Thanks to visit codestin.com
Credit goes to github.com

Skip to content

Commit f4ab491

Browse files
ezio-melottiezio-melotti
committed
Improve handling of declarations in HTMLParser.
1 parent 86f6712 commit f4ab491

2 files changed

Lines changed: 51 additions & 29 deletions

File tree

‎Lib/html/parser.py‎

Lines changed: 22 additions & 8 deletions
Original file line numberDiff line numberDiff line change
@@ -187,17 +187,10 @@ def goahead(self, end):
187187
elif startswith("<?", i):
188188
k = self.parse_pi(i)
189189
elif startswith("<!", i):
190-
# this might fail with things like <! not a comment > or
191-
# <! -- space before '--' -->. When strict is True an
192-
# error is raised, when it's False they will be considered
193-
# as bogus comments and parsed (see parse_bogus_comment).
194190
if self.strict:
195191
k = self.parse_declaration(i)
196192
else:
197-
try:
198-
k = self.parse_declaration(i)
199-
except HTMLParseError:
200-
k = self.parse_bogus_comment(i)
193+
k = self.parse_html_declaration(i)
201194
elif (i + 1) < n:
202195
self.handle_data("<")
203196
k = i + 1
@@ -269,6 +262,27 @@ def goahead(self, end):
269262
i = self.updatepos(i, n)
270263
self.rawdata = rawdata[i:]
271264

265+
# Internal -- parse html declarations, return length or -1 if not terminated
266+
# See w3.org/TR/html5/tokenization.html#markup-declaration-open-state
267+
# See also parse_declaration in _markupbase
268+
def parse_html_declaration(self, i):
269+
rawdata = self.rawdata
270+
if rawdata[i:i+2] != '<!':
271+
self.error('unexpected call to parse_html_declaration()')
272+
if rawdata[i:i+4] == '<!--':
273+
return self.parse_comment(i)
274+
elif rawdata[i:i+3] == '<![':
275+
return self.parse_marked_section(i)
276+
elif rawdata[i:i+9].lower() == '<!doctype':
277+
# find the closing >
278+
gtpos = rawdata.find('>', 9)
279+
if gtpos == -1:
280+
return -1
281+
self.handle_decl(rawdata[i+2:gtpos])
282+
return gtpos+1
283+
else:
284+
return self.parse_bogus_comment(i)
285+
272286
# Internal -- parse bogus comment, return length or -1 if not terminated
273287
# see http://www.w3.org/TR/html5/tokenization.html#bogus-comment-state
274288
def parse_bogus_comment(self, i, report=1):

‎Lib/test/test_htmlparser.py‎

Lines changed: 29 additions & 21 deletions
Original file line numberDiff line numberDiff line change
@@ -122,7 +122,7 @@ def test_simple_html(self):
122122
<Img sRc='Bar' isMAP>sample
123123
text
124124
&#x201C;
125-
<!--comment2a-- --comment2b--><!>
125+
<!--comment2a-- --comment2b-->
126126
</Html>
127127
""", [
128128
("data", "\n"),
@@ -157,24 +157,6 @@ def test_unclosed_entityref(self):
157157
("data", " foo"),
158158
])
159159

160-
def test_doctype_decl(self):
161-
inside = """\
162-
DOCTYPE html [
163-
<!ELEMENT html - O EMPTY>
164-
<!ATTLIST html
165-
version CDATA #IMPLIED
166-
profile CDATA 'DublinCore'>
167-
<!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
168-
<!ENTITY myEntity 'internal parsed entity'>
169-
<!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
170-
<!ENTITY % paramEntity 'name|name|name'>
171-
%paramEntity;
172-
<!-- comment -->
173-
]"""
174-
self._run_check("<!%s>" % inside, [
175-
("decl", inside),
176-
])
177-
178160
def test_bad_nesting(self):
179161
# Strangely, this *is* supposed to test that overlapping
180162
# elements are allowed. HTMLParser is more geared toward
@@ -247,6 +229,30 @@ def test_starttag_junk_chars(self):
247229
self._parse_error("<a foo='>'")
248230
self._parse_error("<a foo='>")
249231

232+
def test_valid_doctypes(self):
233+
# from http://www.w3.org/QA/2002/04/valid-dtd-list.html
234+
dtds = ['HTML', # HTML5 doctype
235+
('HTML PUBLIC "-//W3C//DTD HTML 4.01//EN" '
236+
'"http://www.w3.org/TR/html4/strict.dtd"'),
237+
('HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" '
238+
'"http://www.w3.org/TR/html4/loose.dtd"'),
239+
('html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" '
240+
'"http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd"'),
241+
('html PUBLIC "-//W3C//DTD XHTML 1.0 Frameset//EN" '
242+
'"http://www.w3.org/TR/xhtml1/DTD/xhtml1-frameset.dtd"'),
243+
('math PUBLIC "-//W3C//DTD MathML 2.0//EN" '
244+
'"http://www.w3.org/Math/DTD/mathml2/mathml2.dtd"'),
245+
('html PUBLIC "-//W3C//DTD '
246+
'XHTML 1.1 plus MathML 2.0 plus SVG 1.1//EN" '
247+
'"http://www.w3.org/2002/04/xhtml-math-svg/xhtml-math-svg.dtd"'),
248+
('svg PUBLIC "-//W3C//DTD SVG 1.1//EN" '
249+
'"http://www.w3.org/Graphics/SVG/1.1/DTD/svg11.dtd"'),
250+
'html PUBLIC "-//IETF//DTD HTML 2.0//EN"',
251+
'html PUBLIC "-//W3C//DTD HTML 3.2 Final//EN"']
252+
for dtd in dtds:
253+
self._run_check("<!DOCTYPE %s>" % dtd,
254+
[('decl', 'DOCTYPE ' + dtd)])
255+
250256
def test_declaration_junk_chars(self):
251257
self._parse_error("<!DOCTYPE foo $ >")
252258

@@ -384,8 +390,7 @@ def test_starttag_junk_chars(self):
384390
self._run_check("<a foo='>", [('data', "<a foo='>")])
385391

386392
def test_declaration_junk_chars(self):
387-
# XXX this is wrong
388-
self._run_check("<!DOCTYPE foo $ >", [('comment', 'DOCTYPE foo $ ')])
393+
self._run_check("<!DOCTYPE foo $ >", [('decl', 'DOCTYPE foo $ ')])
389394

390395
def test_illegal_declarations(self):
391396
# XXX this might be wrong
@@ -510,11 +515,14 @@ def test_broken_comments(self):
510515
html = ('<! not really a comment >'
511516
'<! not a comment either -->'
512517
'<! -- close enough -->'
518+
'<!><!<-- this was an empty comment>'
513519
'<!!! another bogus comment !!!>')
514520
expected = [
515521
('comment', ' not really a comment '),
516522
('comment', ' not a comment either --'),
517523
('comment', ' -- close enough --'),
524+
('comment', ''),
525+
('comment', '<-- this was an empty comment'),
518526
('comment', '!! another bogus comment !!!'),
519527
]
520528
self._run_check(html, expected)

0 commit comments

Comments
 (0)