Show More
@@ -1,305 +1,306 b'' | |||
|
1 | 1 | # -*- coding: utf-8 -*- |
|
2 | 2 | |
|
3 | 3 | # Copyright (C) 2011-2016 RhodeCode GmbH |
|
4 | 4 | # |
|
5 | 5 | # This program is free software: you can redistribute it and/or modify |
|
6 | 6 | # it under the terms of the GNU Affero General Public License, version 3 |
|
7 | 7 | # (only), as published by the Free Software Foundation. |
|
8 | 8 | # |
|
9 | 9 | # This program is distributed in the hope that it will be useful, |
|
10 | 10 | # but WITHOUT ANY WARRANTY; without even the implied warranty of |
|
11 | 11 | # MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the |
|
12 | 12 | # GNU General Public License for more details. |
|
13 | 13 | # |
|
14 | 14 | # You should have received a copy of the GNU Affero General Public License |
|
15 | 15 | # along with this program. If not, see <http://www.gnu.org/licenses/>. |
|
16 | 16 | # |
|
17 | 17 | # This program is dual-licensed. If you wish to learn more about the |
|
18 | 18 | # RhodeCode Enterprise Edition, including its added features, Support services, |
|
19 | 19 | # and proprietary license terms, please see https://rhodecode.com/licenses/ |
|
20 | 20 | |
|
21 | 21 | |
|
22 | 22 | """ |
|
23 | 23 | Renderer for markup languages with ability to parse using rst or markdown |
|
24 | 24 | """ |
|
25 | 25 | |
|
26 | 26 | import re |
|
27 | 27 | import os |
|
28 | 28 | import logging |
|
29 | 29 | import itertools |
|
30 | 30 | |
|
31 | 31 | from mako.lookup import TemplateLookup |
|
32 | 32 | |
|
33 | 33 | from docutils.core import publish_parts |
|
34 | 34 | from docutils.parsers.rst import directives |
|
35 | 35 | import markdown |
|
36 | 36 | |
|
37 | 37 | from rhodecode.lib.markdown_ext import ( |
|
38 | 38 | UrlizeExtension, GithubFlavoredMarkdownExtension) |
|
39 | 39 | from rhodecode.lib.utils2 import safe_unicode, md5_safe, MENTIONS_REGEX |
|
40 | 40 | |
|
41 | 41 | log = logging.getLogger(__name__) |
|
42 | 42 | |
|
43 | 43 | # default renderer used to generate automated comments |
|
44 | 44 | DEFAULT_COMMENTS_RENDERER = 'rst' |
|
45 | 45 | |
|
46 | 46 | |
|
47 | 47 | class MarkupRenderer(object): |
|
48 | 48 | RESTRUCTUREDTEXT_DISALLOWED_DIRECTIVES = ['include', 'meta', 'raw'] |
|
49 | 49 | |
|
50 | 50 | MARKDOWN_PAT = re.compile(r'\.(md|mkdn?|mdown|markdown)$', re.IGNORECASE) |
|
51 | 51 | RST_PAT = re.compile(r'\.re?st$', re.IGNORECASE) |
|
52 | 52 | PLAIN_PAT = re.compile(r'^readme$', re.IGNORECASE) |
|
53 | 53 | |
|
54 | 54 | # list of readme files to search in file tree and display in summary |
|
55 | 55 | # attached weights defines the search order lower is first |
|
56 | 56 | ALL_READMES = [ |
|
57 | 57 | ('readme', 0), ('README', 0), ('Readme', 0), |
|
58 | 58 | ('doc/readme', 1), ('doc/README', 1), ('doc/Readme', 1), |
|
59 | 59 | ('Docs/readme', 2), ('Docs/README', 2), ('Docs/Readme', 2), |
|
60 | 60 | ('DOCS/readme', 2), ('DOCS/README', 2), ('DOCS/Readme', 2), |
|
61 | 61 | ('docs/readme', 2), ('docs/README', 2), ('docs/Readme', 2), |
|
62 | 62 | ] |
|
63 | 63 | # extension together with weights. Lower is first means we control how |
|
64 | 64 | # extensions are attached to readme names with those. |
|
65 | 65 | PLAIN_EXTS = [ |
|
66 | 66 | ('', 0), # special case that renders READMES names without extension |
|
67 | 67 | ('.text', 2), ('.TEXT', 2), |
|
68 | 68 | ('.txt', 3), ('.TXT', 3) |
|
69 | 69 | ] |
|
70 | 70 | |
|
71 | 71 | RST_EXTS = [ |
|
72 | 72 | ('.rst', 1), ('.rest', 1), |
|
73 | 73 | ('.RST', 2), ('.REST', 2) |
|
74 | 74 | ] |
|
75 | 75 | |
|
76 | 76 | MARKDOWN_EXTS = [ |
|
77 | 77 | ('.md', 1), ('.MD', 1), |
|
78 | 78 | ('.mkdn', 2), ('.MKDN', 2), |
|
79 | 79 | ('.mdown', 3), ('.MDOWN', 3), |
|
80 | 80 | ('.markdown', 4), ('.MARKDOWN', 4) |
|
81 | 81 | ] |
|
82 | 82 | |
|
83 | 83 | ALL_EXTS = PLAIN_EXTS + MARKDOWN_EXTS + RST_EXTS |
|
84 | 84 | |
|
85 | 85 | def _detect_renderer(self, source, filename=None): |
|
86 | 86 | """ |
|
87 | 87 | runs detection of what renderer should be used for generating html |
|
88 | 88 | from a markup language |
|
89 | 89 | |
|
90 | 90 | filename can be also explicitly a renderer name |
|
91 | 91 | |
|
92 | 92 | :param source: |
|
93 | 93 | :param filename: |
|
94 | 94 | """ |
|
95 | 95 | |
|
96 | 96 | if MarkupRenderer.MARKDOWN_PAT.findall(filename): |
|
97 | 97 | detected_renderer = 'markdown' |
|
98 | 98 | elif MarkupRenderer.RST_PAT.findall(filename): |
|
99 | 99 | detected_renderer = 'rst' |
|
100 | 100 | elif MarkupRenderer.PLAIN_PAT.findall(filename): |
|
101 | 101 | detected_renderer = 'rst' |
|
102 | 102 | else: |
|
103 | 103 | detected_renderer = 'plain' |
|
104 | 104 | |
|
105 | 105 | return getattr(MarkupRenderer, detected_renderer) |
|
106 | 106 | |
|
107 | 107 | @classmethod |
|
108 | 108 | def renderer_from_filename(cls, filename, exclude): |
|
109 | 109 | """ |
|
110 |
Detect renderer from filename and optionally use ex |
|
|
111 | remove some options. This is mostly used in helpers | |
|
110 | Detect renderer markdown/rst from filename and optionally use exclude | |
|
111 | list to remove some options. This is mostly used in helpers. | |
|
112 | Returns None when no renderer can be detected. | |
|
112 | 113 | """ |
|
113 | 114 | def _filter(elements): |
|
114 | 115 | if isinstance(exclude, (list, tuple)): |
|
115 | 116 | return [x for x in elements if x not in exclude] |
|
116 | 117 | return elements |
|
117 | 118 | |
|
118 | 119 | if filename.endswith( |
|
119 | 120 | tuple(_filter([x[0] for x in cls.MARKDOWN_EXTS if x[0]]))): |
|
120 | 121 | return 'markdown' |
|
121 | 122 | if filename.endswith(tuple(_filter([x[0] for x in cls.RST_EXTS if x[0]]))): |
|
122 | 123 | return 'rst' |
|
123 | 124 | |
|
124 |
return |
|
|
125 | return None | |
|
125 | 126 | |
|
126 | 127 | @classmethod |
|
127 | 128 | def generate_readmes(cls, all_readmes, extensions): |
|
128 | 129 | combined = itertools.product(all_readmes, extensions) |
|
129 | 130 | # sort by filename weight(y[0][1]) + extensions weight(y[1][1]) |
|
130 | 131 | prioritized_readmes = sorted(combined, key=lambda y: y[0][1] + y[1][1]) |
|
131 | 132 | # filename, extension |
|
132 | 133 | return [''.join([x[0][0], x[1][0]]) for x in prioritized_readmes] |
|
133 | 134 | |
|
134 | 135 | def pick_readme_order(self, default_renderer): |
|
135 | 136 | |
|
136 | 137 | if default_renderer == 'markdown': |
|
137 | 138 | markdown = self.generate_readmes(self.ALL_READMES, self.MARKDOWN_EXTS) |
|
138 | 139 | readme_order = markdown + self.generate_readmes( |
|
139 | 140 | self.ALL_READMES, self.RST_EXTS + self.PLAIN_EXTS) |
|
140 | 141 | elif default_renderer == 'rst': |
|
141 | 142 | markdown = self.generate_readmes(self.ALL_READMES, self.RST_EXTS) |
|
142 | 143 | readme_order = markdown + self.generate_readmes( |
|
143 | 144 | self.ALL_READMES, self.MARKDOWN_EXTS + self.PLAIN_EXTS) |
|
144 | 145 | else: |
|
145 | 146 | readme_order = self.generate_readmes(self.ALL_READMES, self.ALL_EXTS) |
|
146 | 147 | |
|
147 | 148 | return readme_order |
|
148 | 149 | |
|
149 | 150 | def render(self, source, filename=None): |
|
150 | 151 | """ |
|
151 | 152 | Renders a given filename using detected renderer |
|
152 | 153 | it detects renderers based on file extension or mimetype. |
|
153 | 154 | At last it will just do a simple html replacing new lines with <br/> |
|
154 | 155 | |
|
155 | 156 | :param file_name: |
|
156 | 157 | :param source: |
|
157 | 158 | """ |
|
158 | 159 | |
|
159 | 160 | renderer = self._detect_renderer(source, filename) |
|
160 | 161 | readme_data = renderer(source) |
|
161 | 162 | return readme_data |
|
162 | 163 | |
|
163 | 164 | @classmethod |
|
164 | 165 | def _flavored_markdown(cls, text): |
|
165 | 166 | """ |
|
166 | 167 | Github style flavored markdown |
|
167 | 168 | |
|
168 | 169 | :param text: |
|
169 | 170 | """ |
|
170 | 171 | |
|
171 | 172 | # Extract pre blocks. |
|
172 | 173 | extractions = {} |
|
173 | 174 | |
|
174 | 175 | def pre_extraction_callback(matchobj): |
|
175 | 176 | digest = md5_safe(matchobj.group(0)) |
|
176 | 177 | extractions[digest] = matchobj.group(0) |
|
177 | 178 | return "{gfm-extraction-%s}" % digest |
|
178 | 179 | pattern = re.compile(r'<pre>.*?</pre>', re.MULTILINE | re.DOTALL) |
|
179 | 180 | text = re.sub(pattern, pre_extraction_callback, text) |
|
180 | 181 | |
|
181 | 182 | # Prevent foo_bar_baz from ending up with an italic word in the middle. |
|
182 | 183 | def italic_callback(matchobj): |
|
183 | 184 | s = matchobj.group(0) |
|
184 | 185 | if list(s).count('_') >= 2: |
|
185 | 186 | return s.replace('_', r'\_') |
|
186 | 187 | return s |
|
187 | 188 | text = re.sub(r'^(?! {4}|\t)\w+_\w+_\w[\w_]*', italic_callback, text) |
|
188 | 189 | |
|
189 | 190 | # Insert pre block extractions. |
|
190 | 191 | def pre_insert_callback(matchobj): |
|
191 | 192 | return '\n\n' + extractions[matchobj.group(1)] |
|
192 | 193 | text = re.sub(r'\{gfm-extraction-([0-9a-f]{32})\}', |
|
193 | 194 | pre_insert_callback, text) |
|
194 | 195 | |
|
195 | 196 | return text |
|
196 | 197 | |
|
197 | 198 | @classmethod |
|
198 | 199 | def urlify_text(cls, text): |
|
199 | 200 | url_pat = re.compile(r'(http[s]?://(?:[a-zA-Z]|[0-9]|[$-_@.&+]' |
|
200 | 201 | r'|[!*\(\),]|(?:%[0-9a-fA-F][0-9a-fA-F]))+)') |
|
201 | 202 | |
|
202 | 203 | def url_func(match_obj): |
|
203 | 204 | url_full = match_obj.groups()[0] |
|
204 | 205 | return '<a href="%(url)s">%(url)s</a>' % ({'url': url_full}) |
|
205 | 206 | |
|
206 | 207 | return url_pat.sub(url_func, text) |
|
207 | 208 | |
|
208 | 209 | @classmethod |
|
209 | 210 | def plain(cls, source, universal_newline=True): |
|
210 | 211 | source = safe_unicode(source) |
|
211 | 212 | if universal_newline: |
|
212 | 213 | newline = '\n' |
|
213 | 214 | source = newline.join(source.splitlines()) |
|
214 | 215 | |
|
215 | 216 | source = cls.urlify_text(source) |
|
216 | 217 | return '<br />' + source.replace("\n", '<br />') |
|
217 | 218 | |
|
218 | 219 | @classmethod |
|
219 | 220 | def markdown(cls, source, safe=True, flavored=True, mentions=False): |
|
220 | 221 | # It does not allow to insert inline HTML. In presence of HTML tags, it |
|
221 | 222 | # will replace them instead with [HTML_REMOVED]. This is controlled by |
|
222 | 223 | # the safe_mode=True parameter of the markdown method. |
|
223 | 224 | extensions = ['codehilite', 'extra', 'def_list', 'sane_lists'] |
|
224 | 225 | if flavored: |
|
225 | 226 | extensions.append(GithubFlavoredMarkdownExtension()) |
|
226 | 227 | |
|
227 | 228 | if mentions: |
|
228 | 229 | mention_pat = re.compile(MENTIONS_REGEX) |
|
229 | 230 | |
|
230 | 231 | def wrapp(match_obj): |
|
231 | 232 | uname = match_obj.groups()[0] |
|
232 | 233 | return ' **@%(uname)s** ' % {'uname': uname} |
|
233 | 234 | mention_hl = mention_pat.sub(wrapp, source).strip() |
|
234 | 235 | # we extracted mentions render with this using Mentions false |
|
235 | 236 | return cls.markdown(mention_hl, safe=safe, flavored=flavored, |
|
236 | 237 | mentions=False) |
|
237 | 238 | |
|
238 | 239 | source = safe_unicode(source) |
|
239 | 240 | try: |
|
240 | 241 | if flavored: |
|
241 | 242 | source = cls._flavored_markdown(source) |
|
242 | 243 | return markdown.markdown( |
|
243 | 244 | source, extensions, safe_mode=True, enable_attributes=False) |
|
244 | 245 | except Exception: |
|
245 | 246 | log.exception('Error when rendering Markdown') |
|
246 | 247 | if safe: |
|
247 | 248 | log.debug('Fallback to render in plain mode') |
|
248 | 249 | return cls.plain(source) |
|
249 | 250 | else: |
|
250 | 251 | raise |
|
251 | 252 | |
|
252 | 253 | @classmethod |
|
253 | 254 | def rst(cls, source, safe=True, mentions=False): |
|
254 | 255 | if mentions: |
|
255 | 256 | mention_pat = re.compile(MENTIONS_REGEX) |
|
256 | 257 | |
|
257 | 258 | def wrapp(match_obj): |
|
258 | 259 | uname = match_obj.groups()[0] |
|
259 | 260 | return ' **@%(uname)s** ' % {'uname': uname} |
|
260 | 261 | mention_hl = mention_pat.sub(wrapp, source).strip() |
|
261 | 262 | # we extracted mentions render with this using Mentions false |
|
262 | 263 | return cls.rst(mention_hl, safe=safe, mentions=False) |
|
263 | 264 | |
|
264 | 265 | source = safe_unicode(source) |
|
265 | 266 | try: |
|
266 | 267 | docutils_settings = dict( |
|
267 | 268 | [(alias, None) for alias in |
|
268 | 269 | cls.RESTRUCTUREDTEXT_DISALLOWED_DIRECTIVES]) |
|
269 | 270 | |
|
270 | 271 | docutils_settings.update({'input_encoding': 'unicode', |
|
271 | 272 | 'report_level': 4}) |
|
272 | 273 | |
|
273 | 274 | for k, v in docutils_settings.iteritems(): |
|
274 | 275 | directives.register_directive(k, v) |
|
275 | 276 | |
|
276 | 277 | parts = publish_parts(source=source, |
|
277 | 278 | writer_name="html4css1", |
|
278 | 279 | settings_overrides=docutils_settings) |
|
279 | 280 | |
|
280 | 281 | return parts['html_title'] + parts["fragment"] |
|
281 | 282 | except Exception: |
|
282 | 283 | log.exception('Error when rendering RST') |
|
283 | 284 | if safe: |
|
284 | 285 | log.debug('Fallbacking to render in plain mode') |
|
285 | 286 | return cls.plain(source) |
|
286 | 287 | else: |
|
287 | 288 | raise |
|
288 | 289 | |
|
289 | 290 | |
|
290 | 291 | class RstTemplateRenderer(object): |
|
291 | 292 | |
|
292 | 293 | def __init__(self): |
|
293 | 294 | base = os.path.abspath(os.path.dirname(os.path.dirname(__file__))) |
|
294 | 295 | rst_template_dirs = [os.path.join(base, 'templates', 'rst_templates')] |
|
295 | 296 | self.template_store = TemplateLookup( |
|
296 | 297 | directories=rst_template_dirs, |
|
297 | 298 | input_encoding='utf-8', |
|
298 | 299 | imports=['from rhodecode.lib import helpers as h']) |
|
299 | 300 | |
|
300 | 301 | def _get_template(self, templatename): |
|
301 | 302 | return self.template_store.get_template(templatename) |
|
302 | 303 | |
|
303 | 304 | def render(self, template_name, **kwargs): |
|
304 | 305 | template = self._get_template(template_name) |
|
305 | 306 | return template.render(**kwargs) |
General Comments 0
You need to be logged in to leave comments.
Login now