Mercurial > genshi > genshi-test
comparison markup/filters.py @ 1:821114ec4f69
Initial import.
author | cmlenz |
---|---|
date | Sat, 03 Jun 2006 07:16:01 +0000 |
parents | |
children | c5890ef863ba |
comparison
equal
deleted
inserted
replaced
0:20f3417d4171 | 1:821114ec4f69 |
---|---|
1 # -*- coding: utf-8 -*- | |
2 # | |
3 # Copyright (C) 2006 Christopher Lenz | |
4 # All rights reserved. | |
5 # | |
6 # This software is licensed as described in the file COPYING, which | |
7 # you should have received as part of this distribution. The terms | |
8 # are also available at http://trac.edgewall.com/license.html. | |
9 # | |
10 # This software consists of voluntary contributions made by many | |
11 # individuals. For the exact contribution history, see the revision | |
12 # history and logs, available at http://projects.edgewall.com/trac/. | |
13 | |
14 """Implementation of a number of stream filters.""" | |
15 | |
16 try: | |
17 frozenset | |
18 except NameError: | |
19 from sets import ImmutableSet as frozenset | |
20 import re | |
21 | |
22 from markup.core import Attributes, Markup, Stream | |
23 from markup.path import Path | |
24 | |
25 __all__ = ['EvalFilter', 'IncludeFilter', 'MatchFilter', 'WhitespaceFilter', | |
26 'HTMLSanitizer'] | |
27 | |
28 | |
29 class EvalFilter(object): | |
30 """Responsible for evaluating expressions in a template.""" | |
31 | |
32 def __call__(self, stream, ctxt=None): | |
33 for kind, data, pos in stream: | |
34 | |
35 if kind is Stream.START: | |
36 # Attributes may still contain expressions in start tags at | |
37 # this point, so do some evaluation | |
38 tag, attrib = data | |
39 new_attrib = [] | |
40 for name, substream in attrib: | |
41 if isinstance(substream, basestring): | |
42 value = substream | |
43 else: | |
44 values = [] | |
45 for subkind, subdata, subpos in substream: | |
46 if subkind is Stream.EXPR: | |
47 values.append(subdata.evaluate(ctxt)) | |
48 else: | |
49 values.append(subdata) | |
50 value = filter(lambda x: x is not None, values) | |
51 if not value: | |
52 continue | |
53 new_attrib.append((name, ''.join(value))) | |
54 yield kind, (tag, Attributes(new_attrib)), pos | |
55 | |
56 elif kind is Stream.EXPR: | |
57 result = data.evaluate(ctxt) | |
58 if result is None: | |
59 continue | |
60 | |
61 # First check for a string, otherwise the iterable | |
62 # test below succeeds, and the string will be | |
63 # chopped up into characters | |
64 if isinstance(result, basestring): | |
65 yield Stream.TEXT, result, pos | |
66 else: | |
67 # Test if the expression evaluated to an | |
68 # iterable, in which case we yield the | |
69 # individual items | |
70 try: | |
71 yield Stream.SUB, ([], iter(result)), pos | |
72 except TypeError: | |
73 # Neither a string nor an iterable, so just | |
74 # pass it through | |
75 yield Stream.TEXT, unicode(result), pos | |
76 | |
77 else: | |
78 yield kind, data, pos | |
79 | |
80 | |
81 class IncludeFilter(object): | |
82 """Template filter providing (very) basic XInclude support | |
83 (see http://www.w3.org/TR/xinclude/) in templates. | |
84 """ | |
85 | |
86 _NAMESPACE = 'http://www.w3.org/2001/XInclude' | |
87 | |
88 def __init__(self, loader): | |
89 """Initialize the filter. | |
90 | |
91 @param loader: the `TemplateLoader` to use for resolving references to | |
92 external template files | |
93 """ | |
94 self.loader = loader | |
95 | |
96 def __call__(self, stream, ctxt=None): | |
97 """Filter the stream, processing any XInclude directives it may | |
98 contain. | |
99 | |
100 @param ctxt: the template context | |
101 @param stream: the markup event stream to filter | |
102 """ | |
103 from markup.template import TemplateError, TemplateNotFound | |
104 | |
105 in_fallback = False | |
106 include_href, fallback_stream = None, None | |
107 indent = 0 | |
108 | |
109 for kind, data, pos in stream: | |
110 | |
111 if kind is Stream.START and data[0].namespace == self._NAMESPACE \ | |
112 and not in_fallback: | |
113 tag, attrib = data | |
114 if tag.localname == 'include': | |
115 include_href = attrib.get('href') | |
116 indent = pos[1] | |
117 elif tag.localname == 'fallback': | |
118 in_fallback = True | |
119 fallback_stream = [] | |
120 | |
121 elif kind is Stream.END and data.namespace == self._NAMESPACE: | |
122 if data.localname == 'include': | |
123 try: | |
124 if not include_href: | |
125 raise TemplateError('Include misses required ' | |
126 'attribute "href"') | |
127 template = self.loader.load(include_href) | |
128 for ikind, idata, ipos in template.generate(ctxt): | |
129 # Fixup indentation of included markup | |
130 if ikind is Stream.TEXT: | |
131 idata = idata.replace('\n', '\n' + ' ' * indent) | |
132 yield ikind, idata, ipos | |
133 | |
134 # If the included template defines any filters added at | |
135 # runtime (such as py:match templates), those need to be | |
136 # applied to the including template, too. | |
137 for filter_ in template.filters: | |
138 stream = filter_(stream, ctxt) | |
139 | |
140 except TemplateNotFound: | |
141 if fallback_stream is None: | |
142 raise | |
143 for event in fallback_stream: | |
144 yield event | |
145 | |
146 include_href = None | |
147 fallback_stream = None | |
148 indent = 0 | |
149 break | |
150 elif data.localname == 'fallback': | |
151 in_fallback = False | |
152 | |
153 elif in_fallback: | |
154 fallback_stream.append((kind, data, pos)) | |
155 | |
156 elif kind is Stream.START_NS and data[1] == self._NAMESPACE: | |
157 continue | |
158 | |
159 else: | |
160 yield kind, data, pos | |
161 else: | |
162 # The loop exited normally, so there shouldn't be further events to | |
163 # process | |
164 return | |
165 | |
166 for event in self(stream, ctxt): | |
167 yield event | |
168 | |
169 | |
170 class MatchFilter(object): | |
171 """A filter that delegates to a given handler function when the input stream | |
172 matches some path expression. | |
173 """ | |
174 | |
175 def __init__(self, path, handler): | |
176 self.path = Path(path) | |
177 self.handler = handler | |
178 | |
179 def __call__(self, stream, ctxt=None): | |
180 test = self.path.test() | |
181 for kind, data, pos in stream: | |
182 result = test(kind, data, pos) | |
183 if result is True: | |
184 content = [(kind, data, pos)] | |
185 depth = 1 | |
186 while depth > 0: | |
187 ev = stream.next() | |
188 if ev[0] is Stream.START: | |
189 depth += 1 | |
190 elif ev[0] is Stream.END: | |
191 depth -= 1 | |
192 content.append(ev) | |
193 test(*ev) | |
194 | |
195 yield (Stream.SUB, | |
196 ([lambda stream, ctxt: self.handler(content, ctxt)], []), | |
197 pos) | |
198 else: | |
199 yield kind, data, pos | |
200 | |
201 | |
202 class WhitespaceFilter(object): | |
203 """A filter that removes extraneous white space from the stream. | |
204 | |
205 Todo: | |
206 * Support for xml:space | |
207 """ | |
208 | |
209 _TRAILING_SPACE = re.compile('[ \t]+(?=\n)') | |
210 _LINE_COLLAPSE = re.compile('\n{2,}') | |
211 | |
212 def __call__(self, stream, ctxt=None): | |
213 textbuf = [] | |
214 prev_kind = None | |
215 for kind, data, pos in stream: | |
216 if kind is Stream.TEXT: | |
217 textbuf.append(data) | |
218 elif prev_kind is Stream.TEXT: | |
219 text = ''.join(textbuf) | |
220 text = self._TRAILING_SPACE.sub('', text) | |
221 text = self._LINE_COLLAPSE.sub('\n', text) | |
222 yield Stream.TEXT, text, pos | |
223 del textbuf[:] | |
224 prev_kind = kind | |
225 if kind is not Stream.TEXT: | |
226 yield kind, data, pos | |
227 | |
228 if textbuf: | |
229 text = self._LINE_COLLAPSE.sub('\n', ''.join(textbuf)) | |
230 yield Stream.TEXT, text, pos | |
231 | |
232 | |
233 class HTMLSanitizer(object): | |
234 """A filter that removes potentially dangerous HTML tags and attributes | |
235 from the stream. | |
236 """ | |
237 | |
238 _SAFE_TAGS = frozenset(['a', 'abbr', 'acronym', 'address', 'area', 'b', | |
239 'big', 'blockquote', 'br', 'button', 'caption', 'center', 'cite', | |
240 'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', | |
241 'em', 'fieldset', 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', | |
242 'hr', 'i', 'img', 'input', 'ins', 'kbd', 'label', 'legend', 'li', 'map', | |
243 'menu', 'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp', | |
244 'select', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table', | |
245 'tbody', 'td', 'textarea', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u', | |
246 'ul', 'var']) | |
247 | |
248 _SAFE_ATTRS = frozenset(['abbr', 'accept', 'accept-charset', 'accesskey', | |
249 'action', 'align', 'alt', 'axis', 'border', 'cellpadding', | |
250 'cellspacing', 'char', 'charoff', 'charset', 'checked', 'cite', 'class', | |
251 'clear', 'cols', 'colspan', 'color', 'compact', 'coords', 'datetime', | |
252 'dir', 'disabled', 'enctype', 'for', 'frame', 'headers', 'height', | |
253 'href', 'hreflang', 'hspace', 'id', 'ismap', 'label', 'lang', | |
254 'longdesc', 'maxlength', 'media', 'method', 'multiple', 'name', | |
255 'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', 'rel', 'rev', | |
256 'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', | |
257 'span', 'src', 'start', 'style', 'summary', 'tabindex', 'target', | |
258 'title', 'type', 'usemap', 'valign', 'value', 'vspace', 'width']) | |
259 _URI_ATTRS = frozenset(['action', 'background', 'dynsrc', 'href', 'lowsrc', | |
260 'src']) | |
261 _SAFE_SCHEMES = frozenset(['file', 'ftp', 'http', 'https', 'mailto', None]) | |
262 | |
263 def __call__(self, stream, ctxt=None): | |
264 waiting_for = None | |
265 | |
266 for kind, data, pos in stream: | |
267 if kind is Stream.START: | |
268 if waiting_for: | |
269 continue | |
270 tag, attrib = data | |
271 if tag not in self._SAFE_TAGS: | |
272 waiting_for = tag | |
273 continue | |
274 | |
275 new_attrib = [] | |
276 for attr, value in attrib: | |
277 if attr not in self._SAFE_ATTRS: | |
278 continue | |
279 elif attr in self._URI_ATTRS: | |
280 # Don't allow URI schemes such as "javascript:" | |
281 if self._get_scheme(value) not in self._SAFE_SCHEMES: | |
282 continue | |
283 elif attr == 'style': | |
284 # Remove dangerous CSS declarations from inline styles | |
285 decls = [] | |
286 for decl in filter(None, value.split(';')): | |
287 is_evil = False | |
288 if 'expression' in decl: | |
289 is_evil = True | |
290 for m in re.finditer(r'url\s*\(([^)]+)', decl): | |
291 if self._get_scheme(m.group(1)) not in self._SAFE_SCHEMES: | |
292 is_evil = True | |
293 break | |
294 if not is_evil: | |
295 decls.append(decl.strip()) | |
296 if not decls: | |
297 continue | |
298 value = '; '.join(decls) | |
299 new_attrib.append((attr, value)) | |
300 | |
301 yield kind, (tag, new_attrib), pos | |
302 | |
303 elif kind is Stream.END: | |
304 tag = data | |
305 if waiting_for: | |
306 if waiting_for == tag: | |
307 waiting_for = None | |
308 else: | |
309 yield kind, data, pos | |
310 | |
311 else: | |
312 if not waiting_for: | |
313 yield kind, data, pos | |
314 | |
315 def _get_scheme(self, text): | |
316 if ':' not in text: | |
317 return None | |
318 chars = [char for char in text.split(':', 1)[0] if char.isalnum()] | |
319 return ''.join(chars).lower() |