Update docos.

Fix static library build commands - remove archive before building it fresh.

CHANGES.md

@@ -2,19 +2,23 @@ Changes in PDFio
 ================
 
 
-v1.2.0 (Month DD, YYYY)
+v1.2.0 (January 24, 2024)
------------------------
+-------------------------
 
+- Now use autoconf to configure the PDFio sources (Issue #54)
 - Added `pdfioFileCreateNumberObj` and `pdfioFileCreateStringObj` functions
   (Issue #14)
 - Added `pdfioContentTextMeasure` function (Issue #17)
 - Added `pdfioContentTextNewLineShow` and `pdfioContentTextNewLineShowf`
   functions (Issue #24)
 - Renamed `pdfioContentTextNextLine` to `pdfioContentTextNewLine`.
-- Now use autoconf to configure the PDFio sources (Issue #54)
+- Updated the maximum number of object streams in a single file from 4096 to
+  8192 (Issue #58)
 - Updated the token reading code to protect against some obvious abuses of the
   PDF format.
 - Updated the xref reading code to protect against loops.
+- Updated the object handling code to use a binary insertion algorithm -
+  provides a significant (~800x) improvement in open times.
 - Fixed handling of encrypted PDFs with per-object file IDs (Issue #42)
 - Fixed handling of of trailer dictionaries that started immediately after the
   "trailer" keyword (Issue #58)

Makefile.in

@@ -172,6 +172,7 @@ valgrind:	testpdfio
 # pdfio library
 libpdfio.a:		$(LIBOBJS)
 	echo Archiving $@...
+	$(RM) $@
 	$(AR) $(ARFLAGS) $@ $(LIBOBJS)
 	$(RANLIB) $@
 
@@ -226,7 +227,7 @@ ttf.o:			ttf.h
 # Make documentation using Codedoc <https://www.msweet.org/codedoc>
 DOCFLAGS	=	\
 			--author "Michael R Sweet" \
-			--copyright "Copyright (c) 2021-2023 by Michael R Sweet" \
+			--copyright "Copyright (c) 2021-2024 by Michael R Sweet" \
 			--docversion $(PDFIO_VERSION)
 
 .PHONY: doc

NOTICE

@@ -1,6 +1,6 @@
 PDFio - PDF Read/Write Library
 
-Copyright © 2021-2023 by Michael R Sweet.
+Copyright © 2021-2024 by Michael R Sweet.
 
 (Optional) Exceptions to the Apache 2.0 License:
 ================================================

README.md

@@ -89,7 +89,7 @@ generates a static library that will be installed under "/usr/local" with:
 Legal Stuff
 -----------
 
-PDFio is Copyright © 2021-2023 by Michael R Sweet.
+PDFio is Copyright © 2021-2024 by Michael R Sweet.
 
 This software is licensed under the Apache License Version 2.0 with an
 (optional) exception to allow linking against GPL2/LGPL2 software.  See the

SECURITY.md

@@ -54,7 +54,7 @@ starting at 0.  A feature release has a "PATCH" value of 0, for example:
     1.1.0
     2.0.0
 
-Beta releases and release candidates are *not* prodution releases and use
+Beta releases and release candidates are *not* production releases and use
 semantic version numbers of the form:
 
     MAJOR.MINORbNUMBER

configure.ac

@@ -1,7 +1,7 @@
 dnl
 dnl Configuration script for PDFio
 dnl
-dnl Copyright © 2023 by Michael R Sweet
+dnl Copyright © 2023-2024 by Michael R Sweet
 dnl
 dnl Licensed under Apache License v2.0.  See the file "LICENSE" for more
 dnl information.
@@ -105,8 +105,8 @@ dnl ZLIB
 AC_MSG_CHECKING([for zlib via pkg-config])
 AS_IF([$PKGCONFIG --exists zlib], [
     AC_MSG_RESULT([yes])
-    LIBS="$($PKGCONFIG --libs zlib) $LIBS"
     CPPFLAGS="$($PKGCONFIG --cflags zlib) $CPPFLAGS"
+    LIBS="$($PKGCONFIG --libs zlib) $LIBS"
 ],[
     AC_MSG_RESULT([no])
     AC_CHECK_HEADER([zlib.h])
@@ -116,6 +116,7 @@ AS_IF([$PKGCONFIG --exists zlib], [
 	AC_MSG_ERROR([Sorry, this software requires zlib 1.1 or higher.])
     ])
 
+    PKGCONFIG_REQUIRES=""
     PKGCONFIG_LIBS_PRIVATE="-lz $PKGCONFIG_LIBS_PRIVATE"
 ])
 

doc/pdfio.3

@@ -1,4 +1,4 @@
-.TH pdfio 3 "pdf read/write library" "2023-12-05" "pdf read/write library"
+.TH pdfio 3 "pdf read/write library" "2024-01-24" "pdf read/write library"
 .SH NAME
 pdfio \- pdf read/write library
 .SH Introduction
@@ -34,7 +34,7 @@ PDFio is
 .I not
  concerned with rendering or viewing a PDF file, although a PDF RIP or viewer could be written using it.
 .PP
-PDFio is Copyright \[co] 2021\-2023 by Michael R Sweet and is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GPL2/LGPL2 software. See the files "LICENSE" and "NOTICE" for more information.
+PDFio is Copyright \[co] 2021\-2024 by Michael R Sweet and is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GPL2/LGPL2 software. See the files "LICENSE" and "NOTICE" for more information.
 .SS Requirements
 .PP
 PDFio requires the following to build the software:
@@ -3087,4 +3087,4 @@ typedef enum pdfio_valtype_e pdfio_valtype_t;
 Michael R Sweet
 .SH COPYRIGHT
 .PP
-Copyright (c) 2021-2023 by Michael R Sweet
+Copyright (c) 2021-2024 by Michael R Sweet

doc/pdfio.html

@@ -6,7 +6,7 @@
 <meta name="generator" content="codedoc v3.7">
 <meta name="author" content="Michael R Sweet">
 <meta name="language" content="en-US">
-<meta name="copyright" content="Copyright © 2021-2023 by Michael R Sweet">
+<meta name="copyright" content="Copyright © 2021-2024 by Michael R Sweet">
 <meta name="version" content="1.2.0">
 <style type="text/css"><!--
 body {
@@ -247,7 +247,7 @@ span.string {
 <p><img class="title" src="pdfio-512.png"></p>
 <h1 class="title">PDFio Programming Manual v1.2.0</h1>
 <p>Michael R Sweet</p>
-<p>Copyright © 2021-2023 by Michael R Sweet</p>
+<p>Copyright © 2021-2024 by Michael R Sweet</p>
 </div>
 <div class="contents">
 <h2 class="title">Contents</h2>
@@ -501,7 +501,7 @@ span.string {
 </li>
 </ul>
 <p>PDFio is <em>not</em> concerned with rendering or viewing a PDF file, although a PDF RIP or viewer could be written using it.</p>
-<p>PDFio is Copyright © 2021-2023 by Michael R Sweet and is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GPL2/LGPL2 software. See the files &quot;LICENSE&quot; and &quot;NOTICE&quot; for more information.</p>
+<p>PDFio is Copyright © 2021-2024 by Michael R Sweet and is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GPL2/LGPL2 software. See the files &quot;LICENSE&quot; and &quot;NOTICE&quot; for more information.</p>
 <h3 class="title" id="requirements">Requirements</h3>
 <p>PDFio requires the following to build the software:</p>
 <ul>

doc/pdfio.md

@@ -15,7 +15,7 @@ goals of pdfio are:
 PDFio is *not* concerned with rendering or viewing a PDF file, although a PDF
 RIP or viewer could be written using it.
 
-PDFio is Copyright © 2021-2023 by Michael R Sweet and is licensed under the
+PDFio is Copyright © 2021-2024 by Michael R Sweet and is licensed under the
 Apache License Version 2.0 with an (optional) exception to allow linking against
 GPL2/LGPL2 software.  See the files "LICENSE" and "NOTICE" for more information.
 

pdfio-array.c

@@ -327,6 +327,30 @@ pdfioArrayCreate(pdfio_file_t *pdf)	// I - PDF file
 }
 
 
+//
+// '_pdfioArrayDecrypt()' - Decrypt values in an array.
+//
+
+bool					// O - `true` on success, `false` on error
+_pdfioArrayDecrypt(pdfio_file_t  *pdf,	// I - PDF file
+                   pdfio_obj_t   *obj,	// I - Object
+                   pdfio_array_t *a,	// I - Array
+                   size_t        depth)	// I - Depth
+{
+  size_t		i;		// Looping var
+  _pdfio_value_t	*v;		// Current value
+
+
+  for (i = a->num_values, v = a->values; i > 0; i --, v ++)
+  {
+    if (!_pdfioValueDecrypt(pdf, obj, v, depth))
+      return (false);
+  }
+
+  return (true);
+}
+
+
 //
 // '_pdfioArrayDebug()' - Print the contents of an array.
 //

pdfio-content.c

@@ -1131,20 +1131,20 @@ pdfioContentTextMeasure(
       if (ch < 128)
       {
         // ASCII
-	*tempptr++ = ch;
+	*tempptr++ = (char)ch;
       }
       else if (ch < 2048)
       {
         // 2-byte UTF-8
-        *tempptr++ = 0xc0 | ((ch >> 6) & 0x1f);
+        *tempptr++ = (char)(0xc0 | ((ch >> 6) & 0x1f));
-        *tempptr++ = 0x80 | (ch & 0x3f);
+        *tempptr++ = (char)(0x80 | (ch & 0x3f));
       }
       else
       {
         // 3-byte UTF-8
-        *tempptr++ = 0xe0 | ((ch >> 12) & 0x0f);
+        *tempptr++ = (char)(0xe0 | ((ch >> 12) & 0x0f));
-        *tempptr++ = 0x80 | ((ch >> 6) & 0x3f);
+        *tempptr++ = (char)(0x80 | ((ch >> 6) & 0x3f));
-        *tempptr++ = 0x80 | (ch & 0x3f);
+        *tempptr++ = (char)(0x80 | (ch & 0x3f));
       }
     }
 
@@ -1152,7 +1152,7 @@ pdfioContentTextMeasure(
     s        = temp;
   }
 
-  ttfGetExtents(ttf, size, s, &extents);
+  ttfGetExtents(ttf, (float)size, s, &extents);
 
   return (extents.right - extents.left);
 }
@@ -1642,7 +1642,7 @@ pdfioFileCreateFontObjFromFile(
         *bufptr++ = (unsigned char)(cmap[i] >> 8);
         *bufptr++ = (unsigned char)(cmap[i] & 255);
 
-        glyphs[cmap[i]] = i;
+        glyphs[cmap[i]] = (unsigned short)i;
         if (cmap[i] < min_glyph)
           min_glyph = cmap[i];
         if (cmap[i] > max_glyph)
@@ -1727,9 +1727,9 @@ pdfioFileCreateFontObjFromFile(
     if ((w_array = pdfioArrayCreate(pdf)) == NULL)
       goto done;
 
-    for (start = 0, w0 = ttfGetWidth(font, 0), i = 1; i < 65536; start = i, w0 = w1, i ++)
+    for (start = 0, w0 = ttfGetWidth(font, 0), w1 = 0, i = 1; i < 65536; start = i, w0 = w1, i ++)
     {
-      while (i < 65536 && (w1 = ttfGetWidth(font, i)) == w0)
+      while (i < 65536 && (w1 = ttfGetWidth(font, (int)i)) == w0)
         i ++;
 
       if ((i - start) > 1)
@@ -1750,7 +1750,7 @@ pdfioFileCreateFontObjFromFile(
         pdfioArrayAppendNumber(temp_array, w0);
         for (w0 = w1, i ++; i < 65536; w0 = w1, i ++)
         {
-          if ((w1 = ttfGetWidth(font, i)) == w0 && i < 65535)
+          if ((w1 = ttfGetWidth(font, (int)i)) == w0 && i < 65535)
             break;
 
 	  pdfioArrayAppendNumber(temp_array, w0);

pdfio-crypto.c

@@ -399,7 +399,7 @@ _pdfioCryptoMakeRandom(uint8_t *buffer,	// I - Buffer
 //
 
 _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
- _pdfioCryptoMakeReader(
+_pdfioCryptoMakeReader(
      pdfio_file_t        *pdf,		// I  - PDF file
      pdfio_obj_t         *obj,		// I  - PDF object
      _pdfio_crypto_ctx_t *ctx,		// I  - Pointer to crypto context
@@ -409,11 +409,13 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
   uint8_t	data[21];		// Key data
   _pdfio_md5_t	md5;			// MD5 state
   uint8_t	digest[16];		// MD5 digest value
+#if PDFIO_OBJ_CRYPT
   pdfio_array_t	*id_array;		// Object ID array
   unsigned char	*id_value;		// Object ID value
   size_t	id_len;			// Length of object ID
-  uint8_t	*file_key,		// Computed file key to use
+  uint8_t	temp_key[16];		// File key for object
-		temp_key[16];		// File key for object
+#endif // PDFIO_OBJ_CRYPT
+  uint8_t	*file_key;		// Computed file key to use
 
 
   PDFIO_DEBUG("_pdfioCryptoMakeReader(pdf=%p, obj=%p(%d), ctx=%p, iv=%p, ivlen=%p(%d))\n", pdf, obj, (int)obj->number, ctx, iv, ivlen, (int)*ivlen);
@@ -425,6 +427,7 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
     return (NULL);
   }
 
+#if PDFIO_OBJ_CRYPT
   if ((id_array = pdfioDictGetArray(pdfioObjGetDict(obj), "ID")) != NULL)
   {
     // Object has its own ID that will get used for encryption...
@@ -434,6 +437,8 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
 		own_user_key[32],	// Calculated user key
 		pdf_user_key[32];	// Decrypted user key
 
+    PDFIO_DEBUG("_pdfioCryptoMakeReader: Per-object file ID.\n");
+
     if ((id_value = pdfioArrayGetBinary(id_array, 0, &id_len)) == NULL)
     {
       *ivlen = 0;
@@ -451,6 +456,8 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
 
     if (memcmp(own_user_key, pdf->user_key, sizeof(own_user_key)))
     {
+      PDFIO_DEBUG("_pdfioCryptoMakeReader: Not user password, trying owner password.\n");
+
       make_file_key(pdf->encryption, pdf->permissions, id_value, id_len, pdf->password, pdf->owner_key, temp_key);
       make_user_key(id_value, id_len, own_user_key);
 
@@ -467,6 +474,7 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
     file_key = temp_key;
   }
   else
+#endif // PDFIO_OBJ_CRYPT
   {
     // Use the default file key...
     file_key = pdf->file_key;
@@ -543,7 +551,7 @@ _pdfio_crypto_cb_t			// O  - Decryption callback or `NULL` for none
 //
 
 _pdfio_crypto_cb_t			// O  - Encryption callback or `NULL` for none
- _pdfioCryptoMakeWriter(
+_pdfioCryptoMakeWriter(
      pdfio_file_t        *pdf,		// I  - PDF file
      pdfio_obj_t         *obj,		// I  - PDF object
      _pdfio_crypto_ctx_t *ctx,		// I  - Pointer to crypto context
@@ -648,6 +656,8 @@ _pdfioCryptoUnlock(
   revision = (int)pdfioDictGetNumber(encrypt_dict, "R");
   length   = (int)pdfioDictGetNumber(encrypt_dict, "Length");
 
+  PDFIO_DEBUG("_pdfioCryptoUnlock: handler=%p(%s), version=%d, revision=%d, length=%d\n", (void *)handler, handler ? handler : "(null)", version, revision, length);
+
   if (!handler || strcmp(handler, "Standard"))
   {
     _pdfioFileError(pdf, "Unsupported security handler '%s'.", handler ? handler : "(null)");
@@ -696,6 +706,8 @@ _pdfioCryptoUnlock(
     }
     else
     {
+      PDFIO_DEBUG("_pdfioCryptoUnlock: CFM=\"%s\"\n", cfm);
+
       if (length < 40 || length > 128)
 	length = 128;			// Default to 128 bits
 

pdfio-dict.c

@@ -158,6 +158,30 @@ pdfioDictCreate(pdfio_file_t *pdf)	// I - PDF file
 }
 
 
+//
+// '_pdfioDictDecrypt()' - Decrypt the values in a dictionary.
+//
+
+bool					// O - `true` on success, `false` on error
+_pdfioDictDecrypt(pdfio_file_t *pdf,	// I - PDF file
+                  pdfio_obj_t  *obj,	// I - Object
+                  pdfio_dict_t *dict,	// I - Dictionary
+                  size_t       depth)	// I - Depth
+{
+  size_t	i;			// Looping var
+  _pdfio_pair_t	*pair;			// Current pair
+
+
+  for (i = dict->num_pairs, pair = dict->pairs; i > 0; i --, pair ++)
+  {
+    if (strcmp(pair->key, "ID") && !_pdfioValueDecrypt(pdf, obj, &pair->value, depth + 1))
+      return (false);
+  }
+
+  return (true);
+}
+
+
 //
 // '_pdfioDictDebug()' - Dump a dictionary to stderr.
 //
@@ -518,7 +542,7 @@ _pdfioDictRead(pdfio_file_t   *pdf,	// I - PDF file
   _pdfio_value_t	value;		// Dictionary value
 
 
-  PDFIO_DEBUG("_pdfioDictRead(pdf=%p)\n", pdf);
+  PDFIO_DEBUG("_pdfioDictRead(pdf=%p, obj=%p, tb=%p, depth=%lu)\n", pdf, obj, tb, (unsigned long)depth);
 
   // Create a dictionary and start reading...
   if ((dict = pdfioDictCreate(pdf)) == NULL)
@@ -530,6 +554,7 @@ _pdfioDictRead(pdfio_file_t   *pdf,	// I - PDF file
     if (!strcmp(key, ">>"))
     {
       // End of dictionary...
+      PDFIO_DEBUG("_pdfioDictRead: Returning dictionary value...\n");
       return (dict);
     }
     else if (key[0] != '/')
@@ -548,14 +573,14 @@ _pdfioDictRead(pdfio_file_t   *pdf,	// I - PDF file
 
     if (!_pdfioValueRead(pdf, obj, tb, &value, depth))
     {
-      _pdfioFileError(pdf, "Missing value for dictionary key.");
+      _pdfioFileError(pdf, "Missing value for dictionary key '%s'.", key + 1);
       break;
     }
 
     if (!_pdfioDictSetValue(dict, pdfioStringCreate(pdf, key + 1), &value))
       break;
 
-//    PDFIO_DEBUG("_pdfioDictRead: Set %s.\n", key);
+    PDFIO_DEBUG("_pdfioDictRead: Set %s.\n", key);
   }
 
   // Dictionary is invalid - pdfioFileClose will free the memory, return NULL

pdfio-file.c

@@ -19,7 +19,6 @@
 
 static pdfio_obj_t	*add_obj(pdfio_file_t *pdf, size_t number, unsigned short generation, off_t offset);
 static int		compare_objmaps(_pdfio_objmap_t *a, _pdfio_objmap_t *b);
-static int		compare_objs(pdfio_obj_t **a, pdfio_obj_t **b);
 static const char	*get_info_string(pdfio_file_t *pdf, const char *key);
 static bool		load_obj_stream(pdfio_obj_t *obj);
 static bool		load_pages(pdfio_file_t *pdf, pdfio_obj_t *obj, size_t depth);
@@ -916,21 +915,57 @@ pdfioFileFindObj(
     pdfio_file_t *pdf,			// I - PDF file
     size_t       number)		// I - Object number (1 to N)
 {
-  pdfio_obj_t	key,			// Search key
+  size_t	left,			// Left object
-		*keyptr,		// Pointer to key
+		right,			// Right object
-		**match;		// Pointer to match
+		current;		// Current object
 
 
-  if (pdf->num_objs > 0)
+  PDFIO_DEBUG("pdfioFileFindObj(pdf=%p, number=%lu) alloc_objs=%lu, num_objs=%lu, objs=%p\n", (void *)pdf, (unsigned long)number, (unsigned long)(pdf ? pdf->alloc_objs : 0), (unsigned long)(pdf ? pdf->num_objs : 0), (void *)(pdf ? pdf->objs : NULL));
+
+  // Range check input...
+  if (!pdf || pdf->num_objs == 0 || number < 1)
+    return (NULL);
+
+  // Do a binary search for the object...
+  if ((current = number - 1) >= pdf->num_objs)
+    current = pdf->num_objs / 2;
+
+  PDFIO_DEBUG("pdfioFileFindObj: objs[current=%lu]=%p\n", (unsigned long)current, (void *)pdf->objs[current]);
+
+  if (number == pdf->objs[current]->number)
   {
-    key.number = number;
+    // Fast match...
-    keyptr     = &key;
+    return (pdf->objs[current]);
-    match      = (pdfio_obj_t **)bsearch(&keyptr, pdf->objs, pdf->num_objs, sizeof(pdfio_obj_t *), (int (*)(const void *, const void *))compare_objs);
+  }
-
+  else if (number < pdf->objs[current]->number)
-    return (match ? *match : NULL);
+  {
+    left  = 0;
+    right = current;
+  }
+  else
+  {
+    left  = current;
+    right = pdf->num_objs - 1;
   }
 
-  return (NULL);
+  while ((right - left) > 1)
+  {
+    current = (left + right) / 2;
+
+    if (number == pdf->objs[current]->number)
+      return (pdf->objs[current]);
+    else if (number < pdf->objs[current]->number)
+      right = current;
+    else
+      left = current;
+  }
+
+  if (number == pdf->objs[left]->number)
+    return (pdf->objs[left]);
+  else if (number == pdf->objs[right]->number)
+    return (pdf->objs[right]);
+  else
+    return (NULL);
 }
 
 
@@ -1385,6 +1420,9 @@ add_obj(pdfio_file_t   *pdf,		// I - PDF file
 	off_t          offset)		// I - Offset in file
 {
   pdfio_obj_t	*obj;			// Object
+  size_t	left,			// Left object
+		right,			// Right object
+		current;		// Current object (center)
 
 
   // Allocate memory for the object...
@@ -1410,8 +1448,6 @@ add_obj(pdfio_file_t   *pdf,		// I - PDF file
     pdf->alloc_objs += 32;
   }
 
-  pdf->objs[pdf->num_objs ++] = obj;
-
   obj->pdf        = pdf;
   obj->number     = number;
   obj->generation = generation;
@@ -1419,9 +1455,56 @@ add_obj(pdfio_file_t   *pdf,		// I - PDF file
 
   PDFIO_DEBUG("add_obj: obj=%p, ->pdf=%p, ->number=%lu, ->offset=%lu\n", obj, pdf, (unsigned long)obj->number, (unsigned long)offset);
 
-  // Re-sort object array as needed...
+  // Insert object into array as needed...
-  if (pdf->num_objs > 1 && pdf->objs[pdf->num_objs - 2]->number > number)
+  if (pdf->num_objs == 0 || obj->number > pdf->objs[pdf->num_objs - 1]->number)
-    qsort(pdf->objs, pdf->num_objs, sizeof(pdfio_obj_t *), (int (*)(const void *, const void *))compare_objs);
+  {
+    // Append object...
+    PDFIO_DEBUG("add_obj: Appending at %lu\n", (unsigned long)pdf->num_objs);
+
+    pdf->objs[pdf->num_objs] = obj;
+    pdf->last_obj            = pdf->num_objs;
+  }
+  else
+  {
+    // Insert object...
+    if (obj->number < pdf->objs[pdf->last_obj]->number)
+    {
+      left  = 0;
+      right = pdf->last_obj;
+    }
+    else
+    {
+      left  = pdf->last_obj;
+      right = pdf->num_objs - 1;
+    }
+
+    while ((right - left) > 1)
+    {
+      current = (left + right) / 2;
+
+      if (obj->number < pdf->objs[current]->number)
+        right = current;
+      else
+        left = current;
+    }
+
+    if (obj->number < pdf->objs[left]->number)
+      current = left;
+    else if (obj->number < pdf->objs[right]->number)
+      current = right;
+    else
+      current = right;
+
+    PDFIO_DEBUG("add_obj: Inserting at %lu\n", (unsigned long)current);
+
+    if (current < pdf->num_objs)
+      memmove(pdf->objs + current + 1, pdf->objs + current, (pdf->num_objs - current) * sizeof(pdfio_obj_t *));
+
+    pdf->objs[current] = obj;
+    pdf->last_obj      = current;
+  }
+
+  pdf->num_objs ++;
 
   return (obj);
 }
@@ -1448,23 +1531,6 @@ compare_objmaps(_pdfio_objmap_t *a,	// I - First object map
 }
 
 
-//
-// 'compare_objs()' - Compare the object numbers of two objects.
-//
-
-static int				// O - Result of comparison
-compare_objs(pdfio_obj_t **a,		// I - First object
-             pdfio_obj_t **b)		// I - Second object
-{
-  if ((*a)->number < (*b)->number)
-    return (-1);
-  else if ((*a)->number == (*b)->number)
-    return (0);
-  else
-    return (1);
-}
-
-
 //
 // 'get_info_string()' - Get a string value from the Info dictionary.
 //
@@ -1737,7 +1803,7 @@ load_xref(
       pdfio_stream_t	*st;		// Stream
       unsigned char	buffer[32];	// Read buffer
       size_t		num_sobjs = 0,	// Number of object streams
-			sobjs[4096];	// Object streams to load
+			sobjs[8192];	// Object streams to load
       pdfio_obj_t	*current;	// Current object
 
       if ((number = strtoimax(line, &ptr, 10)) < 1)

pdfio-object.c

@@ -453,9 +453,7 @@ _pdfioObjLoad(pdfio_obj_t *obj)		// I - Object
     return (false);
   }
 
-  PDFIO_DEBUG("_pdfioObjLoad: tb.bufptr=%p, tb.bufend=%p, tb.bufptr[0]=0x%02x, tb.bufptr[0]=0x%02x\n", tb.bufptr, tb.bufend, tb.bufptr[0], tb.bufptr[1]);
+  PDFIO_DEBUG("_pdfioObjLoad: tb.bufptr=%p, tb.bufend=%p, tb.bufptr[0]=0x%02x, tb.bufptr[1]=0x%02x\n", tb.bufptr, tb.bufend, tb.bufptr[0], tb.bufptr[1]);
-  if (tb.bufptr && tb.bufptr < tb.bufend && (tb.bufptr[0] == 0x0d || tb.bufptr[0] == 0x0a))
-    tb.bufptr ++;			// Skip trailing CR or LF after token
 
   _pdfioTokenFlush(&tb);
 
@@ -466,6 +464,18 @@ _pdfioObjLoad(pdfio_obj_t *obj)		// I - Object
     PDFIO_DEBUG("_pdfioObjLoad: stream_offset=%lu.\n", (unsigned long)obj->stream_offset);
   }
 
+  // Decrypt as needed...
+  if (obj->pdf->encryption)
+  {
+    PDFIO_DEBUG("_pdfioObjLoad: Decrypting value...\n");
+
+    if (!_pdfioValueDecrypt(obj->pdf, obj, &obj->value, 0))
+    {
+      PDFIO_DEBUG("_pdfioObjLoad: Failed to decrypt.\n");
+      return (false);
+    }
+  }
+
   PDFIO_DEBUG("_pdfioObjLoad: ");
   PDFIO_DEBUG_VALUE(&obj->value);
   PDFIO_DEBUG("\n");

pdfio-private.h

@@ -266,7 +266,8 @@ struct _pdfio_file_s			// PDF file structure
 		alloc_dicts;		// Allocated dictionaries
   pdfio_dict_t	**dicts;		// Dictionaries
   size_t	num_objs,		// Number of objects
-		alloc_objs;		// Allocated objects
+		alloc_objs,		// Allocated objects
+		last_obj;		// Last object added
   pdfio_obj_t	**objs,			// Objects
 		*current_obj;		// Current object being written/read
   size_t	num_objmaps,		// Number of object maps
@@ -321,6 +322,7 @@ struct _pdfio_stream_s			// Stream
 // Functions...
 //
 
+extern bool		_pdfioArrayDecrypt(pdfio_file_t *pdf, pdfio_obj_t *obj, pdfio_array_t *a, size_t depth) _PDFIO_INTERNAL;
 extern void		_pdfioArrayDebug(pdfio_array_t *a, FILE *fp) _PDFIO_INTERNAL;
 extern void		_pdfioArrayDelete(pdfio_array_t *a) _PDFIO_INTERNAL;
 extern _pdfio_value_t	*_pdfioArrayGetValue(pdfio_array_t *a, size_t n) _PDFIO_INTERNAL;
@@ -345,6 +347,7 @@ extern void		_pdfioCryptoSHA256Finish(_pdfio_sha256_t *ctx, uint8_t *Message_Dig
 extern bool		_pdfioCryptoUnlock(pdfio_file_t *pdf, pdfio_password_cb_t password_cb, void *password_data) _PDFIO_INTERNAL;
 
 extern void		_pdfioDictClear(pdfio_dict_t *dict, const char *key) _PDFIO_INTERNAL;
+extern bool		_pdfioDictDecrypt(pdfio_file_t *pdf, pdfio_obj_t *obj, pdfio_dict_t *dict, size_t depth) _PDFIO_INTERNAL;
 extern void		_pdfioDictDebug(pdfio_dict_t *dict, FILE *fp) _PDFIO_INTERNAL;
 extern void		_pdfioDictDelete(pdfio_dict_t *dict) _PDFIO_INTERNAL;
 extern _pdfio_value_t	*_pdfioDictGetValue(pdfio_dict_t *dict, const char *key) _PDFIO_INTERNAL;
@@ -388,6 +391,7 @@ extern void		_pdfioTokenPush(_pdfio_token_t *tb, const char *token) _PDFIO_INTER
 extern bool		_pdfioTokenRead(_pdfio_token_t *tb, char *buffer, size_t bufsize);
 
 extern _pdfio_value_t	*_pdfioValueCopy(pdfio_file_t *pdfdst, _pdfio_value_t *vdst, pdfio_file_t *pdfsrc, _pdfio_value_t *vsrc) _PDFIO_INTERNAL;
+extern bool		_pdfioValueDecrypt(pdfio_file_t *pdf, pdfio_obj_t *obj, _pdfio_value_t *v, size_t depth) _PDFIO_INTERNAL;
 extern void		_pdfioValueDebug(_pdfio_value_t *v, FILE *fp) _PDFIO_INTERNAL;
 extern void		_pdfioValueDelete(_pdfio_value_t *v) _PDFIO_INTERNAL;
 extern _pdfio_value_t	*_pdfioValueRead(pdfio_file_t *pdf, pdfio_obj_t *obj, _pdfio_token_t *ts, _pdfio_value_t *v, size_t depth) _PDFIO_INTERNAL;

pdfio-stream.c

@@ -1066,7 +1066,7 @@ stream_read(pdfio_stream_t *st,		// I - Stream
 
       if ((status = inflate(&(st->flate), Z_NO_FLUSH)) < Z_OK)
       {
-	_pdfioFileError(st->pdf, "Unable to decompress stream data: %s", zstrerror(status));
+	_pdfioFileError(st->pdf, "Unable to decompress stream data for object %ld: %s", (long)st->obj->number, zstrerror(status));
 	return (-1);
       }
       else if (avail_in == st->flate.avail_in && avail_out == st->flate.avail_out)
@@ -1127,7 +1127,7 @@ stream_read(pdfio_stream_t *st,		// I - Stream
 
 	if ((status = inflate(&(st->flate), Z_NO_FLUSH)) < Z_OK)
 	{
-	  _pdfioFileError(st->pdf, "Unable to decompress stream data: %s", zstrerror(status));
+	  _pdfioFileError(st->pdf, "Unable to decompress stream data for object %ld: %s", (long)st->obj->number, zstrerror(status));
 	  return (-1);
 	}
 	else if (status == Z_STREAM_END || (avail_in == st->flate.avail_in && avail_out == st->flate.avail_out))
@@ -1197,7 +1197,7 @@ stream_read(pdfio_stream_t *st,		// I - Stream
 
 	if ((status = inflate(&(st->flate), Z_NO_FLUSH)) < Z_OK)
 	{
-	  _pdfioFileError(st->pdf, "Unable to decompress stream data: %s", zstrerror(status));
+	  _pdfioFileError(st->pdf, "Unable to decompress stream data for object %ld: %s", (long)st->obj->number, zstrerror(status));
 	  return (-1);
 	}
 	else if (status == Z_STREAM_END || (avail_in == st->flate.avail_in && avail_out == st->flate.avail_out))

pdfio-token.c

@@ -447,6 +447,13 @@ _pdfioTokenRead(_pdfio_token_t *tb,	// I - Token buffer/stack
 	    return (false);
 	  }
 	}
+
+	if (ch == '\r')
+	{
+	  // Look for a trailing LF
+	  if ((ch = get_char(tb)) != EOF && ch != '\n')
+	    tb->bufptr --;
+	}
 	break;
 
     case 'N' : // number

pdfio-value.c

@@ -10,6 +10,13 @@
 #include "pdfio-private.h"
 
 
+//
+// Local functions...
+//
+
+static time_t	get_date_time(const char *s);
+
+
 //
 // '_pdfioValueCopy()' - Copy a value to a PDF file.
 //
@@ -105,6 +112,101 @@ _pdfioValueCopy(pdfio_file_t   *pdfdst,	// I - Destination PDF file
 }
 
 
+//
+// '_pdfioValueDecrypt()' - Decrypt a value.
+//
+
+bool					// O - `true` on success, `false` on error
+_pdfioValueDecrypt(pdfio_file_t   *pdf,	// I - PDF file
+                   pdfio_obj_t    *obj,	// I - Object
+                   _pdfio_value_t *v,	// I - Value
+                   size_t         depth)// I - Depth
+{
+  _pdfio_crypto_ctx_t	ctx;		// Decryption context
+  _pdfio_crypto_cb_t	cb;		// Decryption callback
+  size_t		ivlen;		// Number of initialization vector bytes
+  uint8_t		temp[32768];	// Temporary buffer for decryption
+  size_t		templen;	// Number of actual data bytes
+  time_t		timeval;	// Date/time value
+
+
+  if (depth > PDFIO_MAX_DEPTH)
+  {
+    _pdfioFileError(pdf, "Value too deep.");
+    return (false);
+  }
+
+  switch (v->type)
+  {
+    default :
+        // Do nothing
+        break;
+
+    case PDFIO_VALTYPE_ARRAY :
+        return (_pdfioArrayDecrypt(pdf, obj, v->value.array, depth + 1));
+        break;
+
+    case PDFIO_VALTYPE_DICT :
+        return (_pdfioDictDecrypt(pdf, obj, v->value.dict, depth + 1));
+        break;
+
+    case PDFIO_VALTYPE_BINARY :
+	// Decrypt the binary string...
+	if (v->value.binary.datalen > (sizeof(temp) - 32))
+	{
+	  _pdfioFileError(pdf, "Unable to read encrypted binary string - too long.");
+	  return (false);
+	}
+
+	ivlen = v->value.binary.datalen;
+	if ((cb = _pdfioCryptoMakeReader(pdf, obj, &ctx, v->value.binary.data, &ivlen)) == NULL)
+	  return (false);
+
+	templen = (cb)(&ctx, temp, v->value.binary.data + ivlen, v->value.binary.datalen - ivlen);
+
+	// Copy the decrypted string back to the value and adjust the length...
+	memcpy(v->value.binary.data, temp, templen);
+
+	if (pdf->encryption >= PDFIO_ENCRYPTION_AES_128)
+	  v->value.binary.datalen = templen - temp[templen - 1];
+	else
+	  v->value.binary.datalen = templen;
+	break;
+
+    case PDFIO_VALTYPE_STRING :
+        // Decrypt regular string...
+        templen = strlen(v->value.string);
+	if (templen > (sizeof(temp) - 33))
+	{
+	  _pdfioFileError(pdf, "Unable to read encrypted string - too long.");
+	  return (false);
+	}
+
+	ivlen = templen;
+	if ((cb = _pdfioCryptoMakeReader(pdf, obj, &ctx, (uint8_t *)v->value.string, &ivlen)) == NULL)
+	  return (false);
+
+	templen = (cb)(&ctx, temp, (uint8_t *)v->value.string + ivlen, templen - ivlen);
+	temp[templen] = '\0';
+
+        if ((timeval = get_date_time((char *)temp)) != 0)
+        {
+          // Change the type to date...
+          v->type       = PDFIO_VALTYPE_DATE;
+          v->value.date = timeval;
+        }
+        else
+        {
+          // Copy the decrypted string back to the value...
+	  v->value.string = pdfioStringCreate(pdf, (char *)temp);
+	}
+        break;
+  }
+
+  return (true);
+}
+
+
 //
 // '_pdfioValueDebug()' - Print the contents of a value.
 //
@@ -196,6 +298,7 @@ _pdfioValueRead(pdfio_file_t   *pdf,	// I - PDF file
                 size_t         depth)	// I - Depth of value
 {
   char		token[32768];		// Token buffer
+  time_t	timeval;		// Date/time value
 #ifdef DEBUG
   static const char * const valtypes[] =
   {
@@ -245,73 +348,10 @@ _pdfioValueRead(pdfio_file_t   *pdf,	// I - PDF file
     if ((v->value.dict = _pdfioDictRead(pdf, obj, tb, depth + 1)) == NULL)
       return (NULL);
   }
-  else if (!strncmp(token, "(D:", 3))
+  else if (!strncmp(token, "(D:", 3) && (timeval = get_date_time(token + 1)) != 0)
   {
-    // Possible date value of the form:
+    v->type       = PDFIO_VALTYPE_DATE;
-    //
+    v->value.date = timeval;
-    //   (D:YYYYMMDDhhmmssZ)
-    //   (D:YYYYMMDDhhmmss+HH'mm)
-    //   (D:YYYYMMDDhhmmss-HH'mm)
-    //
-    int		i;			// Looping var
-    struct tm	dateval;		// Date value
-    int		offset;			// Date offset
-
-    for (i = 3; i < 17; i ++)
-    {
-      if (!isdigit(token[i] & 255))
-        break;
-    }
-
-    if (i >= 17)
-    {
-      if (token[i] == 'Z')
-      {
-        i ++;
-      }
-      else if (token[i] == '-' || token[i] == '+')
-      {
-        if (isdigit(token[i + 1] & 255) && isdigit(token[i + 2] & 255) && token[i + 3] == '\'' && isdigit(token[i + 4] & 255) && isdigit(token[i + 5] & 255))
-        {
-          i += 6;
-          if (token[i] == '\'')
-            i ++;
-	}
-      }
-    }
-
-    if (token[i])
-    {
-      // Just a string...
-      v->type         = PDFIO_VALTYPE_STRING;
-      v->value.string = pdfioStringCreate(pdf, token + 1);
-    }
-    else
-    {
-      // Date value...
-      memset(&dateval, 0, sizeof(dateval));
-
-      dateval.tm_year = (token[3] - '0') * 1000 + (token[4] - '0') * 100 + (token[5] - '0') * 10 + token[6] - '0' - 1900;
-      dateval.tm_mon  = (token[7] - '0') * 10 + token[8] - '0' - 1;
-      dateval.tm_mday = (token[9] - '0') * 10 + token[10] - '0';
-      dateval.tm_hour = (token[11] - '0') * 10 + token[12] - '0';
-      dateval.tm_min  = (token[13] - '0') * 10 + token[14] - '0';
-      dateval.tm_sec  = (token[15] - '0') * 10 + token[16] - '0';
-
-      if (token[17] == 'Z')
-      {
-        offset = 0;
-      }
-      else
-      {
-        offset = (token[18] - '0') * 600 + (token[19] - '0') * 60 + (token[20] - '0') * 10 + token[21] - '0';
-        if (token[17] == '-')
-          offset = -offset;
-      }
-
-      v->type       = PDFIO_VALTYPE_DATE;
-      v->value.date = mktime(&dateval) + offset;
-    }
   }
   else if (token[0] == '(')
   {
@@ -363,36 +403,6 @@ _pdfioValueRead(pdfio_file_t   *pdf,	// I - PDF file
 
       *dataptr++ = (unsigned char)d;
     }
-
-    if (obj && pdf->encryption)
-    {
-      // Decrypt the string...
-      _pdfio_crypto_ctx_t ctx;		// Decryption context
-      _pdfio_crypto_cb_t cb;		// Decryption callback
-      size_t	ivlen;			// Number of initialization vector bytes
-      uint8_t	temp[32768];		// Temporary buffer for decryption
-      size_t	templen;		// Number of actual data bytes
-
-      if (v->value.binary.datalen > (sizeof(temp) - 32))
-      {
-	_pdfioFileError(pdf, "Unable to read encrypted binary string - too long.");
-	return (false);
-      }
-
-      ivlen = v->value.binary.datalen;
-      if ((cb = _pdfioCryptoMakeReader(pdf, obj, &ctx, v->value.binary.data, &ivlen)) == NULL)
-	return (false);
-
-      templen = (cb)(&ctx, temp, v->value.binary.data + ivlen, v->value.binary.datalen - ivlen);
-
-      // Copy the decrypted string back to the value and adjust the length...
-      memcpy(v->value.binary.data, temp, templen);
-
-      if (pdf->encryption >= PDFIO_ENCRYPTION_AES_128)
-        v->value.binary.datalen = templen - temp[templen - 1];
-      else
-	v->value.binary.datalen = templen;
-    }
   }
   else if (strchr("0123456789-+.", token[0]) != NULL)
   {
@@ -728,3 +738,76 @@ _pdfioValueWrite(pdfio_file_t   *pdf,	// I - PDF file
 
   return (false);
 }
+
+
+//
+// 'get_date_time()' - Convert PDF date/time value to time_t.
+//
+
+static time_t				// O - Time in seconds
+get_date_time(const char *s)		// I - PDF date/time value
+{
+  int		i;			// Looping var
+  struct tm	dateval;		// Date value
+  int		offset;			// Date offset
+
+
+  // Possible date value of the form:
+  //
+  //   (D:YYYYMMDDhhmmssZ)
+  //   (D:YYYYMMDDhhmmss+HH'mm)
+  //   (D:YYYYMMDDhhmmss-HH'mm)
+  //
+
+  for (i = 2; i < 16; i ++)
+  {
+    if (!isdigit(s[i] & 255) || !s[i])
+      break;
+  }
+
+  if (i >= 16)
+  {
+    if (s[i] == 'Z')
+    {
+      i ++;
+    }
+    else if (s[i] == '-' || s[i] == '+')
+    {
+      if (isdigit(s[i + 1] & 255) && isdigit(s[i + 2] & 255) && s[i + 3] == '\'' && isdigit(s[i + 4] & 255) && isdigit(s[i + 5] & 255))
+      {
+	i += 6;
+	if (s[i] == '\'')
+	  i ++;
+      }
+    }
+  }
+
+  if (s[i])
+  {
+    // Just a string...
+    return (0);
+  }
+
+  // Date value...
+  memset(&dateval, 0, sizeof(dateval));
+
+  dateval.tm_year = (s[2] - '0') * 1000 + (s[3] - '0') * 100 + (s[4] - '0') * 10 + s[5] - '0' - 1900;
+  dateval.tm_mon  = (s[6] - '0') * 10 + s[7] - '0' - 1;
+  dateval.tm_mday = (s[8] - '0') * 10 + s[9] - '0';
+  dateval.tm_hour = (s[10] - '0') * 10 + s[11] - '0';
+  dateval.tm_min  = (s[12] - '0') * 10 + s[13] - '0';
+  dateval.tm_sec  = (s[14] - '0') * 10 + s[15] - '0';
+
+  if (s[16] == 'Z')
+  {
+    offset = 0;
+  }
+  else
+  {
+    offset = (s[17] - '0') * 600 + (s[18] - '0') * 60 + (s[19] - '0') * 10 + s[20] - '0';
+    if (s[16] == '-')
+      offset = -offset;
+  }
+
+  return (mktime(&dateval) + offset);
+}

pdfio_native.nuspec

@@ -13,7 +13,7 @@
         <requireLicenseAcceptance>false</requireLicenseAcceptance>
         <description>PDFio Library for VS2019+</description>
         <summary>PDFio is a simple C library for reading and writing PDF files. PDFio is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GNU GPL2-only software.</summary>
-        <copyright>Copyright © 2019-2023 by Michael R Sweet</copyright>
+        <copyright>Copyright © 2019-2024 by Michael R Sweet</copyright>
         <tags>pdf file native</tags>
 	<dependencies>
 	    <dependency id="pdfio_native.redist" version="1.2.0" />

pdfio_native.redist.nuspec

@@ -13,7 +13,7 @@
         <requireLicenseAcceptance>false</requireLicenseAcceptance>
         <description>PDFio Library for VS2019+</description>
         <summary>PDFio is a simple C library for reading and writing PDF files. This package provides the redistributable content for the PDFio library. PDFio is licensed under the Apache License Version 2.0 with an (optional) exception to allow linking against GNU GPL2-only software.</summary>
-        <copyright>Copyright © 2019-2023 by Michael R Sweet</copyright>
+        <copyright>Copyright © 2019-2024 by Michael R Sweet</copyright>
         <tags>pdf file native</tags>
 	<dependencies>
 	    <dependency id="zlib_native.redist" version="1.2.11" />

ttf.c

@@ -1307,7 +1307,7 @@ read_cmap(ttf_t *font)			// I - Font
               {
                 // Use an "obscure indexing trick" (words from the spec, not
                 // mine) to look up the glyph index...
-                temp = segment->idRangeOffset / 2 - segCount + (ch - segment->startCode) + (segment - segments);
+                temp = (int)(segment->idRangeOffset / 2 - segCount + (ch - segment->startCode) + (segment - segments));
 
                 TTF_DEBUG("read_cmap: ch=%d, temp=%d\n", ch, temp);
                 if (temp < 0 || temp >= numGlyphIdArray)