Mercurial > genshi > genshi-test
annotate markup/filters.py @ 10:c5890ef863ba
Moved the template-specific stream event kinds into the template module.
| author | cmlenz |
|---|---|
| date | Sun, 04 Jun 2006 12:34:17 +0000 |
| parents | 821114ec4f69 |
| children | 87238328a71d |
| rev | line source |
|---|---|
| 1 | 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): | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
33 from markup.template import Template |
|
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
34 |
| 1 | 35 for kind, data, pos in stream: |
| 36 | |
| 37 if kind is Stream.START: | |
| 38 # Attributes may still contain expressions in start tags at | |
| 39 # this point, so do some evaluation | |
| 40 tag, attrib = data | |
| 41 new_attrib = [] | |
| 42 for name, substream in attrib: | |
| 43 if isinstance(substream, basestring): | |
| 44 value = substream | |
| 45 else: | |
| 46 values = [] | |
| 47 for subkind, subdata, subpos in substream: | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
48 if subkind is Template.EXPR: |
| 1 | 49 values.append(subdata.evaluate(ctxt)) |
| 50 else: | |
| 51 values.append(subdata) | |
| 52 value = filter(lambda x: x is not None, values) | |
| 53 if not value: | |
| 54 continue | |
| 55 new_attrib.append((name, ''.join(value))) | |
| 56 yield kind, (tag, Attributes(new_attrib)), pos | |
| 57 | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
58 elif kind is Template.EXPR: |
| 1 | 59 result = data.evaluate(ctxt) |
| 60 if result is None: | |
| 61 continue | |
| 62 | |
| 63 # First check for a string, otherwise the iterable | |
| 64 # test below succeeds, and the string will be | |
| 65 # chopped up into characters | |
| 66 if isinstance(result, basestring): | |
| 67 yield Stream.TEXT, result, pos | |
| 68 else: | |
| 69 # Test if the expression evaluated to an | |
| 70 # iterable, in which case we yield the | |
| 71 # individual items | |
| 72 try: | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
73 yield Template.SUB, ([], iter(result)), pos |
| 1 | 74 except TypeError: |
| 75 # Neither a string nor an iterable, so just | |
| 76 # pass it through | |
| 77 yield Stream.TEXT, unicode(result), pos | |
| 78 | |
| 79 else: | |
| 80 yield kind, data, pos | |
| 81 | |
| 82 | |
| 83 class IncludeFilter(object): | |
| 84 """Template filter providing (very) basic XInclude support | |
| 85 (see http://www.w3.org/TR/xinclude/) in templates. | |
| 86 """ | |
| 87 | |
| 88 _NAMESPACE = 'http://www.w3.org/2001/XInclude' | |
| 89 | |
| 90 def __init__(self, loader): | |
| 91 """Initialize the filter. | |
| 92 | |
| 93 @param loader: the `TemplateLoader` to use for resolving references to | |
| 94 external template files | |
| 95 """ | |
| 96 self.loader = loader | |
| 97 | |
| 98 def __call__(self, stream, ctxt=None): | |
| 99 """Filter the stream, processing any XInclude directives it may | |
| 100 contain. | |
| 101 | |
| 102 @param ctxt: the template context | |
| 103 @param stream: the markup event stream to filter | |
| 104 """ | |
| 105 from markup.template import TemplateError, TemplateNotFound | |
| 106 | |
| 107 in_fallback = False | |
| 108 include_href, fallback_stream = None, None | |
| 109 indent = 0 | |
| 110 | |
| 111 for kind, data, pos in stream: | |
| 112 | |
| 113 if kind is Stream.START and data[0].namespace == self._NAMESPACE \ | |
| 114 and not in_fallback: | |
| 115 tag, attrib = data | |
| 116 if tag.localname == 'include': | |
| 117 include_href = attrib.get('href') | |
| 118 indent = pos[1] | |
| 119 elif tag.localname == 'fallback': | |
| 120 in_fallback = True | |
| 121 fallback_stream = [] | |
| 122 | |
| 123 elif kind is Stream.END and data.namespace == self._NAMESPACE: | |
| 124 if data.localname == 'include': | |
| 125 try: | |
| 126 if not include_href: | |
| 127 raise TemplateError('Include misses required ' | |
| 128 'attribute "href"') | |
| 129 template = self.loader.load(include_href) | |
| 130 for ikind, idata, ipos in template.generate(ctxt): | |
| 131 # Fixup indentation of included markup | |
| 132 if ikind is Stream.TEXT: | |
| 133 idata = idata.replace('\n', '\n' + ' ' * indent) | |
| 134 yield ikind, idata, ipos | |
| 135 | |
| 136 # If the included template defines any filters added at | |
| 137 # runtime (such as py:match templates), those need to be | |
| 138 # applied to the including template, too. | |
| 139 for filter_ in template.filters: | |
| 140 stream = filter_(stream, ctxt) | |
| 141 | |
| 142 except TemplateNotFound: | |
| 143 if fallback_stream is None: | |
| 144 raise | |
| 145 for event in fallback_stream: | |
| 146 yield event | |
| 147 | |
| 148 include_href = None | |
| 149 fallback_stream = None | |
| 150 indent = 0 | |
| 151 break | |
| 152 elif data.localname == 'fallback': | |
| 153 in_fallback = False | |
| 154 | |
| 155 elif in_fallback: | |
| 156 fallback_stream.append((kind, data, pos)) | |
| 157 | |
| 158 elif kind is Stream.START_NS and data[1] == self._NAMESPACE: | |
| 159 continue | |
| 160 | |
| 161 else: | |
| 162 yield kind, data, pos | |
| 163 else: | |
| 164 # The loop exited normally, so there shouldn't be further events to | |
| 165 # process | |
| 166 return | |
| 167 | |
| 168 for event in self(stream, ctxt): | |
| 169 yield event | |
| 170 | |
| 171 | |
| 172 class MatchFilter(object): | |
| 173 """A filter that delegates to a given handler function when the input stream | |
| 174 matches some path expression. | |
| 175 """ | |
| 176 | |
| 177 def __init__(self, path, handler): | |
| 178 self.path = Path(path) | |
| 179 self.handler = handler | |
| 180 | |
| 181 def __call__(self, stream, ctxt=None): | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
182 from markup.template import Template |
|
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
183 |
| 1 | 184 test = self.path.test() |
| 185 for kind, data, pos in stream: | |
| 186 result = test(kind, data, pos) | |
| 187 if result is True: | |
| 188 content = [(kind, data, pos)] | |
| 189 depth = 1 | |
| 190 while depth > 0: | |
| 191 ev = stream.next() | |
| 192 if ev[0] is Stream.START: | |
| 193 depth += 1 | |
| 194 elif ev[0] is Stream.END: | |
| 195 depth -= 1 | |
| 196 content.append(ev) | |
| 197 test(*ev) | |
| 198 | |
|
10
c5890ef863ba
Moved the template-specific stream event kinds into the template module.
cmlenz
parents:
1
diff
changeset
|
199 yield (Template.SUB, |
| 1 | 200 ([lambda stream, ctxt: self.handler(content, ctxt)], []), |
| 201 pos) | |
| 202 else: | |
| 203 yield kind, data, pos | |
| 204 | |
| 205 | |
| 206 class WhitespaceFilter(object): | |
| 207 """A filter that removes extraneous white space from the stream. | |
| 208 | |
| 209 Todo: | |
| 210 * Support for xml:space | |
| 211 """ | |
| 212 | |
| 213 _TRAILING_SPACE = re.compile('[ \t]+(?=\n)') | |
| 214 _LINE_COLLAPSE = re.compile('\n{2,}') | |
| 215 | |
| 216 def __call__(self, stream, ctxt=None): | |
| 217 textbuf = [] | |
| 218 prev_kind = None | |
| 219 for kind, data, pos in stream: | |
| 220 if kind is Stream.TEXT: | |
| 221 textbuf.append(data) | |
| 222 elif prev_kind is Stream.TEXT: | |
| 223 text = ''.join(textbuf) | |
| 224 text = self._TRAILING_SPACE.sub('', text) | |
| 225 text = self._LINE_COLLAPSE.sub('\n', text) | |
| 226 yield Stream.TEXT, text, pos | |
| 227 del textbuf[:] | |
| 228 prev_kind = kind | |
| 229 if kind is not Stream.TEXT: | |
| 230 yield kind, data, pos | |
| 231 | |
| 232 if textbuf: | |
| 233 text = self._LINE_COLLAPSE.sub('\n', ''.join(textbuf)) | |
| 234 yield Stream.TEXT, text, pos | |
| 235 | |
| 236 | |
| 237 class HTMLSanitizer(object): | |
| 238 """A filter that removes potentially dangerous HTML tags and attributes | |
| 239 from the stream. | |
| 240 """ | |
| 241 | |
| 242 _SAFE_TAGS = frozenset(['a', 'abbr', 'acronym', 'address', 'area', 'b', | |
| 243 'big', 'blockquote', 'br', 'button', 'caption', 'center', 'cite', | |
| 244 'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', | |
| 245 'em', 'fieldset', 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', | |
| 246 'hr', 'i', 'img', 'input', 'ins', 'kbd', 'label', 'legend', 'li', 'map', | |
| 247 'menu', 'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp', | |
| 248 'select', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table', | |
| 249 'tbody', 'td', 'textarea', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u', | |
| 250 'ul', 'var']) | |
| 251 | |
| 252 _SAFE_ATTRS = frozenset(['abbr', 'accept', 'accept-charset', 'accesskey', | |
| 253 'action', 'align', 'alt', 'axis', 'border', 'cellpadding', | |
| 254 'cellspacing', 'char', 'charoff', 'charset', 'checked', 'cite', 'class', | |
| 255 'clear', 'cols', 'colspan', 'color', 'compact', 'coords', 'datetime', | |
| 256 'dir', 'disabled', 'enctype', 'for', 'frame', 'headers', 'height', | |
| 257 'href', 'hreflang', 'hspace', 'id', 'ismap', 'label', 'lang', | |
| 258 'longdesc', 'maxlength', 'media', 'method', 'multiple', 'name', | |
| 259 'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', 'rel', 'rev', | |
| 260 'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', | |
| 261 'span', 'src', 'start', 'style', 'summary', 'tabindex', 'target', | |
| 262 'title', 'type', 'usemap', 'valign', 'value', 'vspace', 'width']) | |
| 263 _URI_ATTRS = frozenset(['action', 'background', 'dynsrc', 'href', 'lowsrc', | |
| 264 'src']) | |
| 265 _SAFE_SCHEMES = frozenset(['file', 'ftp', 'http', 'https', 'mailto', None]) | |
| 266 | |
| 267 def __call__(self, stream, ctxt=None): | |
| 268 waiting_for = None | |
| 269 | |
| 270 for kind, data, pos in stream: | |
| 271 if kind is Stream.START: | |
| 272 if waiting_for: | |
| 273 continue | |
| 274 tag, attrib = data | |
| 275 if tag not in self._SAFE_TAGS: | |
| 276 waiting_for = tag | |
| 277 continue | |
| 278 | |
| 279 new_attrib = [] | |
| 280 for attr, value in attrib: | |
| 281 if attr not in self._SAFE_ATTRS: | |
| 282 continue | |
| 283 elif attr in self._URI_ATTRS: | |
| 284 # Don't allow URI schemes such as "javascript:" | |
| 285 if self._get_scheme(value) not in self._SAFE_SCHEMES: | |
| 286 continue | |
| 287 elif attr == 'style': | |
| 288 # Remove dangerous CSS declarations from inline styles | |
| 289 decls = [] | |
| 290 for decl in filter(None, value.split(';')): | |
| 291 is_evil = False | |
| 292 if 'expression' in decl: | |
| 293 is_evil = True | |
| 294 for m in re.finditer(r'url\s*\(([^)]+)', decl): | |
| 295 if self._get_scheme(m.group(1)) not in self._SAFE_SCHEMES: | |
| 296 is_evil = True | |
| 297 break | |
| 298 if not is_evil: | |
| 299 decls.append(decl.strip()) | |
| 300 if not decls: | |
| 301 continue | |
| 302 value = '; '.join(decls) | |
| 303 new_attrib.append((attr, value)) | |
| 304 | |
| 305 yield kind, (tag, new_attrib), pos | |
| 306 | |
| 307 elif kind is Stream.END: | |
| 308 tag = data | |
| 309 if waiting_for: | |
| 310 if waiting_for == tag: | |
| 311 waiting_for = None | |
| 312 else: | |
| 313 yield kind, data, pos | |
| 314 | |
| 315 else: | |
| 316 if not waiting_for: | |
| 317 yield kind, data, pos | |
| 318 | |
| 319 def _get_scheme(self, text): | |
| 320 if ':' not in text: | |
| 321 return None | |
| 322 chars = [char for char in text.split(':', 1)[0] if char.isalnum()] | |
| 323 return ''.join(chars).lower() |