Merge pull request #1 from diffbot/performance-fixes

praveen-diffbot · web-flow · commit e8fdf0b7d26b · 2024-09-17T09:17:03.000-07:00
Use a string function rather than regex for better performance
diff --git a/flexmark-html2md-converter/pom.xml b/flexmark-html2md-converter/pom.xml
@@ -7,6 +7,7 @@
         <version>0.64.8</version>
     </parent>
 
+    <groupId>com.diffbot</groupId>
     <artifactId>flexmark-html2md-converter</artifactId>
     <name>flexmark-java HTML to Markdown extensible converter</name>
     <description>flexmark-java customizable extension to convert HTML to Markdown</description>
diff --git a/flexmark-html2md-converter/src/main/java/com/vladsch/flexmark/html2md/converter/internal/HtmlConverterCoreNodeRenderer.java b/flexmark-html2md-converter/src/main/java/com/vladsch/flexmark/html2md/converter/internal/HtmlConverterCoreNodeRenderer.java
@@ -1199,7 +1199,7 @@ private void handleTableCaption(Element element, HtmlNodeConverterContext contex
     }
 
     private void handleTableCell(Element element, HtmlNodeConverterContext context, HtmlMarkdownWriter out) {
-        String cellText = context.processTextNodes(element).trim().replaceAll("\\s*\n\\s*", " ");
+        String cellText = replaceMultipleBlankSpace(context.processTextNodes(element).trim());
         int colSpan = 1;
         int rowSpan = 1;
         CellAlignment alignment = null;
@@ -1260,6 +1260,25 @@ private void handleTableCell(Element element, HtmlNodeConverterContext context,
         }
     }
 
+    private String replaceMultipleBlankSpace(String cellText) {
+        StringBuilder result = new StringBuilder();
+        boolean wasSpace = false;
+
+        for (char c : cellText.toCharArray()) {
+            if (Character.isWhitespace(c)) {
+                if (!wasSpace) {
+                    result.append(' ');
+                    wasSpace = true;
+                }
+            } else {
+                result.append(c);
+                wasSpace = false;
+            }
+        }
+
+        return result.toString();
+    }
+
     private boolean matchingText(Pattern pattern, String text, String[] match) {
         Matcher matcher = pattern.matcher(text);
         if (matcher.matches()) {
diff --git a/pom.xml b/pom.xml
@@ -93,12 +93,10 @@
 
     <distributionManagement>
         <snapshotRepository>
-            <id>ossrh</id>
-            <url>https://oss.sonatype.org/content/repositories/snapshots</url>
+            <id>diffbot-nexus-snapshots</id>
         </snapshotRepository>
         <repository>
-            <id>ossrh</id>
-            <url>https://oss.sonatype.org/service/local/staging/deploy/maven2/</url>
+            <id>diffbot-nexus-releases</id>
         </repository>
     </distributionManagement>
 
