fp.replace('\\','/') |
|
459 | 457 | else: |
|
460 | 458 | fp_cleaner = None |
|
461 | 459 | |
|
462 | 460 | return self._get_display_formatter(dirname_output_format, |
|
463 | 461 | fname_output_format, |
|
464 | 462 | fp_format, |
|
465 | 463 | fp_cleaner) |
|
466 | 464 | |
|
467 | 465 | def _get_terminal_display_formatter(self, |
|
468 | 466 | spacer=" "): |
|
469 | 467 | """ generate function to use for terminal formatting |
|
470 | 468 | """ |
|
471 | 469 | dirname_output_format = "%s/" |
|
472 | 470 | fname_output_format = spacer + "%s" |
|
473 | 471 | fp_format = '%s/%s' |
|
474 | 472 | |
|
475 | 473 | return self._get_display_formatter(dirname_output_format, |
|
476 | 474 | fname_output_format, |
|
477 | 475 | fp_format) |
|
478 | 476 | |
|
479 | 477 | def _format_path(self): |
|
480 | 478 | result_lines = [] |
|
481 | 479 | walked_dir = list(walk(self.path)) |
|
482 | 480 | walked_dir.sort() |
|
483 | 481 | for dirname, subdirs, fnames in walked_dir: |
|
484 | 482 | result_lines += self.notebook_display_formatter(dirname, fnames, self.included_suffixes) |
|
485 | 483 | return '\n'.join(result_lines) |
|
486 | 484 | |
|
487 | 485 | def __repr__(self): |
|
488 | 486 | """return newline-separated absolute paths |
|
489 | 487 | """ |
|
490 | 488 | result_lines = [] |
|
491 | 489 | walked_dir = list(walk(self.path)) |
|
492 | 490 | walked_dir.sort() |
|
493 | 491 | for dirname, subdirs, fnames in walked_dir: |
|
494 | 492 | result_lines += self.terminal_display_formatter(dirname, fnames, self.included_suffixes) |
|
495 | 493 | return '\n'.join(result_lines) |
@@ -1,439 +1,439 b'' | |||
|
1 | 1 | """Patched version of standard library tokenize, to deal with various bugs. |
|
2 | 2 | |
|
3 | 3 | Patches |
|
4 | 4 | |
|
5 | 5 | - Relevant parts of Gareth Rees' patch for Python issue #12691 (untokenizing), |
|
6 | 6 | manually applied. |
|
7 | 7 | - Newlines in comments and blank lines should be either NL or NEWLINE, depending |
|
8 | 8 | on whether they are in a multi-line statement. Filed as Python issue #17061. |
|
9 | 9 | |
|
10 | 10 | ------------------------------------------------------------------------------- |
|
11 | 11 | Tokenization help for Python programs. |
|
12 | 12 | |
|
13 | 13 | generate_tokens(readline) is a generator that breaks a stream of |
|
14 | 14 | text into Python tokens. It accepts a readline-like method which is called |
|
15 | 15 | repeatedly to get the next line of input (or "" for EOF). It generates |
|
16 | 16 | 5-tuples with these members: |
|
17 | 17 | |
|
18 | 18 | the token type (see token.py) |
|
19 | 19 | the token (a string) |
|
20 | 20 | the starting (row, column) indices of the token (a 2-tuple of ints) |
|
21 | 21 | the ending (row, column) indices of the token (a 2-tuple of ints) |
|
22 | 22 | the original line (string) |
|
23 | 23 | |
|
24 | 24 | It is designed to match the working of the Python tokenizer exactly, except |
|
25 | 25 | that it produces COMMENT tokens for comments and gives type OP for all |
|
26 | 26 | operators |
|
27 | 27 | |
|
28 | 28 | Older entry points |
|
29 | 29 | tokenize_loop(readline, tokeneater) |
|
30 | 30 | tokenize(readline, tokeneater=printtoken) |
|
31 | 31 | are the same, except instead of generating tokens, tokeneater is a callback |
|
32 | 32 | function to which the 5 fields described above are passed as 5 arguments, |
|
33 | 33 | each time a new token is found.""" |
|
34 | 34 | from __future__ import print_function |
|
35 | 35 | |
|
36 | 36 | __author__ = 'Ka-Ping Yee <ping@lfw.org>' |
|
37 | 37 | __credits__ = ('GvR, ESR, Tim Peters, Thomas Wouters, Fred Drake, ' |
|
38 | 38 | 'Skip Montanaro, Raymond Hettinger') |
|
39 | 39 | |
|
40 | 40 | import string, re |
|
41 | 41 | from token import * |
|
42 | 42 | |
|
43 | 43 | import token |
|
44 | 44 | __all__ = [x for x in dir(token) if not x.startswith("_")] |
|
45 | 45 | __all__ += ["COMMENT", "tokenize", "generate_tokens", "NL", "untokenize"] |
|
46 | 46 | del x |
|
47 | 47 | del token |
|
48 | 48 | |
|
49 | 49 | __all__ += ["TokenError"] |
|
50 | 50 | |
|
51 | 51 | COMMENT = N_TOKENS |
|
52 | 52 | tok_name[COMMENT] = 'COMMENT' |
|
53 | 53 | NL = N_TOKENS + 1 |
|
54 | 54 | tok_name[NL] = 'NL' |
|
55 | 55 | N_TOKENS += 2 |
|
56 | 56 | |
|
57 | 57 | def group(*choices): return '(' + '|'.join(choices) + ')' |
|
58 | 58 | def any(*choices): return group(*choices) + '*' |
|
59 | 59 | def maybe(*choices): return group(*choices) + '?' |
|
60 | 60 | |
|
61 | 61 | Whitespace = r'[ \f\t]*' |
|
62 | 62 | Comment = r'#[^\r\n]*' |
|
63 | 63 | Ignore = Whitespace + any(r'\\\r?\n' + Whitespace) + maybe(Comment) |
|
64 | 64 | Name = r'[a-zA-Z_]\w*' |
|
65 | 65 | |
|
66 | 66 | Hexnumber = r'0[xX][\da-fA-F]+[lL]?' |
|
67 | 67 | Octnumber = r'(0[oO][0-7]+)|(0[0-7]*)[lL]?' |
|
68 | 68 | Binnumber = r'0[bB][01]+[lL]?' |
|
69 | 69 | Decnumber = r'[1-9]\d*[lL]?' |
|
70 | 70 | Intnumber = group(Hexnumber, Binnumber, Octnumber, Decnumber) |
|
71 | 71 | Exponent = r'[eE][-+]?\d+' |
|
72 | 72 | Pointfloat = group(r'\d+\.\d*', r'\.\d+') + maybe(Exponent) |
|
73 | 73 | Expfloat = r'\d+' + Exponent |
|
74 | 74 | Floatnumber = group(Pointfloat, Expfloat) |
|
75 | 75 | Imagnumber = group(r'\d+[jJ]', Floatnumber + r'[jJ]') |
|
76 | 76 | Number = group(Imagnumber, Floatnumber, Intnumber) |
|
77 | 77 | |
|
78 | 78 | # Tail end of ' string. |
|
79 | 79 | Single = r"[^'\\]*(?:\\.[^'\\]*)*'" |
|
80 | 80 | # Tail end of " string. |
|
81 | 81 | Double = r'[^"\\]*(?:\\.[^"\\]*)*"' |
|
82 | 82 | # Tail end of ''' string. |
|
83 | 83 | Single3 = r"[^'\\]*(?:(?:\\.|'(?!''))[^'\\]*)*'''" |
|
84 | 84 | # Tail end of """ string. |
|
85 | 85 | Double3 = r'[^"\\]*(?:(?:\\.|"(?!""))[^"\\]*)*"""' |
|
86 | 86 | Triple = group("[uUbB]?[rR]?'''", '[uUbB]?[rR]?"""') |
|
87 | 87 | # Single-line ' or " string. |
|
88 | 88 | String = group(r"[uUbB]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*'", |
|
89 | 89 | r'[uUbB]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*"') |
|
90 | 90 | |
|
91 | 91 | # Because of leftmost-then-longest match semantics, be sure to put the |
|
92 | 92 | # longest operators first (e.g., if = came before ==, == would get |
|
93 | 93 | # recognized as two instances of =). |
|
94 | 94 | Operator = group(r"\*\*=?", r">>=?", r"<<=?", r"<>", r"!=", |
|
95 | 95 | r"//=?", |
|
96 | 96 | r"[+\-*/%&|^=<>]=?", |
|
97 | 97 | r"~") |
|
98 | 98 | |
|
99 | 99 | Bracket = '[][(){}]' |
|
100 | 100 | Special = group(r'\r?\n', r'[:;.,`@]') |
|
101 | 101 | Funny = group(Operator, Bracket, Special) |
|
102 | 102 | |
|
103 | 103 | PlainToken = group(Number, Funny, String, Name) |
|
104 | 104 | Token = Ignore + PlainToken |
|
105 | 105 | |
|
106 | 106 | # First (or only) line of ' or " string. |
|
107 | 107 | ContStr = group(r"[uUbB]?[rR]?'[^\n'\\]*(?:\\.[^\n'\\]*)*" + |
|
108 | 108 | group("'", r'\\\r?\n'), |
|
109 | 109 | r'[uUbB]?[rR]?"[^\n"\\]*(?:\\.[^\n"\\]*)*' + |
|
110 | 110 | group('"', r'\\\r?\n')) |
|
111 | 111 | PseudoExtras = group(r'\\\r?\n', Comment, Triple) |
|
112 | 112 | PseudoToken = Whitespace + group(PseudoExtras, Number, Funny, ContStr, Name) |
|
113 | 113 | |
|
114 | 114 | tokenprog, pseudoprog, single3prog, double3prog = map( |
|
115 | 115 | re.compile, (Token, PseudoToken, Single3, Double3)) |
|
116 | 116 | endprogs = {"'": re.compile(Single), '"': re.compile(Double), |
|
117 | 117 | "'''": single3prog, '"""': double3prog, |
|
118 | 118 | "r'''": single3prog, 'r"""': double3prog, |
|
119 | 119 | "u'''": single3prog, 'u"""': double3prog, |
|
120 | 120 | "ur'''": single3prog, 'ur"""': double3prog, |
|
121 | 121 | "R'''": single3prog, 'R"""': double3prog, |
|
122 | 122 | "U'''": single3prog, 'U"""': double3prog, |
|
123 | 123 | "uR'''": single3prog, 'uR"""': double3prog, |
|
124 | 124 | "Ur'''": single3prog, 'Ur"""': double3prog, |
|
125 | 125 | "UR'''": single3prog, 'UR"""': double3prog, |
|
126 | 126 | "b'''": single3prog, 'b"""': double3prog, |
|
127 | 127 | "br'''": single3prog, 'br"""': double3prog, |
|
128 | 128 | "B'''": single3prog, 'B"""': double3prog, |
|
129 | 129 | "bR'''": single3prog, 'bR"""': double3prog, |
|
130 | 130 | "Br'''": single3prog, 'Br"""': double3prog, |
|
131 | 131 | "BR'''": single3prog, 'BR"""': double3prog, |
|
132 | 132 | 'r': None, 'R': None, 'u': None, 'U': None, |
|
133 | 133 | 'b': None, 'B': None} |
|
134 | 134 | |
|
135 | 135 | triple_quoted = {} |
|
136 | 136 | for t in ("'''", '"""', |
|
137 | 137 | "r'''", 'r"""', "R'''", 'R"""', |
|
138 | 138 | "u'''", 'u"""', "U'''", 'U"""', |
|
139 | 139 | "ur'''", 'ur"""', "Ur'''", 'Ur"""', |
|
140 | 140 | "uR'''", 'uR"""', "UR'''", 'UR"""', |
|
141 | 141 | "b'''", 'b"""', "B'''", 'B"""', |
|
142 | 142 | "br'''", 'br"""', "Br'''", 'Br"""', |
|
143 | 143 | "bR'''", 'bR"""', "BR'''", 'BR"""'): |
|
144 | 144 | triple_quoted[t] = t |
|
145 | 145 | single_quoted = {} |
|
146 | 146 | for t in ("'", '"', |
|
147 | 147 | "r'", 'r"', "R'", 'R"', |
|
148 | 148 | "u'", 'u"', "U'", 'U"', |
|
149 | 149 | "ur'", 'ur"', "Ur'", 'Ur"', |
|
150 | 150 | "uR'", 'uR"', "UR'", 'UR"', |
|
151 | 151 | "b'", 'b"', "B'", 'B"', |
|
152 | 152 | "br'", 'br"', "Br'", 'Br"', |
|
153 | 153 | "bR'", 'bR"', "BR'", 'BR"' ): |
|
154 | 154 | single_quoted[t] = t |
|
155 | 155 | |
|
156 | 156 | tabsize = 8 |
|
157 | 157 | |
|
158 | 158 | class TokenError(Exception): pass |
|
159 | 159 | |
|
160 | 160 | class StopTokenizing(Exception): pass |
|
161 | 161 | |
|
162 | 162 | def printtoken(type, token, srow_scol, erow_ecol, line): # for testing |
|
163 | 163 | srow, scol = srow_scol |
|
164 | 164 | erow, ecol = erow_ecol |
|
165 | 165 | print("%d,%d-%d,%d:\t%s\t%s" % \ |
|
166 | 166 | (srow, scol, erow, ecol, tok_name[type], repr(token))) |
|
167 | 167 | |
|
168 | 168 | def tokenize(readline, tokeneater=printtoken): |
|
169 | 169 | """ |
|
170 | 170 | The tokenize() function accepts two parameters: one representing the |
|
171 | 171 | input stream, and one providing an output mechanism for tokenize(). |
|
172 | 172 | |
|
173 | 173 | The first parameter, readline, must be a callable object which provides |
|
174 | 174 | the same interface as the readline() method of built-in file objects. |
|
175 | 175 | Each call to the function should return one line of input as a string. |
|
176 | 176 | |
|
177 | 177 | The second parameter, tokeneater, must also be a callable object. It is |
|
178 | 178 | called once for each token, with five arguments, corresponding to the |
|
179 | 179 | tuples generated by generate_tokens(). |
|
180 | 180 | """ |
|
181 | 181 | try: |
|
182 | 182 | tokenize_loop(readline, tokeneater) |
|
183 | 183 | except StopTokenizing: |
|
184 | 184 | pass |
|
185 | 185 | |
|
186 | 186 | # backwards compatible interface |
|
187 | 187 | def tokenize_loop(readline, tokeneater): |
|
188 | 188 | for token_info in generate_tokens(readline): |
|
189 | 189 | tokeneater(*token_info) |
|
190 | 190 | |
|
191 | 191 | class Untokenizer: |
|
192 | 192 | |
|
193 | 193 | def __init__(self): |
|
194 | 194 | self.tokens = [] |
|
195 | 195 | self.prev_row = 1 |
|
196 | 196 | self.prev_col = 0 |
|
197 | 197 | |
|
198 | 198 | def add_whitespace(self, start): |
|
199 | 199 | row, col = start |
|
200 | 200 | assert row >= self.prev_row |
|
201 | 201 | col_offset = col - self.prev_col |
|
202 | 202 | if col_offset > 0: |
|
203 | 203 | self.tokens.append(" " * col_offset) |
|
204 | 204 | elif row > self.prev_row and tok_type not in (NEWLINE, NL, ENDMARKER): |
|
205 | 205 | # Line was backslash-continued |
|
206 | 206 | self.tokens.append(" ") |
|
207 | 207 | |
|
208 | 208 | def untokenize(self, tokens): |
|
209 | 209 | iterable = iter(tokens) |
|
210 | 210 | for t in iterable: |
|
211 | 211 | if len(t) == 2: |
|
212 | 212 | self.compat(t, iterable) |
|
213 | 213 | break |
|
214 | 214 | tok_type, token, start, end = t[:4] |
|
215 | 215 | self.add_whitespace(start) |
|
216 | 216 | self.tokens.append(token) |
|
217 | 217 | self.prev_row, self.prev_col = end |
|
218 | 218 | if tok_type in (NEWLINE, NL): |
|
219 | 219 | self.prev_row += 1 |
|
220 | 220 | self.prev_col = 0 |
|
221 | 221 | return "".join(self.tokens) |
|
222 | 222 | |
|
223 | 223 | def compat(self, token, iterable): |
|
224 | 224 | # This import is here to avoid problems when the itertools |
|
225 | 225 | # module is not built yet and tokenize is imported. |
|
226 | 226 | from itertools import chain |
|
227 | 227 | startline = False |
|
228 | 228 | prevstring = False |
|
229 | 229 | indents = [] |
|
230 | 230 | toks_append = self.tokens.append |
|
231 | 231 | for tok in chain([token], iterable): |
|
232 | 232 | toknum, tokval = tok[:2] |
|
233 | 233 | |
|
234 | 234 | if toknum in (NAME, NUMBER): |
|
235 | 235 | tokval += ' ' |
|
236 | 236 | |
|
237 | 237 | # Insert a space between two consecutive strings |
|
238 | 238 | if toknum == STRING: |
|
239 | 239 | if prevstring: |
|
240 | 240 | tokval = ' ' + tokval |
|
241 | 241 | prevstring = True |
|
242 | 242 | else: |
|
243 | 243 | prevstring = False |
|
244 | 244 | |
|
245 | 245 | if toknum == INDENT: |
|
246 | 246 | indents.append(tokval) |
|
247 | 247 | continue |
|
248 | 248 | elif toknum == DEDENT: |
|
249 | 249 | indents.pop() |
|
250 | 250 | continue |
|
251 | 251 | elif toknum in (NEWLINE, NL): |
|
252 | 252 | startline = True |
|
253 | 253 | elif startline and indents: |
|
254 | 254 | toks_append(indents[-1]) |
|
255 | 255 | startline = False |
|
256 | 256 | toks_append(tokval) |
|
257 | 257 | |
|
258 | 258 | def untokenize(iterable): |
|
259 | 259 | """Transform tokens back into Python source code. |
|
260 | 260 | |
|
261 | 261 | Each element returned by the iterable must be a token sequence |
|
262 | 262 | with at least two elements, a token number and token value. If |
|
263 | 263 | only two tokens are passed, the resulting output is poor. |
|
264 | 264 | |
|
265 | 265 | Round-trip invariant for full input: |
|
266 | 266 | Untokenized source will match input source exactly |
|
267 | 267 | |
|
268 | 268 | Round-trip invariant for limited intput: |
|
269 | 269 | # Output text will tokenize the back to the input |
|
270 | 270 | t1 = [tok[:2] for tok in generate_tokens(f.readline)] |
|
271 | 271 | newcode = untokenize(t1) |
|
272 | 272 | readline = iter(newcode.splitlines(1)).next |
|
273 | 273 | t2 = [tok[:2] for tok in generate_tokens(readline)] |
|
274 | 274 | assert t1 == t2 |
|
275 | 275 | """ |
|
276 | 276 | ut = Untokenizer() |
|
277 | 277 | return ut.untokenize(iterable) |
|
278 | 278 | |
|
279 | 279 | def generate_tokens(readline): |
|
280 | 280 | """ |
|
281 | 281 | The generate_tokens() generator requires one argment, readline, which |
|
282 | 282 | must be a callable object which provides the same interface as the |
|
283 | 283 | readline() method of built-in file objects. Each call to the function |
|
284 | 284 | should return one line of input as a string. Alternately, readline |
|
285 | 285 | can be a callable function terminating with StopIteration: |
|
286 | 286 | readline = open(myfile).next # Example of alternate readline |
|
287 | 287 | |
|
288 | 288 | The generator produces 5-tuples with these members: the token type; the |
|
289 | 289 | token string; a 2-tuple (srow, scol) of ints specifying the row and |
|
290 | 290 | column where the token begins in the source; a 2-tuple (erow, ecol) of |
|
291 | 291 | ints specifying the row and column where the token ends in the source; |
|
292 | 292 | and the line on which the token was found. The line passed is the |
|
293 | 293 | logical line; continuation lines are included. |
|
294 | 294 | """ |
|
295 | 295 | lnum = parenlev = continued = 0 |
|
296 | 296 | namechars, numchars = string.ascii_letters + '_', '0123456789' |
|
297 | 297 | contstr, needcont = '', 0 |
|
298 | 298 | contline = None |
|
299 | 299 | indents = [0] |
|
300 | 300 | |
|
301 | 301 | while 1: # loop over lines in stream |
|
302 | 302 | try: |
|
303 | 303 | line = readline() |
|
304 | 304 | except StopIteration: |
|
305 | 305 | line = '' |
|
306 | 306 | lnum += 1 |
|
307 | 307 | pos, max = 0, len(line) |
|
308 | 308 | |
|
309 | 309 | if contstr: # continued string |
|
310 | 310 | if not line: |
|
311 |
raise TokenError |
|
|
311 | raise TokenError("EOF in multi-line string", strstart) | |
|
312 | 312 | endmatch = endprog.match(line) |
|
313 | 313 | if endmatch: |
|
314 | 314 | pos = end = endmatch.end(0) |
|
315 | 315 | yield (STRING, contstr + line[:end], |
|
316 | 316 | strstart, (lnum, end), contline + line) |
|
317 | 317 | contstr, needcont = '', 0 |
|
318 | 318 | contline = None |
|
319 | 319 | elif needcont and line[-2:] != '\\\n' and line[-3:] != '\\\r\n': |
|
320 | 320 | yield (ERRORTOKEN, contstr + line, |
|
321 | 321 | strstart, (lnum, len(line)), contline) |
|
322 | 322 | contstr = '' |
|
323 | 323 | contline = None |
|
324 | 324 | continue |
|
325 | 325 | else: |
|
326 | 326 | contstr = contstr + line |
|
327 | 327 | contline = contline + line |
|
328 | 328 | continue |
|
329 | 329 | |
|
330 | 330 | elif parenlev == 0 and not continued: # new statement |
|
331 | 331 | if not line: break |
|
332 | 332 | column = 0 |
|
333 | 333 | while pos < max: # measure leading whitespace |
|
334 | 334 | if line[pos] == ' ': |
|
335 | 335 | column += 1 |
|
336 | 336 | elif line[pos] == '\t': |
|
337 | 337 | column = (column//tabsize + 1)*tabsize |
|
338 | 338 | elif line[pos] == '\f': |
|
339 | 339 | column = 0 |
|
340 | 340 | else: |
|
341 | 341 | break |
|
342 | 342 | pos += 1 |
|
343 | 343 | if pos == max: |
|
344 | 344 | break |
|
345 | 345 | |
|
346 | 346 | if line[pos] in '#\r\n': # skip comments or blank lines |
|
347 | 347 | if line[pos] == '#': |
|
348 | 348 | comment_token = line[pos:].rstrip('\r\n') |
|
349 | 349 | nl_pos = pos + len(comment_token) |
|
350 | 350 | yield (COMMENT, comment_token, |
|
351 | 351 | (lnum, pos), (lnum, pos + len(comment_token)), line) |
|
352 | 352 | yield (NEWLINE, line[nl_pos:], |
|
353 | 353 | (lnum, nl_pos), (lnum, len(line)), line) |
|
354 | 354 | else: |
|
355 | 355 | yield (NEWLINE, line[pos:], |
|
356 | 356 | (lnum, pos), (lnum, len(line)), line) |
|
357 | 357 | continue |
|
358 | 358 | |
|
359 | 359 | if column > indents[-1]: # count indents or dedents |
|
360 | 360 | indents.append(column) |
|
361 | 361 | yield (INDENT, line[:pos], (lnum, 0), (lnum, pos), line) |
|
362 | 362 | while column < indents[-1]: |
|
363 | 363 | if column not in indents: |
|
364 | 364 | raise IndentationError( |
|
365 | 365 | "unindent does not match any outer indentation level", |
|
366 | 366 | ("<tokenize>", lnum, pos, line)) |
|
367 | 367 | indents = indents[:-1] |
|
368 | 368 | yield (DEDENT, '', (lnum, pos), (lnum, pos), line) |
|
369 | 369 | |
|
370 | 370 | else: # continued statement |
|
371 | 371 | if not line: |
|
372 |
raise TokenError |
|
|
372 | raise TokenError("EOF in multi-line statement", (lnum, 0)) | |
|
373 | 373 | continued = 0 |
|
374 | 374 | |
|
375 | 375 | while pos < max: |
|
376 | 376 | pseudomatch = pseudoprog.match(line, pos) |
|
377 | 377 | if pseudomatch: # scan for tokens |
|
378 | 378 | start, end = pseudomatch.span(1) |
|
379 | 379 | spos, epos, pos = (lnum, start), (lnum, end), end |
|
380 | 380 | token, initial = line[start:end], line[start] |
|
381 | 381 | |
|
382 | 382 | if initial in numchars or \ |
|
383 | 383 | (initial == '.' and token != '.'): # ordinary number |
|
384 | 384 | yield (NUMBER, token, spos, epos, line) |
|
385 | 385 | elif initial in '\r\n': |
|
386 | 386 | yield (NL if parenlev > 0 else NEWLINE, |
|
387 | 387 | token, spos, epos, line) |
|
388 | 388 | elif initial == '#': |
|
389 | 389 | assert not token.endswith("\n") |
|
390 | 390 | yield (COMMENT, token, spos, epos, line) |
|
391 | 391 | elif token in triple_quoted: |
|
392 | 392 | endprog = endprogs[token] |
|
393 | 393 | endmatch = endprog.match(line, pos) |
|
394 | 394 | if endmatch: # all on one line |
|
395 | 395 | pos = endmatch.end(0) |
|
396 | 396 | token = line[start:pos] |
|
397 | 397 | yield (STRING, token, spos, (lnum, pos), line) |
|
398 | 398 | else: |
|
399 | 399 | strstart = (lnum, start) # multiple lines |
|
400 | 400 | contstr = line[start:] |
|
401 | 401 | contline = line |
|
402 | 402 | break |
|
403 | 403 | elif initial in single_quoted or \ |
|
404 | 404 | token[:2] in single_quoted or \ |
|
405 | 405 | token[:3] in single_quoted: |
|
406 | 406 | if token[-1] == '\n': # continued string |
|
407 | 407 | strstart = (lnum, start) |
|
408 | 408 | endprog = (endprogs[initial] or endprogs[token[1]] or |
|
409 | 409 | endprogs[token[2]]) |
|
410 | 410 | contstr, needcont = line[start:], 1 |
|
411 | 411 | contline = line |
|
412 | 412 | break |
|
413 | 413 | else: # ordinary string |
|
414 | 414 | yield (STRING, token, spos, epos, line) |
|
415 | 415 | elif initial in namechars: # ordinary name |
|
416 | 416 | yield (NAME, token, spos, epos, line) |
|
417 | 417 | elif initial == '\\': # continued stmt |
|
418 | 418 | continued = 1 |
|
419 | 419 | else: |
|
420 | 420 | if initial in '([{': |
|
421 | 421 | parenlev += 1 |
|
422 | 422 | elif initial in ')]}': |
|
423 | 423 | parenlev -= 1 |
|
424 | 424 | yield (OP, token, spos, epos, line) |
|
425 | 425 | else: |
|
426 | 426 | yield (ERRORTOKEN, line[pos], |
|
427 | 427 | (lnum, pos), (lnum, pos+1), line) |
|
428 | 428 | pos += 1 |
|
429 | 429 | |
|
430 | 430 | for indent in indents[1:]: # pop remaining indent levels |
|
431 | 431 | yield (DEDENT, '', (lnum, 0), (lnum, 0), '') |
|
432 | 432 | yield (ENDMARKER, '', (lnum, 0), (lnum, 0), '') |
|
433 | 433 | |
|
434 | 434 | if __name__ == '__main__': # testing |
|
435 | 435 | import sys |
|
436 | 436 | if len(sys.argv) > 1: |
|
437 | 437 | tokenize(open(sys.argv[1]).readline) |
|
438 | 438 | else: |
|
439 | 439 | tokenize(sys.stdin.readline) |
General Comments 0
You need to be logged in to leave comments.
Login now