Merge 8b2b013b36ff05b3f98afed416fe3e87023b121a into 48fe8d1bc9d56189001600065cad84cc05851886

Bump version.
Prep for release.
2025-04-08 18:06:44 +02:00 · 2025-01-25 12:41:16 +04:00 · 2025-01-24 15:31:31 -05:00 · 2025-01-24 15:30:59 -05:00 · 2025-01-24 14:44:44 -05:00 · 2025-01-24 14:42:41 -05:00
11 changed files with 174 additions and 26 deletions
--- a/CHANGES.md
+++ b/CHANGES.md
@ -1,8 +1,7 @@
 Changes in PDFio
 ================

-
-v1.4.1 - YYYY-MM-DD
+v1.4.1 - 2025-01-24
 -------------------

 - Added license files for the example fonts now bundled with PDFio (Issue #91)
@ -10,6 +9,8 @@ v1.4.1 - YYYY-MM-DD
 - Fixed handling of the Info object (Issue #87)
 - Fixed opening of PDF files less than 1024 bytes in length (Issue #87)
 - Fixed potential `NULL` dereference when reading (Issue #89)
+- Fixed reading of compressed object streams (Issue #92)
+- Fixed reading of UTF-16 string values (Issue #92)


 v1.4.0 - 2024-12-26
--- a/Makefile.in
+++ b/Makefile.in
@ -117,12 +117,14 @@ DOCFILES	=	\
 			NOTICE
 EXAMPLES	=	\
 			examples/Makefile \
+			examples/Roboto-LICENSE.txt \
 			examples/Roboto-Bold.ttf \
 			examples/Roboto-Italic.ttf \
 			examples/Roboto-Regular.ttf \
 			examples/RobotoMono-Regular.ttf \
 			examples/code128.c \
 			examples/code128.ttf \
+			examples/code128-LICENSE.txt \
 			examples/image2pdf.c \
 			examples/md2pdf.c \
 			examples/md2pdf.md \
--- a/doc/pdfio.3
+++ b/doc/pdfio.3
@ -1,4 +1,4 @@
-.TH pdfio 3 "pdf read/write library" "2025-01-17" "pdf read/write library"
+.TH pdfio 3 "pdf read/write library" "2025-01-24" "pdf read/write library"
 .SH NAME
 pdfio \- pdf read/write library
 .SH Introduction
@ -2074,7 +2074,7 @@ The render_line function adds content from the linefrag_t array to a PDF page. I
    }
 .fi
 .PP
-We then loops through the fragments for the current line, drawing checkboxes, images, and text as needed. Whan a hyperlink is present, we add the link to the links array in the docdata_t structure, mapping "@" and "@@" to an internal link corresponding to the linked text:
+We then loop through the fragments for the current line, drawing checkboxes, images, and text as needed. When a hyperlink is present, we add the link to the links array in the docdata_t structure, mapping "@" and "@@" to an internal link corresponding to the linked text:
 .nf

    if (frag\->url && dd\->num_links < DOCLINK_MAX)
@ -4567,4 +4567,4 @@ typedef enum pdfio_valtype_e pdfio_valtype_t;
 Michael R Sweet
 .SH COPYRIGHT
 .PP
-Copyright (c) 2021-2024 by Michael R Sweet
+Copyright (c) 2021-2025 by Michael R Sweet
--- a/doc/pdfio.html
+++ b/doc/pdfio.html
@ -1,13 +1,13 @@
 <!DOCTYPE html>
 <html lang="en-US">
 <head>
-<title>PDFio Programming Manual v1.4.0</title>
+<title>PDFio Programming Manual v1.4.1</title>
 <meta http-equiv="Content-Type" content="text/html;charset=utf-8">
 <meta name="generator" content="codedoc v3.8">
 <meta name="author" content="Michael R Sweet">
 <meta name="language" content="en-US">
-<meta name="copyright" content="Copyright Â© 2021-2024 by Michael R Sweet">
-<meta name="version" content="1.4.0">
+<meta name="copyright" content="Copyright © 2021-2025 by Michael R Sweet">
+<meta name="version" content="1.4.1">
 <style type="text/css"><!--
 body {
  background: white;
@ -251,9 +251,9 @@ span.string {
 <body>
 <div class="header">
 <p><img class="title" src="pdfio-512.png"></p>
-<h1 class="title">PDFio Programming Manual v1.4.0</h1>
+<h1 class="title">PDFio Programming Manual v1.4.1</h1>
 <p>Michael R Sweet</p>
-<p>Copyright Â© 2021-2024 by Michael R Sweet</p>
+<p>Copyright © 2021-2025 by Michael R Sweet</p>
 </div>
 <div class="contents">
 <h2 class="title">Contents</h2>
@ -2034,7 +2034,7 @@ dd-&gt;y -= margin_top + lineheight;
  dd-&gt;y -= lineheight;
 }
 </code></pre>
-<p>We then loops through the fragments for the current line, drawing checkboxes, images, and text as needed. WhÑn a hyperlink is present, we add the link to the <code>links</code> array in the <code>docdata_t</code> structure, mapping &quot;@&quot; and &quot;@@&quot; to an internal link corresponding to the linked text:</p>
+<p>We then loop through the fragments for the current line, drawing checkboxes, images, and text as needed. When a hyperlink is present, we add the link to the <code>links</code> array in the <code>docdata_t</code> structure, mapping &quot;@&quot; and &quot;@@&quot; to an internal link corresponding to the linked text:</p>
 <pre><code class="language-c"><span class="reserved">if</span> (frag-&gt;url &amp;&amp; dd-&gt;num_links &lt; DOCLINK_MAX)
 {
  doclink_t *l = dd-&gt;links + dd-&gt;num_links;
--- a/doc/pdfio.md
+++ b/doc/pdfio.md
@ -2005,7 +2005,7 @@ if ((dd->y - need_bottom) < dd->art_box.y1)
 }
 ```

-We then loops through the fragments for the current line, drawing checkboxes,
+We then loop through the fragments for the current line, drawing checkboxes,
 images, and text as needed.  When a hyperlink is present, we add the link to the
 `links` array in the `docdata_t` structure, mapping "@" and "@@" to an internal
 link corresponding to the linked text:
--- a/examples/code128-LICENSE.txt
+++ b/examples/code128-LICENSE.txt
@ -1,3 +1,7 @@
+Copyright 2003 Grandzebu, All Rights Reserved
+
+http://grandzebu.net/informatique/codbar-en/code128.htm
+
                    GNU GENERAL PUBLIC LICENSE
                       Version 2, June 1991

--- a/9
+++ b/9
@ -29,6 +29,15 @@ if test $(grep AC_INIT configure.ac | awk '{print $2}') != "[$version],"; then
    exit 1
 fi

+if test $(head -4 CHANGES.md | tail -1 | awk '{print $1}') != "v$version"; then
+    echo "Still need to update CHANGES.md version number."
+    exit 1
+fi
+if test $(head -4 CHANGES.md | tail -1 | awk '{print $3}') = "YYYY-MM-DD"; then
+    echo "Still need to update CHANGES.md release date."
+    exit 1
+fi
+
 if test $(grep PDFIO_VERSION= configure | awk -F \" '{print $2}') != "$version"; then
    echo "Still need to run 'autoconf -f'."
    exit 1
--- a/pdfio-dict.c
+++ b/pdfio-dict.c
@ -465,10 +465,134 @@ pdfioDictGetString(pdfio_dict_t *dict,	// I - Dictionary
  else if (value && value->type == PDFIO_VALTYPE_BINARY && value->value.binary.datalen < 4096)
  {
    // Convert binary string to regular string...
-    char	temp[4096];		// Temporary string
+    char	temp[4096],		// Temporary string
+		*tempptr;		// Pointer into temporary string
+    unsigned char *dataptr;		// Pointer into the data string

-    memcpy(temp, value->value.binary.data, value->value.binary.datalen);
-    temp[value->value.binary.datalen] = '\0';
+    if (!(value->value.binary.datalen & 1) && !memcmp(value->value.binary.data, "\377\376", 2))
+    {
+      // Copy UTF-16 BE
+      int	ch;			// Unicode character
+      size_t	remaining;		// Remaining bytes
+
+      for (dataptr = value->value.binary.data + 2, remaining = value->value.binary.datalen - 2, tempptr = temp; remaining > 1 && tempptr < (temp + sizeof(temp) - 5); dataptr += 2, remaining -= 2)
+      {
+        ch = (dataptr[0] << 8) | dataptr[1];
+
+        if (ch >= 0xd800 && ch <= 0xdbff && remaining > 3)
+	{
+	  // Multi-word UTF-16 char...
+          int lch;			// Lower bits
+
+	  lch = (dataptr[2] << 8) | dataptr[3];
+
+          if (lch < 0xdc00 || lch >= 0xdfff)
+	    break;
+
+          ch        = (((ch & 0x3ff) << 10) | (lch & 0x3ff)) + 0x10000;
+          dataptr   += 2;
+          remaining -= 2;
+	}
+	else if (ch >= 0xfffe)
+	{
+	  continue;
+	}
+
+        if (ch < 128)
+        {
+          // ASCII
+          *tempptr++ = (char)ch;
+        }
+        else if (ch < 4096)
+        {
+          // 2-byte UTF-8
+          *tempptr++ = (char)(0xc0 | (ch >> 6));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+        else if (ch < 65536)
+        {
+          // 3-byte UTF-8
+          *tempptr++ = (char)(0xe0 | (ch >> 12));
+          *tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+        else
+        {
+          // 4-byte UTF-8
+          *tempptr++ = (char)(0xe0 | (ch >> 18));
+          *tempptr++ = (char)(0x80 | ((ch >> 12) & 0x3f));
+          *tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+      }
+
+      *tempptr = '\0';
+    }
+    else if (!(value->value.binary.datalen & 1) && !memcmp(value->value.binary.data, "\376\377", 2))
+    {
+      // Copy UTF-16 LE
+      int	ch;			// Unicode character
+      size_t	remaining;		// Remaining bytes
+
+      for (dataptr = value->value.binary.data + 2, remaining = value->value.binary.datalen - 2, tempptr = temp; remaining > 1 && tempptr < (temp + sizeof(temp) - 5); dataptr += 2, remaining -= 2)
+      {
+        ch = (dataptr[1] << 8) | dataptr[0];
+
+        if (ch >= 0xd800 && ch <= 0xdbff && remaining > 3)
+	{
+	  // Multi-word UTF-16 char...
+          int lch;			// Lower bits
+
+	  lch = (dataptr[3] << 8) | dataptr[2];
+
+          if (lch < 0xdc00 || lch >= 0xdfff)
+	    break;
+
+          ch        = (((ch & 0x3ff) << 10) | (lch & 0x3ff)) + 0x10000;
+          dataptr   += 2;
+          remaining -= 2;
+	}
+	else if (ch >= 0xfffe)
+	{
+	  continue;
+	}
+
+        if (ch < 128)
+        {
+          // ASCII
+          *tempptr++ = (char)ch;
+        }
+        else if (ch < 4096)
+        {
+          // 2-byte UTF-8
+          *tempptr++ = (char)(0xc0 | (ch >> 6));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+        else if (ch < 65536)
+        {
+          // 3-byte UTF-8
+          *tempptr++ = (char)(0xe0 | (ch >> 12));
+          *tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+        else
+        {
+          // 4-byte UTF-8
+          *tempptr++ = (char)(0xe0 | (ch >> 18));
+          *tempptr++ = (char)(0x80 | ((ch >> 12) & 0x3f));
+          *tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
+          *tempptr++ = (char)(0x80 | (ch & 0x3f));
+        }
+      }
+
+      *tempptr = '\0';
+    }
+    else
+    {
+      // Copy as-is...
+      memcpy(temp, value->value.binary.data, value->value.binary.datalen);
+      temp[value->value.binary.datalen] = '\0';
+    }

    free(value->value.binary.data);
    value->type         = PDFIO_VALTYPE_STRING;
--- a/pdfio-file.c
+++ b/pdfio-file.c
@ -1529,6 +1529,7 @@ load_obj_stream(pdfio_obj_t *obj)	// I - Object to load
 			cur_obj,	// Current object
 			num_objs = 0;	// Number of objects
  pdfio_obj_t		*objs[16384];	// Objects
+  int			count;		// Count of objects


  PDFIO_DEBUG("load_obj_stream(obj=%p(%d))\n", obj, (int)obj->number);
@ -1540,12 +1541,17 @@ load_obj_stream(pdfio_obj_t *obj)	// I - Object to load
    return (false);
  }

+  count = (int)pdfioDictGetNumber(pdfioObjGetDict(obj), "N");
+
+  PDFIO_DEBUG("load_obj_stream: N=%d\n", count);
+
  _pdfioTokenInit(&tb, obj->pdf, (_pdfio_tconsume_cb_t)pdfioStreamConsume, (_pdfio_tpeek_cb_t)pdfioStreamPeek, st);

  // Read the object numbers from the beginning of the stream...
-  while (_pdfioTokenGet(&tb, buffer, sizeof(buffer)))
+  while (count > 0 && _pdfioTokenGet(&tb, buffer, sizeof(buffer)))
  {
    // Stop if this isn't an object number...
+    PDFIO_DEBUG("load_obj_stream: %s\n", buffer);
    if (!isdigit(buffer[0] & 255))
      break;

@ -1568,21 +1574,19 @@ load_obj_stream(pdfio_obj_t *obj)	// I - Object to load
    // Skip offset
    _pdfioTokenGet(&tb, buffer, sizeof(buffer));
    PDFIO_DEBUG("load_obj_stream: %ld at offset %s\n", (long)number, buffer);
+
+    // One less compressed object...
+    count --;
  }

-  if (!buffer[0])
-  {
-    pdfioStreamClose(st);
-    return (false);
-  }
-
-  _pdfioTokenPush(&tb, buffer);
+  PDFIO_DEBUG("load_obj_stream: num_objs=%lu\n", (unsigned long)num_objs);

  // Read the objects themselves...
  for (cur_obj = 0; cur_obj < num_objs; cur_obj ++)
  {
    if (!_pdfioValueRead(obj->pdf, obj, &tb, &(objs[cur_obj]->value), 0))
    {
+      _pdfioFileError(obj->pdf, "Unable to read compressed object.");
      pdfioStreamClose(st);
      return (false);
    }
@ -1732,7 +1736,7 @@ load_xref(
      pdfio_stream_t	*st;		// Stream
      unsigned char	buffer[32];	// Read buffer
      size_t		num_sobjs = 0,	// Number of object streams
-			sobjs[8192];	// Object streams to load
+			sobjs[16384];	// Object streams to load
      pdfio_obj_t	*current;	// Current object

      if ((number = strtoimax(line, &ptr, 10)) < 1)
--- a/pdfio.h
+++ b/pdfio.h
@ -1,7 +1,7 @@
 //
 // Public header file for PDFio.
 //
-// Copyright © 2021-2024 by Michael R Sweet.
+// Copyright © 2021-2025 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@ -23,7 +23,7 @@ extern "C" {
 // Version number...
 //

-#  define PDFIO_VERSION		"1.4.0"
+#  define PDFIO_VERSION		"1.4.1"


 //
--- a/testfiles/iso22028-2-romm-rgb.txt
+++ b/testfiles/iso22028-2-romm-rgb.txt
@ -1,3 +1,7 @@
+https://www.color.org/chardata/rgb/rommrgb.xalter
+
+Copyright © 2006 Hewlett-Packard
+
 Terms of use

 This profile is made available by ICC, and may be copied, distributed, embedded, made, used, and sold without restriction. Altered versions of this profile shall have the original identification and copyright information removed and shall not be misrepresented as the original profile.
Author	SHA1	Message	Date
Thierry LARONDE	269160d745	Merge 8b2b013b36ff05b3f98afed416fe3e87023b121a into 48fe8d1bc9d56189001600065cad84cc05851886	2025-01-25 12:41:16 +04:00
Michael R Sweet	48fe8d1bc9	Bump version.	2025-01-24 15:31:31 -05:00
Michael R Sweet	a4026bfe00	Prep for release.	2025-01-24 15:30:59 -05:00
Michael R Sweet	1e945cb750	Add LICENSE files to example install list.	2025-01-24 14:44:44 -05:00
Michael R Sweet	4cb4ceaadd	Update docos with fixed codedoc.	2025-01-24 14:42:41 -05:00
Michael R Sweet	cca7383c73	Fix support for UTF-16 string values in dictionaries (Issue #92 ) Specifically to support Unicode Title and Author values.	2025-01-24 10:43:41 -05:00
Michael R Sweet	6c68b9fa5a	Add URLs and copyrights for Code 128 font and ProPhoto ICC profile (Issue #91 )	2025-01-24 09:56:51 -05:00
Michael R Sweet	dd7ed67ec1	Update makesrcdist to validate CHANGES.md.	2025-01-23 15:34:43 -05:00
Michael R Sweet	9e2f3aba10	Fix reading of compressed object streams (Issue #92 )	2025-01-23 15:27:22 -05:00