fix(_shadow): use _content_span to strip head and tail

fixes #126

CHANGELOG.rst

@@ -1,5 +1,6 @@
 Unreleased
 ----------
+- Fixed a bug in ``plain_text``. (#126)
 - Fixed another bug in parsing tables that end without a ``|}`` mark. (#125)
 
 v0.55.6

tests/wikitext/test_get_bolds_and_italics.py

@@ -1,36 +1,42 @@
 from wikitextparser import Bold, Italic, parse
 
 
+def assert_bold(
+    input_string: str, expected_bold_string: str, recursive: bool = True, /
+):
+    assert (
+        parse(input_string).get_bolds(recursive)[0].string
+        == expected_bold_string
+    )
+
+
+def assert_no_bold(input_string: str):
+    assert not parse(input_string).get_bolds(True)
+
+
 def test_get_bolds():
-    def ab(s: str, o: str, r: bool = True):
-        assert parse(s).get_bolds(r)[0].string == o
-
-    def anb(s: str):
-        assert not parse(s).get_bolds(True)
-
-    ab("A''''''''''B", "'''B")
-    ab("''''''a''''''", "'''a''''")  # '<i><b>a'</b></i>
-    ab("a'''<!--b-->'''BI", "'''BI")
-    ab("'''b'''", "'''b'''")
-    anb("''i1'''s")
-    anb("<!--'''b'''-->")
-    ab(
+    assert_bold("A''''''''''B", "'''B")
+    assert_bold("''''''a''''''", "'''a''''")  # '<i><b>a'</b></i>
+    assert_bold("a'''<!--b-->'''BI", "'''BI")
+    assert_bold("'''b'''", "'''b'''")
+    assert_no_bold("''i1'''s")
+    assert_no_bold("<!--'''b'''-->")
+    assert_bold(
         "a<!---->'<!---->'<!---->'<!---->" "b<!---->'<!---->'<!---->'<!---->d",
         "'<!---->'<!---->'<!---->b<!---->'<!---->'<!---->'",
     )
-    ab("'''b{{a|'''}}", "'''b{{a|'''}}")  # ?
-    ab("a'''b{{text|c|d}}e'''f", "'''b{{text|c|d}}e'''")
-    ab("{{text|'''b'''}}", "'''b'''")
-    ab("{{text|'''b}}", "'''b")  # ?
-    ab("[[a|'''b]] c", "'''b")
-    ab("{{{PARAM|'''b}}} c", "'''b")  # ?
+    assert_bold("'''b{{a|'''}}", "'''b{{a|'''}}")  # ?
+    assert_bold("a'''b{{text|c|d}}e'''f", "'''b{{text|c|d}}e'''")
+    assert_bold("{{text|'''b'''}}", "'''b'''")
+    assert_bold("{{text|'''b}}", "'''b")  # ?
+    assert_bold("{{{PARAM|'''b}}} c", "'''b")  # ?
     assert (
         repr(parse("'''b\na'''c").get_bolds())
         == """[Bold("'''b"), Bold("'''c")]"""
     )
-    ab("'''<S>b</S>'''", "'''<S>b</S>'''")
-    ab("'''b<S>r'''c</S>", "'''b<S>r'''")
-    ab("'''''b'''i", "'''b'''")
+    assert_bold("'''<S>b</S>'''", "'''<S>b</S>'''")
+    assert_bold("'''b<S>r'''c</S>", "'''b<S>r'''")
+    assert_bold("'''''b'''i", "'''b'''")
     assert (
         repr(parse("'''b<ref>r'''c</ref>a").get_bolds())
         == """[Bold("'''b<ref>r'''c</ref>a"), Bold("'''c")]"""
@@ -39,12 +45,16 @@ def anb(s: str):
         repr(parse("'''b<ref>r'''c</ref>a").get_bolds(False))
         == """[Bold("'''b<ref>r'''c</ref>a")]"""
     )
-    ab("'''b{{{p|'''}}}", "'''b{{{p|'''}}}")  # ?
-    ab("<nowiki>'''a</nowiki>'''b", "'''a")
-    anb("' ' ' a ' ' '")
-    ab("x''' '''y", "''' '''")
-    ab("x''''''y", "'''y")
-    ab("{{text|{{text|'''b'''}}}}", "'''b'''")
+    assert_bold("'''b{{{p|'''}}}", "'''b{{{p|'''}}}")  # ?
+    assert_bold("<nowiki>'''a</nowiki>'''b", "'''a")
+    assert_no_bold("' ' ' a ' ' '")
+    assert_bold("x''' '''y", "''' '''")
+    assert_bold("x''''''y", "'''y")
+    assert_bold("{{text|{{text|'''b'''}}}}", "'''b'''")
+
+
+def test_no_end_in_wikilink():
+    assert_bold("[[a|'''b]] c", "'''b")
 
 
 def test_get_italics():

tests/wikitext/test_plain_text.py

@@ -217,3 +217,16 @@ def test_file_links():
     # Fails for the following cases:
     # assert parse('[[Media:Example.jpg]]').plain_text() == 'Media:Example.jpg'
     # assert parse('[[Media:n.jpg|Sunflowers]]').plain_text() == 'Sunflowers'
+
+
+def test_tag_containing_comment_with_no_end():  # 126
+    parsed = parse(
+        """
+        [[a|b]]
+        <gallery>
+        <!-- 
+        </gallery>
+        """
+    )
+    del parsed.wikilinks[0][:]
+    assert parsed.plain_text().strip() == ''

wikitextparser/_comment_bold_italic.py

@@ -1,4 +1,4 @@
-from typing import Dict, List, MutableSequence, Optional, Union
+from typing import Dict, List, MutableSequence, Optional, Union, Tuple
 
 from regex import DOTALL, MULTILINE
 
@@ -49,7 +49,7 @@ def text(self, s: str):
         self[b:e] = s
 
     @property
-    def _relative_contents_end(self) -> tuple:
+    def _content_span(self) -> Tuple[int, int]:
         # noinspection PyUnresolvedReferences
         return self._match.span(1)
 

wikitextparser/_parameter.py

@@ -1,4 +1,4 @@
-from typing import List, Optional
+from typing import List, Optional, Tuple
 
 from ._wikitext import WS, SubWikiText
 
@@ -104,5 +104,5 @@ def parameters(self) -> List['Parameter']:
         return super().parameters[1:]
 
     @property
-    def _relative_contents_end(self) -> tuple:
+    def _content_span(self) -> Tuple[int, int]:
         return 3, -3

wikitextparser/_parser_function.py

@@ -1,5 +1,5 @@
 from bisect import insort
-from typing import Iterable, List, Union
+from typing import Iterable, List, Union, Tuple
 
 from ._argument import Argument
 from ._wikilist import WikiList
@@ -19,6 +19,10 @@ class SubWikiTextWithArgs(SubWikiText):
     _name_args_matcher = NotImplemented
     _first_arg_sep = 0
 
+    @property
+    def _content_span(self) -> Tuple[int, int]:
+        return 2, -2
+
     @property
     def nesting_level(self) -> int:
         """Return the nesting level of self.
@@ -95,10 +99,6 @@ def name(self) -> str:
     def name(self, newname: str) -> None:
         self[2 : 2 + len(self.name)] = newname
 
-    @property
-    def _relative_contents_end(self) -> tuple:
-        return 2, -2
-
 
 class ParserFunction(SubWikiTextWithArgs):
     __slots__ = ()

wikitextparser/_tag.py

@@ -8,7 +8,7 @@
 * https://www.mediawiki.org/wiki/HTML_restriction
 """
 
-from typing import Any, Dict, List, Optional
+from typing import Any, Dict, List, Optional, Tuple
 
 from regex import DOTALL, VERBOSE
 
@@ -215,5 +215,6 @@ def get_tags(self, name=None) -> List['Tag']:
         return super().get_tags(name)[1:]
 
     @property
-    def _relative_contents_end(self) -> tuple:
-        return self._match.span('contents')
+    def _content_span(self) -> Tuple[int, int]:
+        s = self.string
+        return s.find('>') + 1, s.rfind('<')

wikitextparser/_template.py

@@ -29,6 +29,10 @@ class Template(SubWikiTextWithArgs):
     _name_args_matcher = TL_NAME_ARGS_FULLMATCH
     _first_arg_sep = 124
 
+    @property
+    def _content_span(self) -> Tuple[int, int]:
+        return 2, -2
+
     def normal_name(
         self,
         rm_namespaces=('Template',),

wikitextparser/_wikilink.py

@@ -1,7 +1,7 @@
 """Define the WikiLink class."""
 
 
-from typing import List, Optional
+from typing import List, Optional, Tuple
 
 from regex import DOTALL
 
@@ -22,6 +22,13 @@
 class WikiLink(SubWikiText):
     __slots__ = '_cached_match'
 
+    @property
+    def _content_span(self) -> Tuple[int, int]:
+        s = self.string
+        f = s.find
+        rf = s.rfind
+        return f('[', f('[') + 1) + 1, rf(']', None, rf(']'))
+
     @property
     def _match(self):
         shadow = self._shadow
@@ -141,7 +148,3 @@ def title(self) -> None:
     @property
     def wikilinks(self) -> List['WikiLink']:
         return super().wikilinks[1:]
-
-    @property
-    def _relative_contents_end(self) -> tuple:
-        return self._match.span(4)

wikitextparser/_wikitext.py

@@ -575,6 +575,11 @@ def _nesting_level(self, parent_types) -> int:
                     level += 1
         return level
 
+    @property
+    def _content_span(self) -> Tuple[int, int]:
+        # return content_start, self_len, self_end
+        return 0, len(self)
+
     @property
     def _shadow(self) -> bytearray:
         """Return a copy of self.string with specific sub-spans replaced.
@@ -597,12 +602,14 @@ def _shadow(self) -> bytearray:
             self._lststr[0][ss:se], 'ascii', 'replace'
         )
         if self._type in SPAN_PARSER_TYPES:
-            head = shadow[:2]
-            tail = shadow[-2:]
-            shadow[:2] = shadow[-2:] = b'__'
+            cs, ce = self._content_span
+            head = shadow[:cs]
+            tail = shadow[ce:]
+            shadow[:cs] = b'_' * cs
+            shadow[ce:] = b'_' * len(tail)
             parse_to_spans(shadow)
-            shadow[:2] = head
-            shadow[-2:] = tail
+            shadow[:cs] = head
+            shadow[ce:] = tail
         else:
             parse_to_spans(shadow)
         return shadow
@@ -1002,8 +1009,6 @@ def comments(self) -> List['Comment']:
             for span in self._subspans('Comment')
         ]
 
-    _relative_contents_end = span
-
     @property
     def _balanced_quotes_shadow(self):
         """Return bold and italic match objects according MW's algorithm.
@@ -1110,7 +1115,7 @@ def get_bolds_and_italics(
         type_to_spans = self._type_to_spans
         tts_setdefault = type_to_spans.setdefault
         balanced_shadow = self._balanced_quotes_shadow
-        rs, re = self._relative_contents_end
+        rs, re = self._content_span
 
         if filter_cls is None or filter_cls is Bold:
             bold_spans = tts_setdefault('Bold', [])