jenkins-bot merged this change.

View Change

textlib._tag_pattern: Do not mistake self-closing tags with start tag

Bug: T203568
Change-Id: I58ac107691b2c6866d2b7568983e8760f7e6683a
---
M pywikibot/textlib.py
M tests/textlib_tests.py
2 files changed, 14 insertions(+), 1 deletion(-)

diff --git a/pywikibot/textlib.py b/pywikibot/textlib.py
index c9d416b..b58ae27 100644
--- a/pywikibot/textlib.py
+++ b/pywikibot/textlib.py
@@ -258,7 +258,11 @@
 
 def _tag_pattern(tag_name):
     """Return a tag pattern for the given tag name."""
-    return r'<{0}[ >][\s\S]*?</{0}\s*>'.format(_ignore_case(tag_name))
+    return (
+        r'<{0}(?:>|\s+[^>]*(?<!/)>)'  # start tag
+        r'[\s\S]*?'  # contents
+        r'</{0}\s*>'  # end tag
+        .format(_ignore_case(tag_name)))
 
 
 def _tag_regex(tag_name):
diff --git a/tests/textlib_tests.py b/tests/textlib_tests.py
index 67db849..0e56cc8 100644
--- a/tests/textlib_tests.py
+++ b/tests/textlib_tests.py
@@ -1216,6 +1216,15 @@
         self.assertEqual(textlib.replaceExcept('{{#invoke:x}}', 'x', 'y',
                                                ['invoke'], site=self.site),
                          '{{#invoke:x}}')
+        self.assertEqual(
+            textlib.replaceExcept(
+                '<ref name=etwa /> not_in_ref <ref> in_ref </ref>',
+                'not_in_ref', 'text', ['ref'], site=self.site),
+            '<ref name=etwa /> text <ref> in_ref </ref>')
+        self.assertEqual(
+            textlib.replaceExcept(
+                '<ab> content </a>', 'content', 'text', ['a'], site=self.site),
+            '<ab> text </a>')
 
     def test_replace_with_count(self):
         """Test replacing with count argument."""

To view, visit change 458636. To unsubscribe, or for help writing mail filters, visit settings.