Ignore non-breaking spaces and word joiners in s-041

When the text being compared is from the image's alt attribute, since
they are not used in that context.

se/se_epub_lint.py

@@ -1517,13 +1517,16 @@ def _lint_special_file_checks(self, filename: Path, dom: se.easy_xml.EasyXmlTree
 			loi_text_matches_figure = False
 			for child in figure.xpath("./img|./figcaption"):
 				figure_text = ""
+				loi_text_to_compare = loi_text
 				if child.tag == "img":
 					figure_text = child.get_attr("alt")
+					# Replace/remove characters that don't appear in alt attributes.
+					loi_text_to_compare = loi_text_to_compare.replace(se.NO_BREAK_SPACE, ' ').replace(se.WORD_JOINER, '')
 				elif child.tag == "figcaption":
 					# Replace tabs and newlines with a single space to better match figcaptions that contain <br/>
 					figure_text = regex.sub(r"[ \n\t]+", " ", child.inner_text())
 
-				if loi_text == figure_text:
+				if loi_text_to_compare == figure_text:
 					loi_text_matches_figure = True
 					break
 

tests/lint/semantic/s-041/golden/s-041-out.txt

@@ -1,9 +1,12 @@
 s-004 [Error] chapter-1.xhtml `img` element missing `alt` attribute.
         <img>
         <img>
+        <img>
 s-041 [Manual Review] chapter-1.xhtml The text in `#f-5`'s LoI entry does not 
 match either its `<figcaption>` element or its `<img>` `alt` attribute.
 s-041 [Manual Review] chapter-1.xhtml The text in `#f-6`'s LoI entry does not 
 match either its `<figcaption>` element or its `<img>` `alt` attribute.
 s-041 [Manual Review] chapter-1.xhtml The text in `#f-7`'s LoI entry does not 
 match either its `<figcaption>` element or its `<img>` `alt` attribute.
+s-041 [Manual Review] chapter-1.xhtml The text in `#f-9`'s LoI entry does not 
+match either its `<figcaption>` element or its `<img>` `alt` attribute.

tests/lint/semantic/s-041/in/src/epub/text/chapter-1.xhtml

@@ -16,6 +16,9 @@
 			<!-- will cause s-004 to be emitted, but we are deliberately testing behavior with missing alt --> 
 			<figure id="f-6"><img/><figcaption>t-6-cap</figcaption></figure>
 			<figure id="f-7"><img alt="t-7-alt."/><figcaption>t-7-cap</figcaption></figure>
+
+			<figure id="f-8"><img alt="Mr. Smith 2–3 years ago."/></figure>
+			<figure id="f-9"><img/><figcaption>Mr. Smith 2–3 years ago.</figcaption></figure>
 		</section>
 	</body>
 </html>

tests/lint/semantic/s-041/in/src/epub/text/loi.xhtml

@@ -17,6 +17,12 @@
 				<li><p><a href="chapter-1.xhtml#f-5">x</a></p></li>
 				<li><p><a href="chapter-1.xhtml#f-6">x</a></p></li>
 				<li><p><a href="chapter-1.xhtml#f-7">x</a></p></li>
+
+				<!-- text matches alt once tags and word joiners and non-breaking spaces are removed -->
+				<li><p><a href="chapter-1.xhtml#f-8"><abbr epub:type="z3998:name-title">Mr.</abbr> Smith 2⁠–⁠3 years ago.</a></p></li>
+
+				<!-- text does not match caption due to differing word joiners and non-breaking spaces -->
+				<li><p><a href="chapter-1.xhtml#f-9"><abbr epub:type="z3998:name-title">Mr.</abbr> Smith 2⁠–⁠3 years ago.</a></p></li>
 			</ol>
 		</nav>
 	</body>