lint: Add t-076, grapheme or phoneme not italicized

standardebooks · Oct 15, 2024 · d8b1b7d · d8b1b7d
1 parent 2276813
commit d8b1b7d
Show file tree

Hide file tree

Showing 3 changed files with 27 additions and 0 deletions.
diff --git a/se/se_epub_lint.py b/se/se_epub_lint.py
@@ -431,6 +431,7 @@
 "t-073", "Possible transcription error in Greek."
 "t-074", "Extended sound using hyphen-minus [text]-[/] instead of non-breaking hyphen [text]‑[/]."
 "t-075", "Word in verse with acute accent for scansion instead of grave accent."
+"t-076", "Grapheme or phoneme not italicized. Hint: Dialect with missing letters should mark missing letters with [text]’[/]."
 
 
 XHTML
@@ -2792,6 +2793,11 @@ def _lint_xhtml_typography_checks(filename: Path, dom: se.easy_xml.EasyXmlTree,
 	if filtered_nodes:
 		messages.append(LintMessage("t-075", "Word in verse with acute accent for scansion instead of grave accent.", se.MESSAGE_TYPE_WARNING, filename, [node.to_string() for node in filtered_nodes]))
 
+	# Check for graphemes or phonemes that are not italicized.
+	nodes = dom.xpath("/html/body//text()[re:test(., '\\s[a-z]’s\\b')]")
+	if nodes:
+		messages.append(LintMessage("t-076", "Grapheme or phoneme not italicized. Hint: Dialect with missing letters should mark missing letters with [text]’[/].", se.MESSAGE_TYPE_ERROR, filename, nodes))
+
 	return (messages, missing_files)
 
 def _lint_xhtml_xhtml_checks(filename: Path, dom: se.easy_xml.EasyXmlTree, file_contents: str, local_css_path: str) -> list:

diff --git a/tests/lint/typography/t-076/golden/t-076-out.txt b/tests/lint/typography/t-076/golden/t-076-out.txt
@@ -0,0 +1,2 @@
+t-076 [Error] chapter-1.xhtml Grapheme or phoneme not italicized.
+        Two f’s.
diff --git a/tests/lint/typography/t-076/in/src/epub/text/chapter-1.xhtml b/tests/lint/typography/t-076/in/src/epub/text/chapter-1.xhtml
@@ -0,0 +1,19 @@
+<?xml version="1.0" encoding="utf-8"?>
+<html xmlns="http://www.w3.org/1999/xhtml" xmlns:epub="http://www.idpf.org/2007/ops" epub:prefix="z3998: http://www.daisy.org/z3998/2012/vocab/structure/, se: https://standardebooks.org/vocab/1.0" xml:lang="en-GB">
+	<head>
+		<title>I</title>
+		<link href="../css/core.css" rel="stylesheet" type="text/css"/>
+		<link href="../css/local.css" rel="stylesheet" type="text/css"/>
+	</head>
+	<body epub:type="bodymatter z3998:fiction">
+		<section id="chapter-1" epub:type="chapter">
+			<h2 epub:type="ordinal z3998:roman">I</h2>
+			<!-- Match -->
+			<p>Two f’s.</p>
+			<!-- Ignore uppercase -->
+			<p>The three R’s.</p>
+			<!-- Ignore common dialect -->
+			<p>“I’s hungry,” he said.</p>
+		</section>
+	</body>
+</html>