Mercurial > genshi > mirror
annotate genshi/filters.py @ 415:b9f9a22484f0 trunk
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
author | cmlenz |
---|---|
date | Mon, 12 Mar 2007 12:29:17 +0000 |
parents | 4675d5cf6c67 |
children | 073640758a42 |
rev | line source |
---|---|
1 | 1 # -*- coding: utf-8 -*- |
2 # | |
408 | 3 # Copyright (C) 2006-2007 Edgewall Software |
1 | 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 | |
230 | 8 # are also available at http://genshi.edgewall.org/wiki/License. |
1 | 9 # |
10 # This software consists of voluntary contributions made by many | |
11 # individuals. For the exact contribution history, see the revision | |
230 | 12 # history and logs, available at http://genshi.edgewall.org/log/. |
1 | 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 | |
403
228907abb726
Remove some magic/overhead from `Attrs` creation and manipulation by not automatically wrapping attribute names in `QName`.
cmlenz
parents:
363
diff
changeset
|
22 from genshi.core import Attrs, QName, stripentities |
363
37e4b4bb0b53
Parse template includes at parse time to avoid some runtime overhead.
cmlenz
parents:
345
diff
changeset
|
23 from genshi.core import END, START, TEXT |
1 | 24 |
363
37e4b4bb0b53
Parse template includes at parse time to avoid some runtime overhead.
cmlenz
parents:
345
diff
changeset
|
25 __all__ = ['HTMLFormFiller', 'HTMLSanitizer'] |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
26 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
27 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
28 class HTMLFormFiller(object): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
29 """A stream filter that can populate HTML forms from a dictionary of values. |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
30 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
31 >>> from genshi.input import HTML |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
32 >>> html = HTML('''<form> |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
33 ... <p><input type="text" name="foo" /></p> |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
34 ... </form>''') |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
35 >>> filler = HTMLFormFiller(data={'foo': 'bar'}) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
36 >>> print html | filler |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
37 <form> |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
38 <p><input type="text" name="foo" value="bar"/></p> |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
39 </form> |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
40 """ |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
41 # TODO: only select the first radio button, and the first select option |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
42 # (if not in a multiple-select) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
43 # TODO: only apply to elements in the XHTML namespace (or no namespace)? |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
44 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
45 def __init__(self, name=None, id=None, data=None): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
46 """Create the filter. |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
47 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
48 @param name: The name of the form that should be populated. If this |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
49 parameter is given, only forms where the ``name`` attribute value |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
50 matches the parameter are processed. |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
51 @param id: The ID of the form that should be populated. If this |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
52 parameter is given, only forms where the ``id`` attribute value |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
53 matches the parameter are processed. |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
54 @param data: The dictionary of form values, where the keys are the names |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
55 of the form fields, and the values are the values to fill in. |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
56 """ |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
57 self.name = name |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
58 self.id = id |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
59 if data is None: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
60 data = {} |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
61 self.data = data |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
62 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
63 def __call__(self, stream, ctxt=None): |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
64 """Apply the filter to the given stream. |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
65 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
66 @param stream: the markup event stream to filter |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
67 @param ctxt: the template context (unused) |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
68 """ |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
69 in_form = in_select = in_option = in_textarea = False |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
70 select_value = option_value = textarea_value = None |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
71 option_start = option_text = None |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
72 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
73 for kind, data, pos in stream: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
74 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
75 if kind is START: |
345 | 76 tag, attrs = data |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
77 tagname = tag.localname |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
78 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
79 if tagname == 'form' and ( |
345 | 80 self.name and attrs.get('name') == self.name or |
81 self.id and attrs.get('id') == self.id or | |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
82 not (self.id or self.name)): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
83 in_form = True |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
84 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
85 elif in_form: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
86 if tagname == 'input': |
345 | 87 type = attrs.get('type') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
88 if type in ('checkbox', 'radio'): |
345 | 89 name = attrs.get('name') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
90 if name: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
91 value = self.data.get(name) |
345 | 92 declval = attrs.get('value') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
93 checked = False |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
94 if isinstance(value, (list, tuple)): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
95 if declval: |
415
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
96 checked = declval in [str(v) for v |
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
97 in value] |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
98 else: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
99 checked = bool(filter(None, value)) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
100 else: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
101 if declval: |
415
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
102 checked = declval == str(value) |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
103 elif type == 'checkbox': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
104 checked = bool(value) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
105 if checked: |
403
228907abb726
Remove some magic/overhead from `Attrs` creation and manipulation by not automatically wrapping attribute names in `QName`.
cmlenz
parents:
363
diff
changeset
|
106 attrs |= [(QName('checked'), 'checked')] |
345 | 107 elif 'checked' in attrs: |
108 attrs -= 'checked' | |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
109 elif type in (None, 'hidden', 'text'): |
345 | 110 name = attrs.get('name') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
111 if name: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
112 value = self.data.get(name) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
113 if isinstance(value, (list, tuple)): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
114 value = value[0] |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
115 if value is not None: |
403
228907abb726
Remove some magic/overhead from `Attrs` creation and manipulation by not automatically wrapping attribute names in `QName`.
cmlenz
parents:
363
diff
changeset
|
116 attrs |= [(QName('value'), unicode(value))] |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
117 elif tagname == 'select': |
345 | 118 name = attrs.get('name') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
119 select_value = self.data.get(name) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
120 in_select = True |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
121 elif tagname == 'textarea': |
345 | 122 name = attrs.get('name') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
123 textarea_value = self.data.get(name) |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
124 if isinstance(textarea_value, (list, tuple)): |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
125 textarea_value = textarea_value[0] |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
126 in_textarea = True |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
127 elif in_select and tagname == 'option': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
128 option_start = kind, data, pos |
345 | 129 option_value = attrs.get('value') |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
130 in_option = True |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
131 continue |
345 | 132 yield kind, (tag, attrs), pos |
133 | |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
134 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
135 elif in_form and kind is TEXT: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
136 if in_select and in_option: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
137 if option_value is None: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
138 option_value = data |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
139 option_text = kind, data, pos |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
140 continue |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
141 elif in_textarea: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
142 continue |
345 | 143 yield kind, data, pos |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
144 |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
145 elif in_form and kind is END: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
146 tagname = data.localname |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
147 if tagname == 'form': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
148 in_form = False |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
149 elif tagname == 'select': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
150 in_select = False |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
151 select_value = None |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
152 elif in_select and tagname == 'option': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
153 if isinstance(select_value, (tuple, list)): |
415
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
154 selected = option_value in [str(v) for v |
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
155 in select_value] |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
156 else: |
415
b9f9a22484f0
`HTMLFormFiller` now correctly deals with non-string values in the data dictionary for select/checkbox/radio controls.
cmlenz
parents:
408
diff
changeset
|
157 selected = option_value == str(select_value) |
345 | 158 okind, (tag, attrs), opos = option_start |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
159 if selected: |
403
228907abb726
Remove some magic/overhead from `Attrs` creation and manipulation by not automatically wrapping attribute names in `QName`.
cmlenz
parents:
363
diff
changeset
|
160 attrs |= [(QName('selected'), 'selected')] |
345 | 161 elif 'selected' in attrs: |
162 attrs -= 'selected' | |
163 yield okind, (tag, attrs), opos | |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
164 if option_text: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
165 yield option_text |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
166 in_option = False |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
167 option_start = option_text = option_value = None |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
168 elif tagname == 'textarea': |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
169 if textarea_value: |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
170 yield TEXT, unicode(textarea_value), pos |
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
171 in_textarea = False |
345 | 172 yield kind, data, pos |
275
d91cbdeb75e9
Integrated `HTMLFormFiller` filter initially presented as a [wiki:FormFilling#Usingatemplatefilter recipe].
cmlenz
parents:
230
diff
changeset
|
173 |
345 | 174 else: |
175 yield kind, data, pos | |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
176 |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
177 |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
178 class HTMLSanitizer(object): |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
179 """A filter that removes potentially dangerous HTML tags and attributes |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
180 from the stream. |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
181 """ |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
182 |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
183 SAFE_TAGS = frozenset(['a', 'abbr', 'acronym', 'address', 'area', 'b', |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
184 'big', 'blockquote', 'br', 'button', 'caption', 'center', 'cite', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
185 'code', 'col', 'colgroup', 'dd', 'del', 'dfn', 'dir', 'div', 'dl', 'dt', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
186 'em', 'fieldset', 'font', 'form', 'h1', 'h2', 'h3', 'h4', 'h5', 'h6', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
187 'hr', 'i', 'img', 'input', 'ins', 'kbd', 'label', 'legend', 'li', 'map', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
188 'menu', 'ol', 'optgroup', 'option', 'p', 'pre', 'q', 's', 'samp', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
189 'select', 'small', 'span', 'strike', 'strong', 'sub', 'sup', 'table', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
190 'tbody', 'td', 'textarea', 'tfoot', 'th', 'thead', 'tr', 'tt', 'u', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
191 'ul', 'var']) |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
192 |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
193 SAFE_ATTRS = frozenset(['abbr', 'accept', 'accept-charset', 'accesskey', |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
194 'action', 'align', 'alt', 'axis', 'bgcolor', 'border', 'cellpadding', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
195 'cellspacing', 'char', 'charoff', 'charset', 'checked', 'cite', 'class', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
196 'clear', 'cols', 'colspan', 'color', 'compact', 'coords', 'datetime', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
197 'dir', 'disabled', 'enctype', 'for', 'frame', 'headers', 'height', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
198 'href', 'hreflang', 'hspace', 'id', 'ismap', 'label', 'lang', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
199 'longdesc', 'maxlength', 'media', 'method', 'multiple', 'name', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
200 'nohref', 'noshade', 'nowrap', 'prompt', 'readonly', 'rel', 'rev', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
201 'rows', 'rowspan', 'rules', 'scope', 'selected', 'shape', 'size', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
202 'span', 'src', 'start', 'style', 'summary', 'tabindex', 'target', |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
203 'title', 'type', 'usemap', 'valign', 'value', 'vspace', 'width']) |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
204 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
205 SAFE_SCHEMES = frozenset(['file', 'ftp', 'http', 'https', 'mailto', None]) |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
206 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
207 URI_ATTRS = frozenset(['action', 'background', 'dynsrc', 'href', 'lowsrc', |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
208 'src']) |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
209 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
210 def __init__(self, safe_tags=SAFE_TAGS, safe_attrs=SAFE_ATTRS, |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
211 safe_schemes=SAFE_SCHEMES, uri_attrs=URI_ATTRS): |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
212 """Create the sanitizer. |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
213 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
214 The exact set of allowed elements and attributes can be configured. |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
215 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
216 @param safe_tags: a set of tag names that are considered safe |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
217 @param safe_attrs: a set of attribute names that are considered safe |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
218 @param safe_schemes: a set of URI schemes that are considered safe |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
219 @param uri_attrs: a set of names of attributes that contain URIs |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
220 """ |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
221 self.safe_tags = safe_tags |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
222 self.safe_attrs = safe_attrs |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
223 self.uri_attrs = uri_attrs |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
224 self.safe_schemes = safe_schemes |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
225 |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
226 def __call__(self, stream, ctxt=None): |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
227 """Apply the filter to the given stream. |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
228 |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
229 @param stream: the markup event stream to filter |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
230 @param ctxt: the template context (unused) |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
231 """ |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
232 waiting_for = None |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
233 |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
234 def _get_scheme(href): |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
235 if ':' not in href: |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
236 return None |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
237 chars = [char for char in href.split(':', 1)[0] if char.isalnum()] |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
238 return ''.join(chars).lower() |
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
239 |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
240 for kind, data, pos in stream: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
241 if kind is START: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
242 if waiting_for: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
243 continue |
345 | 244 tag, attrs = data |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
245 if tag not in self.safe_tags: |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
246 waiting_for = tag |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
247 continue |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
248 |
345 | 249 new_attrs = [] |
250 for attr, value in attrs: | |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
251 value = stripentities(value) |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
252 if attr not in self.safe_attrs: |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
253 continue |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
254 elif attr in self.uri_attrs: |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
255 # Don't allow URI schemes such as "javascript:" |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
256 if _get_scheme(value) not in self.safe_schemes: |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
257 continue |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
258 elif attr == 'style': |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
259 # Remove dangerous CSS declarations from inline styles |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
260 decls = [] |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
261 for decl in filter(None, value.split(';')): |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
262 is_evil = False |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
263 if 'expression' in decl: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
264 is_evil = True |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
265 for m in re.finditer(r'url\s*\(([^)]+)', decl): |
277
7e30bfa966ab
The `HTMLSanitizer` now lets you override the default set of tag and attribute names that are considered safe.
cmlenz
parents:
275
diff
changeset
|
266 if _get_scheme(m.group(1)) not in self.safe_schemes: |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
267 is_evil = True |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
268 break |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
269 if not is_evil: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
270 decls.append(decl.strip()) |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
271 if not decls: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
272 continue |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
273 value = '; '.join(decls) |
345 | 274 new_attrs.append((attr, value)) |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
275 |
345 | 276 yield kind, (tag, Attrs(new_attrs)), pos |
123
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
277 |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
278 elif kind is END: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
279 tag = data |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
280 if waiting_for: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
281 if waiting_for == tag: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
282 waiting_for = None |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
283 else: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
284 yield kind, data, pos |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
285 |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
286 else: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
287 if not waiting_for: |
10279d2eeec9
Fix for #18: whitespace in space-sensitive elements such as `<pre>` and `<textarea>` is now preserved.
cmlenz
parents:
113
diff
changeset
|
288 yield kind, data, pos |