OLD | NEW |
| (Empty) |
1 """ | |
2 Python Markdown | |
3 =============== | |
4 | |
5 Python Markdown converts Markdown to HTML and can be used as a library or | |
6 called from the command line. | |
7 | |
8 ## Basic usage as a module: | |
9 | |
10 import markdown | |
11 html = markdown.markdown(your_text_string) | |
12 | |
13 See <https://pythonhosted.org/Markdown/> for more | |
14 information and instructions on how to extend the functionality of | |
15 Python Markdown. Read that before you try modifying this file. | |
16 | |
17 ## Authors and License | |
18 | |
19 Started by [Manfred Stienstra](http://www.dwerg.net/). Continued and | |
20 maintained by [Yuri Takhteyev](http://www.freewisdom.org), [Waylan | |
21 Limberg](http://achinghead.com/) and [Artem Yunusov](http://blog.splyer.com). | |
22 | |
23 Contact: markdown@freewisdom.org | |
24 | |
25 Copyright 2007-2013 The Python Markdown Project (v. 1.7 and later) | |
26 Copyright 200? Django Software Foundation (OrderedDict implementation) | |
27 Copyright 2004, 2005, 2006 Yuri Takhteyev (v. 0.2-1.6b) | |
28 Copyright 2004 Manfred Stienstra (the original version) | |
29 | |
30 License: BSD (see LICENSE for details). | |
31 """ | |
32 | |
33 from __future__ import absolute_import | |
34 from __future__ import unicode_literals | |
35 from .__version__ import version, version_info # noqa | |
36 import codecs | |
37 import sys | |
38 import logging | |
39 import warnings | |
40 import importlib | |
41 from . import util | |
42 from .preprocessors import build_preprocessors | |
43 from .blockprocessors import build_block_parser | |
44 from .treeprocessors import build_treeprocessors | |
45 from .inlinepatterns import build_inlinepatterns | |
46 from .postprocessors import build_postprocessors | |
47 from .extensions import Extension | |
48 from .serializers import to_html_string, to_xhtml_string | |
49 | |
50 __all__ = ['Markdown', 'markdown', 'markdownFromFile'] | |
51 | |
52 | |
53 logger = logging.getLogger('MARKDOWN') | |
54 | |
55 | |
56 class Markdown(object): | |
57 """Convert Markdown to HTML.""" | |
58 | |
59 doc_tag = "div" # Element used to wrap document - later removed | |
60 | |
61 option_defaults = { | |
62 'html_replacement_text': '[HTML_REMOVED]', | |
63 'tab_length': 4, | |
64 'enable_attributes': True, | |
65 'smart_emphasis': True, | |
66 'lazy_ol': True, | |
67 } | |
68 | |
69 output_formats = { | |
70 'html': to_html_string, | |
71 'html4': to_html_string, | |
72 'html5': to_html_string, | |
73 'xhtml': to_xhtml_string, | |
74 'xhtml1': to_xhtml_string, | |
75 'xhtml5': to_xhtml_string, | |
76 } | |
77 | |
78 ESCAPED_CHARS = ['\\', '`', '*', '_', '{', '}', '[', ']', | |
79 '(', ')', '>', '#', '+', '-', '.', '!'] | |
80 | |
81 def __init__(self, *args, **kwargs): | |
82 """ | |
83 Creates a new Markdown instance. | |
84 | |
85 Keyword arguments: | |
86 | |
87 * extensions: A list of extensions. | |
88 If they are of type string, the module mdx_name.py will be loaded. | |
89 If they are a subclass of markdown.Extension, they will be used | |
90 as-is. | |
91 * extension_configs: Configuration settings for extensions. | |
92 * output_format: Format of output. Supported formats are: | |
93 * "xhtml1": Outputs XHTML 1.x. Default. | |
94 * "xhtml5": Outputs XHTML style tags of HTML 5 | |
95 * "xhtml": Outputs latest supported version of XHTML | |
96 (currently XHTML 1.1). | |
97 * "html4": Outputs HTML 4 | |
98 * "html5": Outputs HTML style tags of HTML 5 | |
99 * "html": Outputs latest supported version of HTML | |
100 (currently HTML 4). | |
101 Note that it is suggested that the more specific formats ("xhtml1" | |
102 and "html4") be used as "xhtml" or "html" may change in the future | |
103 if it makes sense at that time. | |
104 * safe_mode: Deprecated! Disallow raw html. One of "remove", "replace" | |
105 or "escape". | |
106 * html_replacement_text: Deprecated! Text used when safe_mode is set | |
107 to "replace". | |
108 * tab_length: Length of tabs in the source. Default: 4 | |
109 * enable_attributes: Enable the conversion of attributes. Default: True | |
110 * smart_emphasis: Treat `_connected_words_` intelligently Default: True | |
111 * lazy_ol: Ignore number of first item of ordered lists. Default: True | |
112 | |
113 """ | |
114 | |
115 # For backward compatibility, loop through old positional args | |
116 pos = ['extensions', 'extension_configs', 'safe_mode', 'output_format'] | |
117 for c, arg in enumerate(args): | |
118 if pos[c] not in kwargs: | |
119 kwargs[pos[c]] = arg | |
120 if c+1 == len(pos): # pragma: no cover | |
121 # ignore any additional args | |
122 break | |
123 if len(args): | |
124 warnings.warn('Positional arguments are deprecated in Markdown. ' | |
125 'Use keyword arguments only.', | |
126 DeprecationWarning) | |
127 | |
128 # Loop through kwargs and assign defaults | |
129 for option, default in self.option_defaults.items(): | |
130 setattr(self, option, kwargs.get(option, default)) | |
131 | |
132 self.safeMode = kwargs.get('safe_mode', False) | |
133 if self.safeMode and 'enable_attributes' not in kwargs: | |
134 # Disable attributes in safeMode when not explicitly set | |
135 self.enable_attributes = False | |
136 | |
137 if 'safe_mode' in kwargs: | |
138 warnings.warn('"safe_mode" is deprecated in Python-Markdown. ' | |
139 'Use an HTML sanitizer (like ' | |
140 'Bleach http://bleach.readthedocs.org/) ' | |
141 'if you are parsing untrusted markdown text. ' | |
142 'See the 2.6 release notes for more info', | |
143 DeprecationWarning) | |
144 | |
145 if 'html_replacement_text' in kwargs: | |
146 warnings.warn('The "html_replacement_text" keyword is ' | |
147 'deprecated along with "safe_mode".', | |
148 DeprecationWarning) | |
149 | |
150 self.registeredExtensions = [] | |
151 self.docType = "" | |
152 self.stripTopLevelTags = True | |
153 | |
154 self.build_parser() | |
155 | |
156 self.references = {} | |
157 self.htmlStash = util.HtmlStash() | |
158 self.registerExtensions(extensions=kwargs.get('extensions', []), | |
159 configs=kwargs.get('extension_configs', {})) | |
160 self.set_output_format(kwargs.get('output_format', 'xhtml1')) | |
161 self.reset() | |
162 | |
163 def build_parser(self): | |
164 """ Build the parser from the various parts. """ | |
165 self.preprocessors = build_preprocessors(self) | |
166 self.parser = build_block_parser(self) | |
167 self.inlinePatterns = build_inlinepatterns(self) | |
168 self.treeprocessors = build_treeprocessors(self) | |
169 self.postprocessors = build_postprocessors(self) | |
170 return self | |
171 | |
172 def registerExtensions(self, extensions, configs): | |
173 """ | |
174 Register extensions with this instance of Markdown. | |
175 | |
176 Keyword arguments: | |
177 | |
178 * extensions: A list of extensions, which can either | |
179 be strings or objects. See the docstring on Markdown. | |
180 * configs: A dictionary mapping module names to config options. | |
181 | |
182 """ | |
183 for ext in extensions: | |
184 if isinstance(ext, util.string_type): | |
185 ext = self.build_extension(ext, configs.get(ext, {})) | |
186 if isinstance(ext, Extension): | |
187 ext.extendMarkdown(self, globals()) | |
188 logger.debug( | |
189 'Successfully loaded extension "%s.%s".' | |
190 % (ext.__class__.__module__, ext.__class__.__name__) | |
191 ) | |
192 elif ext is not None: | |
193 raise TypeError( | |
194 'Extension "%s.%s" must be of type: "markdown.Extension"' | |
195 % (ext.__class__.__module__, ext.__class__.__name__)) | |
196 | |
197 return self | |
198 | |
199 def build_extension(self, ext_name, configs): | |
200 """Build extension by name, then return the module. | |
201 | |
202 The extension name may contain arguments as part of the string in the | |
203 following format: "extname(key1=value1,key2=value2)" | |
204 | |
205 """ | |
206 | |
207 configs = dict(configs) | |
208 | |
209 # Parse extensions config params (ignore the order) | |
210 pos = ext_name.find("(") # find the first "(" | |
211 if pos > 0: | |
212 ext_args = ext_name[pos+1:-1] | |
213 ext_name = ext_name[:pos] | |
214 pairs = [x.split("=") for x in ext_args.split(",")] | |
215 configs.update([(x.strip(), y.strip()) for (x, y) in pairs]) | |
216 warnings.warn('Setting configs in the Named Extension string is ' | |
217 'deprecated. It is recommended that you ' | |
218 'pass an instance of the extension class to ' | |
219 'Markdown or use the "extension_configs" keyword. ' | |
220 'The current behavior will raise an error in version 2
.7. ' | |
221 'See the Release Notes for Python-Markdown version ' | |
222 '2.6 for more info.', DeprecationWarning) | |
223 | |
224 # Get class name (if provided): `path.to.module:ClassName` | |
225 ext_name, class_name = ext_name.split(':', 1) \ | |
226 if ':' in ext_name else (ext_name, '') | |
227 | |
228 # Try loading the extension first from one place, then another | |
229 try: | |
230 # Assume string uses dot syntax (`path.to.some.module`) | |
231 module = importlib.import_module(ext_name) | |
232 logger.debug( | |
233 'Successfuly imported extension module "%s".' % ext_name | |
234 ) | |
235 # For backward compat (until deprecation) | |
236 # check that this is an extension. | |
237 if ('.' not in ext_name and not (hasattr(module, 'makeExtension') or | |
238 (class_name and hasattr(module, class_name)))): | |
239 # We have a name conflict | |
240 # eg: extensions=['tables'] and PyTables is installed | |
241 raise ImportError | |
242 except ImportError: | |
243 # Preppend `markdown.extensions.` to name | |
244 module_name = '.'.join(['markdown.extensions', ext_name]) | |
245 try: | |
246 module = importlib.import_module(module_name) | |
247 logger.debug( | |
248 'Successfuly imported extension module "%s".' % | |
249 module_name | |
250 ) | |
251 warnings.warn('Using short names for Markdown\'s builtin ' | |
252 'extensions is deprecated. Use the ' | |
253 'full path to the extension with Python\'s dot ' | |
254 'notation (eg: "%s" instead of "%s"). The ' | |
255 'current behavior will raise an error in version ' | |
256 '2.7. See the Release Notes for ' | |
257 'Python-Markdown version 2.6 for more info.' % | |
258 (module_name, ext_name), | |
259 DeprecationWarning) | |
260 except ImportError: | |
261 # Preppend `mdx_` to name | |
262 module_name_old_style = '_'.join(['mdx', ext_name]) | |
263 try: | |
264 module = importlib.import_module(module_name_old_style) | |
265 logger.debug( | |
266 'Successfuly imported extension module "%s".' % | |
267 module_name_old_style) | |
268 warnings.warn('Markdown\'s behavior of prepending "mdx_" ' | |
269 'to an extension name is deprecated. ' | |
270 'Use the full path to the ' | |
271 'extension with Python\'s dot notation ' | |
272 '(eg: "%s" instead of "%s"). The current ' | |
273 'behavior will raise an error in version 2.7.
' | |
274 'See the Release Notes for Python-Markdown ' | |
275 'version 2.6 for more info.' % | |
276 (module_name_old_style, ext_name), | |
277 DeprecationWarning) | |
278 except ImportError as e: | |
279 message = "Failed loading extension '%s' from '%s', '%s' " \ | |
280 "or '%s'" % (ext_name, ext_name, module_name, | |
281 module_name_old_style) | |
282 e.args = (message,) + e.args[1:] | |
283 raise | |
284 | |
285 if class_name: | |
286 # Load given class name from module. | |
287 return getattr(module, class_name)(**configs) | |
288 else: | |
289 # Expect makeExtension() function to return a class. | |
290 try: | |
291 return module.makeExtension(**configs) | |
292 except AttributeError as e: | |
293 message = e.args[0] | |
294 message = "Failed to initiate extension " \ | |
295 "'%s': %s" % (ext_name, message) | |
296 e.args = (message,) + e.args[1:] | |
297 raise | |
298 | |
299 def registerExtension(self, extension): | |
300 """ This gets called by the extension """ | |
301 self.registeredExtensions.append(extension) | |
302 return self | |
303 | |
304 def reset(self): | |
305 """ | |
306 Resets all state variables so that we can start with a new text. | |
307 """ | |
308 self.htmlStash.reset() | |
309 self.references.clear() | |
310 | |
311 for extension in self.registeredExtensions: | |
312 if hasattr(extension, 'reset'): | |
313 extension.reset() | |
314 | |
315 return self | |
316 | |
317 def set_output_format(self, format): | |
318 """ Set the output format for the class instance. """ | |
319 self.output_format = format.lower() | |
320 try: | |
321 self.serializer = self.output_formats[self.output_format] | |
322 except KeyError as e: | |
323 valid_formats = list(self.output_formats.keys()) | |
324 valid_formats.sort() | |
325 message = 'Invalid Output Format: "%s". Use one of %s.' \ | |
326 % (self.output_format, | |
327 '"' + '", "'.join(valid_formats) + '"') | |
328 e.args = (message,) + e.args[1:] | |
329 raise | |
330 return self | |
331 | |
332 def convert(self, source): | |
333 """ | |
334 Convert markdown to serialized XHTML or HTML. | |
335 | |
336 Keyword arguments: | |
337 | |
338 * source: Source text as a Unicode string. | |
339 | |
340 Markdown processing takes place in five steps: | |
341 | |
342 1. A bunch of "preprocessors" munge the input text. | |
343 2. BlockParser() parses the high-level structural elements of the | |
344 pre-processed text into an ElementTree. | |
345 3. A bunch of "treeprocessors" are run against the ElementTree. One | |
346 such treeprocessor runs InlinePatterns against the ElementTree, | |
347 detecting inline markup. | |
348 4. Some post-processors are run against the text after the ElementTree | |
349 has been serialized into text. | |
350 5. The output is written to a string. | |
351 | |
352 """ | |
353 | |
354 # Fixup the source text | |
355 if not source.strip(): | |
356 return '' # a blank unicode string | |
357 | |
358 try: | |
359 source = util.text_type(source) | |
360 except UnicodeDecodeError as e: | |
361 # Customise error message while maintaining original trackback | |
362 e.reason += '. -- Note: Markdown only accepts unicode input!' | |
363 raise | |
364 | |
365 # Split into lines and run the line preprocessors. | |
366 self.lines = source.split("\n") | |
367 for prep in self.preprocessors.values(): | |
368 self.lines = prep.run(self.lines) | |
369 | |
370 # Parse the high-level elements. | |
371 root = self.parser.parseDocument(self.lines).getroot() | |
372 | |
373 # Run the tree-processors | |
374 for treeprocessor in self.treeprocessors.values(): | |
375 newRoot = treeprocessor.run(root) | |
376 if newRoot is not None: | |
377 root = newRoot | |
378 | |
379 # Serialize _properly_. Strip top-level tags. | |
380 output = self.serializer(root) | |
381 if self.stripTopLevelTags: | |
382 try: | |
383 start = output.index( | |
384 '<%s>' % self.doc_tag) + len(self.doc_tag) + 2 | |
385 end = output.rindex('</%s>' % self.doc_tag) | |
386 output = output[start:end].strip() | |
387 except ValueError: # pragma: no cover | |
388 if output.strip().endswith('<%s />' % self.doc_tag): | |
389 # We have an empty document | |
390 output = '' | |
391 else: | |
392 # We have a serious problem | |
393 raise ValueError('Markdown failed to strip top-level ' | |
394 'tags. Document=%r' % output.strip()) | |
395 | |
396 # Run the text post-processors | |
397 for pp in self.postprocessors.values(): | |
398 output = pp.run(output) | |
399 | |
400 return output.strip() | |
401 | |
402 def convertFile(self, input=None, output=None, encoding=None): | |
403 """Converts a Markdown file and returns the HTML as a Unicode string. | |
404 | |
405 Decodes the file using the provided encoding (defaults to utf-8), | |
406 passes the file content to markdown, and outputs the html to either | |
407 the provided stream or the file with provided name, using the same | |
408 encoding as the source file. The 'xmlcharrefreplace' error handler is | |
409 used when encoding the output. | |
410 | |
411 **Note:** This is the only place that decoding and encoding of Unicode | |
412 takes place in Python-Markdown. (All other code is Unicode-in / | |
413 Unicode-out.) | |
414 | |
415 Keyword arguments: | |
416 | |
417 * input: File object or path. Reads from stdin if `None`. | |
418 * output: File object or path. Writes to stdout if `None`. | |
419 * encoding: Encoding of input and output files. Defaults to utf-8. | |
420 | |
421 """ | |
422 | |
423 encoding = encoding or "utf-8" | |
424 | |
425 # Read the source | |
426 if input: | |
427 if isinstance(input, util.string_type): | |
428 input_file = codecs.open(input, mode="r", encoding=encoding) | |
429 else: | |
430 input_file = codecs.getreader(encoding)(input) | |
431 text = input_file.read() | |
432 input_file.close() | |
433 else: | |
434 text = sys.stdin.read() | |
435 if not isinstance(text, util.text_type): | |
436 text = text.decode(encoding) | |
437 | |
438 text = text.lstrip('\ufeff') # remove the byte-order mark | |
439 | |
440 # Convert | |
441 html = self.convert(text) | |
442 | |
443 # Write to file or stdout | |
444 if output: | |
445 if isinstance(output, util.string_type): | |
446 output_file = codecs.open(output, "w", | |
447 encoding=encoding, | |
448 errors="xmlcharrefreplace") | |
449 output_file.write(html) | |
450 output_file.close() | |
451 else: | |
452 writer = codecs.getwriter(encoding) | |
453 output_file = writer(output, errors="xmlcharrefreplace") | |
454 output_file.write(html) | |
455 # Don't close here. User may want to write more. | |
456 else: | |
457 # Encode manually and write bytes to stdout. | |
458 html = html.encode(encoding, "xmlcharrefreplace") | |
459 try: | |
460 # Write bytes directly to buffer (Python 3). | |
461 sys.stdout.buffer.write(html) | |
462 except AttributeError: | |
463 # Probably Python 2, which works with bytes by default. | |
464 sys.stdout.write(html) | |
465 | |
466 return self | |
467 | |
468 | |
469 """ | |
470 EXPORTED FUNCTIONS | |
471 ============================================================================= | |
472 | |
473 Those are the two functions we really mean to export: markdown() and | |
474 markdownFromFile(). | |
475 """ | |
476 | |
477 | |
478 def markdown(text, *args, **kwargs): | |
479 """Convert a Markdown string to HTML and return HTML as a Unicode string. | |
480 | |
481 This is a shortcut function for `Markdown` class to cover the most | |
482 basic use case. It initializes an instance of Markdown, loads the | |
483 necessary extensions and runs the parser on the given text. | |
484 | |
485 Keyword arguments: | |
486 | |
487 * text: Markdown formatted text as Unicode or ASCII string. | |
488 * Any arguments accepted by the Markdown class. | |
489 | |
490 Returns: An HTML document as a string. | |
491 | |
492 """ | |
493 md = Markdown(*args, **kwargs) | |
494 return md.convert(text) | |
495 | |
496 | |
497 def markdownFromFile(*args, **kwargs): | |
498 """Read markdown code from a file and write it to a file or a stream. | |
499 | |
500 This is a shortcut function which initializes an instance of Markdown, | |
501 and calls the convertFile method rather than convert. | |
502 | |
503 Keyword arguments: | |
504 | |
505 * input: a file name or readable object. | |
506 * output: a file name or writable object. | |
507 * encoding: Encoding of input and output. | |
508 * Any arguments accepted by the Markdown class. | |
509 | |
510 """ | |
511 # For backward compatibility loop through positional args | |
512 pos = ['input', 'output', 'extensions', 'encoding'] | |
513 c = 0 | |
514 for arg in args: | |
515 if pos[c] not in kwargs: | |
516 kwargs[pos[c]] = arg | |
517 c += 1 | |
518 if c == len(pos): | |
519 break | |
520 if len(args): | |
521 warnings.warn('Positional arguments are depreacted in ' | |
522 'Markdown and will raise an error in version 2.7. ' | |
523 'Use keyword arguments only.', | |
524 DeprecationWarning) | |
525 | |
526 md = Markdown(**kwargs) | |
527 md.convertFile(kwargs.get('input', None), | |
528 kwargs.get('output', None), | |
529 kwargs.get('encoding', None)) | |
OLD | NEW |