From 370710657a7e7c771668107d1b6407fc350a2891 Mon Sep 17 00:00:00 2001
From: Jay Berkenbilt <ejb@ql.org>
Date: Tue, 11 Jan 2022 15:06:17 -0500
Subject: Add missing characters from PDF doc encoding (fixes #606)

---
 libqpdf/QUtil.cc | 46 +++++++++++++++++++++++++++++++++++++++++++---
 1 file changed, 43 insertions(+), 3 deletions(-)

(limited to 'libqpdf')

diff --git a/libqpdf/QUtil.cc b/libqpdf/QUtil.cc
index daa663a3..c71e7923 100644
--- a/libqpdf/QUtil.cc
+++ b/libqpdf/QUtil.cc
@@ -37,8 +37,20 @@
 # include <sys/stat.h>
 #endif
 
-// First element is 128
+// First element is 24
+static unsigned short pdf_doc_low_to_unicode[] = {
+    0x02d8,    // 0x18    BREVE
+    0x02c7,    // 0x19    CARON
+    0x02c6,    // 0x1a    MODIFIER LETTER CIRCUMFLEX ACCENT
+    0x02d9,    // 0x1b    DOT ABOVE
+    0x02dd,    // 0x1c    DOUBLE ACUTE ACCENT
+    0x02db,    // 0x1d    OGONEK
+    0x02da,    // 0x1e    RING ABOVE
+    0x02dc,    // 0x1f    SMALL TILDE
+};
+// First element is 127
 static unsigned short pdf_doc_to_unicode[] = {
+    0xfffd,    // 0x7f    UNDEFINED
     0x2022,    // 0x80    BULLET
     0x2020,    // 0x81    DAGGER
     0x2021,    // 0x82    DOUBLE DAGGER
@@ -2032,6 +2044,30 @@ encode_pdfdoc(unsigned long codepoint)
     unsigned char ch = '\0';
     switch (codepoint)
     {
+      case 0x02d8:
+        ch = 0x18;
+        break;
+      case 0x02c7:
+        ch = 0x19;
+        break;
+      case 0x02c6:
+        ch = 0x1a;
+        break;
+      case 0x02d9:
+        ch = 0x1b;
+        break;
+      case 0x02dd:
+        ch = 0x1c;
+        break;
+      case 0x02db:
+        ch = 0x1d;
+        break;
+      case 0x02da:
+        ch = 0x1e;
+        break;
+      case 0x02dc:
+        ch = 0x1f;
+        break;
       case 0x2022:
         ch = 0x80;
         break;
@@ -2427,9 +2463,13 @@ QUtil::pdf_doc_to_utf8(std::string const& val)
     {
         unsigned char ch = static_cast<unsigned char>(val.at(i));
         unsigned short ch_short = ch;
-        if ((ch >= 128) && (ch <= 160))
+        if ((ch >= 127) && (ch <= 160))
+        {
+            ch_short = pdf_doc_to_unicode[ch - 127];
+        }
+        else if ((ch >= 24) && (ch <= 31))
         {
-            ch_short = pdf_doc_to_unicode[ch - 128];
+            ch_short = pdf_doc_low_to_unicode[ch - 24];
         }
         result += QUtil::toUTF8(ch_short);
     }
-- 
cgit v1.2.3-54-g00ecf