Fix export of HTML when using UTF-8 (Issue #526)

michaelrsweet · Nov 21, 2024 · 4c61bca · 4c61bca
1 parent 8d1c067
commit 4c61bca
Show file tree

Hide file tree

Showing 2 changed files with 42 additions and 7 deletions.
diff --git a/CHANGES.md b/CHANGES.md
@@ -5,6 +5,7 @@
   management and fix potential double-free bugs.
 - Updated configure script to look for zlib with pkg-config (Issue #519)
 - Updated markdown support code to mmd.
+- Fixed export of UTF-8 HTML (Issue #526)
 - Fixed handling of whitespace-only nodes (Issue #528)
 - Fixed handling of tabs in PRE nodes (Issue #529)
 - Fixed case sensitivity of link targets (Issue #530)

diff --git a/htmldoc/iso8859.cxx b/htmldoc/iso8859.cxx
@@ -2,17 +2,13 @@
  * ISO-8859-1 conversion routines for HTMLDOC, an HTML document
  * processing program.
  *
- * Copyright 2011-2019 by Michael R Sweet.
+ * Copyright 2011-2024 by Michael R Sweet.
  * Copyright 1997-2010 by Easy Software Products.  All rights reserved.
  *
  * This program is free software.  Distribution and use rights are outlined in
  * the file "COPYING".
  */
 
-/*
- * Include necessary headers.
- */
-
 #include <stdio.h>
 #include <stdlib.h>
 
@@ -441,11 +437,49 @@ iso8859(uchar value)	/* I - ISO-8859-1 equivalent */
 
   if (iso8859_names[value] == NULL)
   {
-    buf[0] = value;
-    buf[1] = '\0';
+    if (value < 127)
+    {
+      // ASCII...
+      buf[0] = value;
+      buf[1] = '\0';
+    }
+    else if (_htmlUTF8)
+    {
+      // UTF-8...
+      int unich = _htmlUnicode[value];	// Unicode character
+
+      if (unich < 0x400)
+      {
+        buf[0] = 0xc0 | (unich >> 6);
+        buf[1] = 0x80 | (unich & 0x3f);
+        buf[2] = '\0';
+      }
+      else if (unich < 0x10000)
+      {
+        buf[0] = 0xe0 | (unich >> 12);
+        buf[1] = 0x80 | ((unich >> 6) & 0x3f);
+        buf[2] = 0x80 | (unich & 0x3f);
+        buf[3] = '\0';
+      }
+      else
+      {
+        buf[0] = 0xf0 | (unich >> 18);
+        buf[1] = 0x80 | ((unich >> 12) & 0x3f);
+        buf[2] = 0x80 | ((unich >> 6) & 0x3f);
+        buf[3] = 0x80 | (unich & 0x3f);
+        buf[4] = '\0';
+      }
+    }
+    else
+    {
+      // Character-set neutral way to map to Unicode...
+      snprintf((char *)buf, sizeof(buf), "&#%d;", _htmlUnicode[value]);
+    }
   }
   else
+  {
     snprintf((char *)buf, sizeof(buf), "&%s;", iso8859_names[value]->name);
+  }
 
   return (buf);
 }