| """Tests for HTMLParser.py.""" |
| |
| import HTMLParser |
| import pprint |
| import unittest |
| from test import test_support |
| |
| |
| class EventCollector(HTMLParser.HTMLParser): |
| |
| def __init__(self): |
| self.events = [] |
| self.append = self.events.append |
| HTMLParser.HTMLParser.__init__(self) |
| |
| def get_events(self): |
| # Normalize the list of events so that buffer artefacts don't |
| # separate runs of contiguous characters. |
| L = [] |
| prevtype = None |
| for event in self.events: |
| type = event[0] |
| if type == prevtype == "data": |
| L[-1] = ("data", L[-1][1] + event[1]) |
| else: |
| L.append(event) |
| prevtype = type |
| self.events = L |
| return L |
| |
| # structure markup |
| |
| 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)) |
| |
| # all other markup |
| |
| 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): |
| parser = collector() |
| for s in source: |
| parser.feed(s) |
| parser.close() |
| 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() |
| parser.feed(source) |
| parser.close() |
| 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): |
| self._run_check(""" |
| <!DOCTYPE html PUBLIC 'foo'> |
| <HTML>&entity;  |
| <!--comment1a |
| -></foo><bar><<?pi?></foo<bar |
| comment1b--> |
| <Img sRc='Bar' isMAP>sample |
| text |
| “ |
| <!--comment2a-- --comment2b--><!> |
| </Html> |
| """, [ |
| ("data", "\n"), |
| ("decl", "DOCTYPE html PUBLIC 'foo'"), |
| ("data", "\n"), |
| ("starttag", "html", []), |
| ("entityref", "entity"), |
| ("charref", "32"), |
| ("data", "\n"), |
| ("comment", "comment1a\n-></foo><bar><<?pi?></foo<bar\ncomment1b"), |
| ("data", "\n"), |
| ("starttag", "img", [("src", "Bar"), ("ismap", None)]), |
| ("data", "sample\ntext\n"), |
| ("charref", "x201C"), |
| ("data", "\n"), |
| ("comment", "comment2a-- --comment2b"), |
| ("data", "\n"), |
| ("endtag", "html"), |
| ("data", "\n"), |
| ]) |
| |
| def test_unclosed_entityref(self): |
| self._run_check("&entityref foo", [ |
| ("entityref", "entityref"), |
| ("data", " foo"), |
| ]) |
| |
| def test_doctype_decl(self): |
| inside = """\ |
| DOCTYPE html [ |
| <!ELEMENT html - O EMPTY> |
| <!ATTLIST html |
| version CDATA #IMPLIED |
| 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'> |
| %paramEntity; |
| <!-- comment --> |
| ]""" |
| self._run_check("<!%s>" % inside, [ |
| ("decl", 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>", [ |
| ("starttag", "a", []), |
| ("starttag", "b", []), |
| ("endtag", "a"), |
| ("endtag", "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): |
| output = [ |
| ("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"), |
| ("c", "yyy\t\nyyy"), |
| ("d", "\txyz\n")]) |
| ]) |
| 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("</$>") |
| self._parse_error("</") |
| self._parse_error("</a") |
| self._parse_error("<a<a>") |
| self._parse_error("</a<a>") |
| self._parse_error("<!") |
| self._parse_error("<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>", [ |
| ("starttag", "p", []), |
| ("endtag", "p"), |
| ]) |
| self._run_check("<p><img src='foo' /></p>", [ |
| ("starttag", "p", []), |
| ("startendtag", "img", [("src", "foo")]), |
| ("endtag", "p"), |
| ]) |
| |
| 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")]), |
| ("starttag_text", s)]) |
| |
| def test_cdata_content(self): |
| s = """<script> <!-- not a comment --> ¬-an-entity-ref; </script>""" |
| self._run_check(s, [ |
| ("starttag", "script", []), |
| ("data", " <!-- not a comment --> ¬-an-entity-ref; "), |
| ("endtag", "script"), |
| ]) |
| s = """<script> <not a='start tag'> </script>""" |
| self._run_check(s, [ |
| ("starttag", "script", []), |
| ("data", " <not a='start tag'> "), |
| ("endtag", "script"), |
| ]) |
| |
| def test_entityrefs_in_attributes(self): |
| self._run_check("<html foo='€&aa&unsupported;'>", [ |
| ("starttag", "html", [("foo", "\u20AC&aa&unsupported;")]) |
| ]) |
| |
| |
| def test_main(): |
| test_support.run_unittest(HTMLParserTestCase) |
| |
| |
| if __name__ == "__main__": |
| test_main() |