"""Tests for HTMLParser.py."""
from test
import test_support
class EventCollector(HTMLParser
.HTMLParser
):
self
.append
= self
.events
.append
HTMLParser
.HTMLParser
.__init
__(self
)
# Normalize the list of events so that buffer artefacts don't
# separate runs of contiguous characters.
for event
in self
.events
:
if type == prevtype
== "data":
L
[-1] = ("data", L
[-1][1] + event
[1])
def handle_starttag(self
, tag
, attrs
):
self
.append(("starttag", tag
, attrs
))
def handle_startendtag(self
, tag
, attrs
):
self
.append(("startendtag", tag
, attrs
))
def handle_endtag(self
, tag
):
self
.append(("endtag", tag
))
def handle_comment(self
, data
):
self
.append(("comment", data
))
def handle_charref(self
, data
):
self
.append(("charref", data
))
def handle_data(self
, data
):
self
.append(("data", data
))
def handle_decl(self
, data
):
self
.append(("decl", data
))
def handle_entityref(self
, data
):
self
.append(("entityref", data
))
def handle_pi(self
, data
):
self
.append(("pi", data
))
def unknown_decl(self
, decl
):
self
.append(("unknown decl", decl
))
class EventCollectorExtra(EventCollector
):
def handle_starttag(self
, tag
, attrs
):
EventCollector
.handle_starttag(self
, tag
, attrs
)
self
.append(("starttag_text", self
.get_starttag_text()))
class TestCaseBase(unittest
.TestCase
):
def _run_check(self
, source
, expected_events
, collector
=EventCollector
):
events
= parser
.get_events()
if events
!= expected_events
:
self
.fail("received events did not match expected events\n"
"Expected:\n" + pprint
.pformat(expected_events
) +
"\nReceived:\n" + pprint
.pformat(events
))
def _run_check_extra(self
, source
, events
):
self
._run
_check
(source
, events
, EventCollectorExtra
)
def _parse_error(self
, source
):
def parse(source
=source
):
parser
= HTMLParser
.HTMLParser()
self
.assertRaises(HTMLParser
.HTMLParseError
, parse
)
class HTMLParserTestCase(TestCaseBase
):
def test_processing_instruction_only(self
):
self
._run
_check
("<?processing instruction>", [
("pi", "processing instruction"),
self
._run
_check
("<?processing instruction ?>", [
("pi", "processing instruction ?"),
def test_simple_html(self
):
<!DOCTYPE html PUBLIC 'foo'>
-></foo><bar><<?pi?></foo<bar
<Img sRc='Bar' isMAP>sample
<!--comment2a-- --comment2b-->
("decl", "DOCTYPE html PUBLIC 'foo'"),
("starttag", "html", []),
("comment", "comment1a\n-></foo><bar><<?pi?></foo<bar\ncomment1b"),
("starttag", "img", [("src", "Bar"), ("ismap", None)]),
("data", "sample\ntext\n"),
("comment", "comment2a-- --comment2b"),
def test_unclosed_entityref(self
):
self
._run
_check
("&entityref foo", [
("entityref", "entityref"),
def test_doctype_decl(self
):
<!ELEMENT html - O EMPTY>
profile CDATA 'DublinCore'>
<!NOTATION datatype SYSTEM 'http://xml.python.org/notations/python-module'>
<!ENTITY myEntity 'internal parsed entity'>
<!ENTITY anEntity SYSTEM 'http://xml.python.org/entities/something.xml'>
<!ENTITY % paramEntity 'name|name|name'>
self
._run
_check
("<!%s>" % inside
, [
def test_bad_nesting(self
):
# Strangely, this *is* supposed to test that overlapping
# elements are allowed. HTMLParser is more geared toward
# lexing the input that parsing the structure.
self
._run
_check
("<a><b></a></b>", [
def test_bare_ampersands(self
):
self
._run
_check
("this text & contains & ampersands &", [
("data", "this text & contains & ampersands &"),
def test_bare_pointy_brackets(self
):
self
._run
_check
("this < text > contains < bare>pointy< brackets", [
("data", "this < text > contains < bare>pointy< brackets"),
def test_attr_syntax(self
):
("starttag", "a", [("b", "v"), ("c", "v"), ("d", "v"), ("e", None)])
self
._run
_check
("""<a b='v' c="v" d=v e>""", output
)
self
._run
_check
("""<a b = 'v' c = "v" d = v e>""", output
)
self
._run
_check
("""<a\nb\n=\n'v'\nc\n=\n"v"\nd\n=\nv\ne>""", output
)
self
._run
_check
("""<a\tb\t=\t'v'\tc\t=\t"v"\td\t=\tv\te>""", output
)
def test_attr_values(self
):
self
._run
_check
("""<a b='xxx\n\txxx' c="yyy\t\nyyy" d='\txyz\n'>""",
[("starttag", "a", [("b", "xxx\n\txxx"),
self
._run
_check
("""<a b='' c="">""", [
("starttag", "a", [("b", ""), ("c", "")]),
# Regression test for SF patch #669683.
self
._run
_check
("<e a=rgb(1,2,3)>", [
("starttag", "e", [("a", "rgb(1,2,3)")]),
# Regression test for SF bug #921657.
self
._run
_check
("<a href=mailto:xyz@example.com>", [
("starttag", "a", [("href", "mailto:xyz@example.com")]),
def test_attr_entity_replacement(self
):
self
._run
_check
("""<a b='&><"''>""", [
("starttag", "a", [("b", "&><\"'")]),
def test_attr_funky_names(self
):
self
._run
_check
("""<a a.b='v' c:d=v e-f=v>""", [
("starttag", "a", [("a.b", "v"), ("c:d", "v"), ("e-f", "v")]),
def test_illegal_declarations(self
):
self
._parse
_error
('<!spacer type="block" height="25">')
def test_starttag_end_boundary(self
):
self
._run
_check
("""<a b='<'>""", [("starttag", "a", [("b", "<")])])
self
._run
_check
("""<a b='>'>""", [("starttag", "a", [("b", ">")])])
def test_buffer_artefacts(self
):
output
= [("starttag", "a", [("b", "<")])]
self
._run
_check
(["<a b='<'>"], output
)
self
._run
_check
(["<a ", "b='<'>"], output
)
self
._run
_check
(["<a b", "='<'>"], output
)
self
._run
_check
(["<a b=", "'<'>"], output
)
self
._run
_check
(["<a b='<", "'>"], output
)
self
._run
_check
(["<a b='<'", ">"], output
)
output
= [("starttag", "a", [("b", ">")])]
self
._run
_check
(["<a b='>'>"], output
)
self
._run
_check
(["<a ", "b='>'>"], output
)
self
._run
_check
(["<a b", "='>'>"], output
)
self
._run
_check
(["<a b=", "'>'>"], output
)
self
._run
_check
(["<a b='>", "'>"], output
)
self
._run
_check
(["<a b='>'", ">"], output
)
output
= [("comment", "abc")]
self
._run
_check
(["", "<!--abc-->"], output
)
self
._run
_check
(["<", "!--abc-->"], output
)
self
._run
_check
(["<!", "--abc-->"], output
)
self
._run
_check
(["<!-", "-abc-->"], output
)
self
._run
_check
(["<!--", "abc-->"], output
)
self
._run
_check
(["<!--a", "bc-->"], output
)
self
._run
_check
(["<!--ab", "c-->"], output
)
self
._run
_check
(["<!--abc", "-->"], output
)
self
._run
_check
(["<!--abc-", "->"], output
)
self
._run
_check
(["<!--abc--", ">"], output
)
self
._run
_check
(["<!--abc-->", ""], output
)
def test_starttag_junk_chars(self
):
self
._parse
_error
("</$>")
self
._parse
_error
("<a<a>")
self
._parse
_error
("</a<a>")
self
._parse
_error
("<a $>")
self
._parse
_error
("<a foo='bar'")
self
._parse
_error
("<a foo='bar")
self
._parse
_error
("<a foo='>'")
self
._parse
_error
("<a foo='>")
self
._parse
_error
("<a foo=>")
def test_declaration_junk_chars(self
):
self
._parse
_error
("<!DOCTYPE foo $ >")
def test_startendtag(self
):
self
._run
_check
("<p/>", [
("startendtag", "p", []),
self
._run
_check
("<p></p>", [
self
._run
_check
("<p><img src='foo' /></p>", [
("startendtag", "img", [("src", "foo")]),
def test_get_starttag_text(self
):
s
= """<foo:bar \n one="1"\ttwo=2 >"""
self
._run
_check
_extra
(s
, [
("starttag", "foo:bar", [("one", "1"), ("two", "2")]),
def test_cdata_content(self
):
s
= """<script> <!-- not a comment --> ¬-an-entity-ref; </script>"""
("starttag", "script", []),
("data", " <!-- not a comment --> ¬-an-entity-ref; "),
s
= """<script> <not a='start tag'> </script>"""
("starttag", "script", []),
("data", " <not a='start tag'> "),
test_support
.run_unittest(HTMLParserTestCase
)
if __name__
== "__main__":