Fix m-070

- Treat glossary values as string literals, not patterns - Search for glossary values surrounded by word boundaries
apasel422 · Jul 13, 2024 · ab2b45e · ab2b45e
1 parent 21dd6d5
commit ab2b45e
Show file tree

Hide file tree

Showing 3 changed files with 2 additions and 4 deletions.
diff --git a/se/se_epub_lint.py b/se/se_epub_lint.py
@@ -3619,7 +3619,7 @@ def lint(self, skip_lint_ignore: bool, allowed_messages: Optional[List[str]] = N
 						nodes = dom.xpath("/html/body//dd[contains(@epub:type, 'glossdef')]")
 						source_text = " ".join([node.inner_text() for node in nodes])
 					for glossary_index, glossary_value in enumerate(glossary_usage):
-						if glossary_value[1] is False and regex.search(glossary_value[0], source_text, flags=regex.IGNORECASE):
+						if glossary_value[1] is False and regex.search(r"\b\L<val>\b", source_text, flags=regex.IGNORECASE, val=[glossary_value[0]]):
 							glossary_usage[glossary_index] = (glossary_value[0], True)
 
 				# Test against word boundaries to not match `halftitlepage`

diff --git a/tests/lint/metadata/m-070/golden/m-070-out.txt b/tests/lint/metadata/m-070/golden/m-070-out.txt
@@ -1,2 +1,2 @@
 m-070 [Error] glossary-search-key-map.xml Glossary entry not found in the text.
-        R+L=J
+        foo
diff --git a/tests/lint/metadata/m-070/in/src/epub/text/chapter-1.xhtml b/tests/lint/metadata/m-070/in/src/epub/text/chapter-1.xhtml
@@ -8,9 +8,7 @@
 	<body epub:type="bodymatter z3998:fiction">
 		<section id="chapter-1" epub:type="chapter">
 			<h2 epub:type="ordinal z3998:roman">I</h2>
-			<!-- erroneously considered to match the glossary term "foo" -->
 			<p>He ate some food.</p>
-			<!-- erroneously unmatched due to treatment of glossary term as a regex -->
 			<p>A common theory was R+L=J.</p>
 		</section>
 	</body>