Do not prefix standalone linebreaks (as often emitted from HTML parsers

without any semantical meaning)
apache · Oct 19, 2024 · 6a78acb · 6a78acb
1 parent 6631e16
commit 6a78acb
Show file tree

Hide file tree

Showing 2 changed files with 8 additions and 8 deletions.
diff --git a/...ia-module-markdown/src/main/java/org/apache/maven/doxia/module/markdown/MarkdownSink.java b/...ia-module-markdown/src/main/java/org/apache/maven/doxia/module/markdown/MarkdownSink.java
@@ -873,9 +873,6 @@ public void nonBreakingSpace() {
 
     @Override
     public void text(String text, SinkEventAttributes attributes) {
-        if (text.equals("\n")) {
-            return;
-        }
         if (attributes != null) {
             inline(attributes);
         }
@@ -885,10 +882,12 @@ public void text(String text, SinkEventAttributes attributes) {
             LOGGER.warn("{}Ignoring unsupported table caption in Markdown", getLocationLogPrefix());
         } else {
             String unifiedText = currentContext.escape(unifyEOLs(text));
-            // each line must potentially be prefixed
-            String prefix = getLinePrefix();
-            if (prefix.length() > 0) {
-                unifiedText = unifiedText.replaceAll(EOL, EOL + prefix);
+            // ignore newlines only, because those are emitted often coming from linebreaks in HTML with no semantical meaning
+            if (!unifiedText.equals(EOL)) {
+                String prefix = getLinePrefix();
+                if (prefix.length() > 0) {
+                    unifiedText = unifiedText.replaceAll(EOL, EOL + prefix);
+                }
             }
             writeUnescaped(unifiedText);
         }

diff --git a/...odule-markdown/src/test/java/org/apache/maven/doxia/module/markdown/MarkdownSinkTest.java b/...odule-markdown/src/test/java/org/apache/maven/doxia/module/markdown/MarkdownSinkTest.java
@@ -371,6 +371,7 @@ public void testRoundtrip() throws IOException, ParseException {
 
         final SinkEventTestingSink originalSink = new SinkEventTestingSink();
         parseFile(parser, "test", originalSink);
+        // strip empty lines from sink content
 
         // compare sink events from parsing original markdown with sink events from re-generated markdown
         try {
@@ -518,7 +519,7 @@ public void testMultilineVerbatimSourceAfterListItem() {
             sink.listItem();
             sink.text("Before");
             sink.verbatim(SinkEventAttributeSet.SOURCE);
-            sink.text("codeline1\ncodeline2");
+            sink.text("codeline1" + EOL + "codeline2");
             sink.verbatim_();
             sink.text("After");
             sink.listItem_();