Add metadata support.

Add UNICODE_VALUE define to allow toggling between Unicode and ISO-8859-1 modes.

.github/workflows/build.yml

@@ -60,3 +60,5 @@ jobs:
       run: nuget restore pdfio.sln
     - name: Build PDFio
       run: msbuild pdfio.sln
+    - name: Test PDFio
+      run: .\runtests.bat x64\Debug

.gitignore

@@ -1,5 +1,6 @@
 *.1.dylib
 *.a
+*.dSYM
 *.log
 *.o
 *.so.1
@@ -8,7 +9,10 @@
 /autom4te.cache
 /config.log
 /config.status
+/configure~
 /doc/pdfio.epub
+/examples/code128
+/examples/md2pdf
 /Makefile
 /packages
 /pdfio.pc

CHANGES.md

@@ -2,8 +2,47 @@ Changes in PDFio
 ================
 
 
-v1.2.0 (January 24, 2024)
--------------------------
+v1.4.0 - YYYY-MM-DD
+-------------------
+
+- Added new `pdfioDictGetKey` and `pdfioDictGetNumPairs` APIs (Issue #63)
+- Added new `pdfioArrayRemove` and `pdfioDictClear` APIs (Issue #74)
+- Added new `pdfioFileCreateNameObj` and `pdfioObjGetName` APIs for creating and
+  getting name object values (Issue #76)
+- Updated documentation (Issue #78)
+- Fixed reading of PDF files whose trailer is missing a newline (Issue #80)
+- Fixed builds with some versions of VC++ (Issue #81)
+
+
+v1.3.2 - 2024-08-15
+-------------------
+
+- Added some more sanity checks to the TrueType font reader.
+- Updated documentation (Issue #77)
+- Fixed an issue when opening certain encrypted PDF files (Issue #62)
+
+
+v1.3.1 - 2024-08-05
+-------------------
+
+- CVE 2024-42358: Updated TrueType font reader to avoid large memory
+  allocations.
+- Fixed some documentation errors and added examples (Issue #68, Issue #69)
+
+
+v1.3.0 - 2024-06-28
+-------------------
+
+- Added `pdfioFileGetCatalog` API for accessing the root/catalog object of a
+  PDF file (Issue #67)
+- Updated number support to avoid locale issues (Issue #61)
+- Updated the PDFio private header to allow compilation with MingW; note that
+  MingW is NOT a supported toolchain for PDFio (Issue #66)
+- Optimized string pool code.
+
+
+v1.2.0 - 2024-01-24
+-------------------
 
 - Now use autoconf to configure the PDFio sources (Issue #54)
 - Added `pdfioFileCreateNumberObj` and `pdfioFileCreateStringObj` functions
@@ -26,8 +65,8 @@ v1.2.0 (January 24, 2024)
   65536 in the xref table (Issue #59)
 
 
-v1.1.4 (December 3, 2023)
--------------------------
+v1.1.4 - 2023-12-03
+-------------------
 
 - Fixed detection of encrypted strings that are too short (Issue #52)
 - Fixed a TrueType CMAP decoding bug.
@@ -35,15 +74,15 @@ v1.1.4 (December 3, 2023)
 - Added a ToUnicode map for Unicode text to support text copying.
 
 
-v1.1.3 (November 15, 2023)
---------------------------
+v1.1.3 - 2023-11-15
+-------------------
 
 - Fixed Unicode font support (Issue #16)
 - Fixed missing initializer for 40-bit RC4 encryption (Issue #51)
 
 
-v1.1.2 (October 10, 2023)
--------------------------
+v1.1.2 - 2023-10-10
+-------------------
 
 - Updated `pdfioContentSetDashPattern` to support setting a solid (0 length)
   dash pattern (Issue #41)
@@ -58,15 +97,15 @@ v1.1.2 (October 10, 2023)
   (Issue #48)
 
 
-v1.1.1 (March 20, 2023)
------------------------
+v1.1.1 - 2023-03-20
+-------------------
 
 - CVE-2023-28428: Fixed a potential denial-of-service with corrupt PDF files.
 - Fixed a few build issues.
 
 
-v1.1.0 (February 6, 2023)
--------------------------
+v1.1.0 - 2023-02-06
+-------------------
 
 - CVE-2023-24808: Fixed a potential denial-of-service with corrupt PDF files.
 - Added `pdfioFileCreateTemporary` function (Issue #29)
@@ -80,28 +119,28 @@ v1.1.0 (February 6, 2023)
 - Fixed `pdfioContentMatrixRotate` function.
 
 
-v1.0.1 (March 2, 2022)
-----------------------
+v1.0.1 - 2022-03-02
+-------------------
 
 - Added missing `pdfioPageGetNumStreams` and `pdfioPageOpenStream` functions.
 - Added demo pdfiototext utility.
 - Fixed bug in `pdfioStreamGetToken`.
 
 
-v1.0.0 (December 14, 2021)
---------------------------
+v1.0.0 - 2021-12-14
+-------------------
 
 - First stable release.
 
 
-v1.0rc1 (November 30, 2021)
----------------------------
+v1.0rc1 - 2021-11-30
+--------------------
 
 - Fixed a few stack/buffer overflow bugs discovered via fuzzing.
 
 
-v1.0b2 (November 7, 2021)
--------------------------
+v1.0b2 - 2021-11-07
+-------------------
 
 - Added `pdfioFileCreateOutput` API to support streaming output of PDF
   (Issue #21)
@@ -112,7 +151,7 @@ v1.0b2 (November 7, 2021)
 - Fixed some issues identified by a Coverity scan.
 
 
-v1.0b1 (August 30, 2021)
-------------------------
+v1.0b1 - 2021-08-30
+-------------------
 
 - Initial release

EXAMPLES.md

@@ -0,0 +1,18 @@
+PDFio Examples
+==============
+
+The "examples" subdirectory contains example code showing how to do different
+things with PDFio.
+
+
+code128.c
+---------
+
+This example shows how to embed and use a barcode font.
+
+
+md2pdf.c
+--------
+
+This example shows how to generate pages with multiple fonts, embedded images,
+and headers and footers.

Makefile.in

@@ -1,7 +1,7 @@
 #
 # Makefile for PDFio.
 #
-# Copyright © 2021-2023 by Michael R Sweet.
+# Copyright © 2021-2024 by Michael R Sweet.
 #
 # Licensed under Apache License v2.0.  See the file "LICENSE" for more
 # information.
@@ -163,7 +163,9 @@ install:	$(TARGETS)
 # Test everything
 test:	testpdfio testttf
 	./testttf 2>test.log
-	./testpdfio 2>test.log
+	./testpdfio 2>>test.log
+	LANG=fr_FR.UTF-8 ./testpdfio 2>>test.log
+
 
 valgrind:	testpdfio
 	valgrind --leak-check=full ./testpdfio
@@ -256,5 +258,5 @@ clang:
 
 # Analyze code using Cppcheck <http://cppcheck.sourceforge.net>
 cppcheck:
-	cppcheck $(CPPFLAGS) --template=gcc --addon=cert.py --suppressions-list=.cppcheck $(OBJS:.o=.c) 2>cppcheck.log
+	cppcheck $(CPPFLAGS) --template=gcc --suppressions-list=.cppcheck $(OBJS:.o=.c) 2>cppcheck.log
 	test -s cppcheck.log && (echo "$(GHA_ERROR)Cppcheck detected issues."; echo ""; cat cppcheck.log; exit 1) || exit 0

configure

@@ -1,6 +1,6 @@
 #! /bin/sh
 # Guess values for system-dependent variables and create Makefiles.
-# Generated by GNU Autoconf 2.71 for pdfio 1.2.0.
+# Generated by GNU Autoconf 2.71 for pdfio 1.4.0.
 #
 # Report bugs to <https://github.com/michaelrsweet/pdfio/issues>.
 #
@@ -610,8 +610,8 @@ MAKEFLAGS=
 # Identity of this package.
 PACKAGE_NAME='pdfio'
 PACKAGE_TARNAME='pdfio'
-PACKAGE_VERSION='1.2.0'
-PACKAGE_STRING='pdfio 1.2.0'
+PACKAGE_VERSION='1.4.0'
+PACKAGE_STRING='pdfio 1.4.0'
 PACKAGE_BUGREPORT='https://github.com/michaelrsweet/pdfio/issues'
 PACKAGE_URL='https://www.msweet.org/pdfio'
 
@@ -1293,7 +1293,7 @@ if test "$ac_init_help" = "long"; then
   # Omit some internal or obsolete options to make the list less imposing.
   # This message is too long to be a string in the A/UX 3.1 sh.
   cat <<_ACEOF
-\`configure' configures pdfio 1.2.0 to adapt to many kinds of systems.
+\`configure' configures pdfio 1.4.0 to adapt to many kinds of systems.
 
 Usage: $0 [OPTION]... [VAR=VALUE]...
 
@@ -1359,7 +1359,7 @@ fi
 
 if test -n "$ac_init_help"; then
   case $ac_init_help in
-     short | recursive ) echo "Configuration of pdfio 1.2.0:";;
+     short | recursive ) echo "Configuration of pdfio 1.4.0:";;
    esac
   cat <<\_ACEOF
 
@@ -1456,7 +1456,7 @@ fi
 test -n "$ac_init_help" && exit $ac_status
 if $ac_init_version; then
   cat <<\_ACEOF
-pdfio configure 1.2.0
+pdfio configure 1.4.0
 generated by GNU Autoconf 2.71
 
 Copyright (C) 2021 Free Software Foundation, Inc.
@@ -1612,7 +1612,7 @@ cat >config.log <<_ACEOF
 This file contains any messages produced by compilers while
 running configure, to aid debugging if configure makes a mistake.
 
-It was created by pdfio $as_me 1.2.0, which was
+It was created by pdfio $as_me 1.4.0, which was
 generated by GNU Autoconf 2.71.  Invocation command line was
 
   $ $0$ac_configure_args_raw
@@ -2368,9 +2368,9 @@ ac_compiler_gnu=$ac_cv_c_compiler_gnu
 
 
 
-PDFIO_VERSION="1.2.0"
-PDFIO_VERSION_MAJOR="`echo 1.2.0 | awk -F. '{print $1}'`"
-PDFIO_VERSION_MINOR="`echo 1.2.0 | awk -F. '{printf("%d\n",$2);}'`"
+PDFIO_VERSION="1.4.0"
+PDFIO_VERSION_MAJOR="`echo 1.4.0 | awk -F. '{print $1}'`"
+PDFIO_VERSION_MINOR="`echo 1.4.0 | awk -F. '{printf("%d\n",$2);}'`"
 
 
 
@@ -4029,8 +4029,8 @@ then :
 
     { printf "%s\n" "$as_me:${as_lineno-$LINENO}: result: yes" >&5
 printf "%s\n" "yes" >&6; }
-    LIBS="$($PKGCONFIG --libs zlib) $LIBS"
     CPPFLAGS="$($PKGCONFIG --cflags zlib) $CPPFLAGS"
+    LIBS="$($PKGCONFIG --libs zlib) $LIBS"
 
 else $as_nop
 
@@ -4093,6 +4093,7 @@ then :
 
 fi
 
+    PKGCONFIG_REQUIRES=""
     PKGCONFIG_LIBS_PRIVATE="-lz $PKGCONFIG_LIBS_PRIVATE"
 
 fi
@@ -4934,7 +4935,7 @@ cat >>$CONFIG_STATUS <<\_ACEOF || ac_write_fail=1
 # report actual input values of CONFIG_FILES etc. instead of their
 # values after options handling.
 ac_log="
-This file was extended by pdfio $as_me 1.2.0, which was
+This file was extended by pdfio $as_me 1.4.0, which was
 generated by GNU Autoconf 2.71.  Invocation command line was
 
   CONFIG_FILES    = $CONFIG_FILES
@@ -4990,7 +4991,7 @@ ac_cs_config_escaped=`printf "%s\n" "$ac_cs_config" | sed "s/^ //; s/'/'\\\\\\\\
 cat >>$CONFIG_STATUS <<_ACEOF || ac_write_fail=1
 ac_cs_config='$ac_cs_config_escaped'
 ac_cs_version="\\
-pdfio config.status 1.2.0
+pdfio config.status 1.4.0
 configured by $0, generated by GNU Autoconf 2.71,
   with options \\"\$ac_cs_config\\"
 

configure.ac

@@ -21,7 +21,7 @@ AC_PREREQ([2.70])
 
 
 dnl Package name and version...
-AC_INIT([pdfio], [1.2.0], [https://github.com/michaelrsweet/pdfio/issues], [pdfio], [https://www.msweet.org/pdfio])
+AC_INIT([pdfio], [1.4.0], [https://github.com/michaelrsweet/pdfio/issues], [pdfio], [https://www.msweet.org/pdfio])
 
 PDFIO_VERSION="AC_PACKAGE_VERSION"
 PDFIO_VERSION_MAJOR="`echo AC_PACKAGE_VERSION | awk -F. '{print $1}'`"

doc/pdfio.3

@@ -1,4 +1,4 @@
-.TH pdfio 3 "pdf read/write library" "2024-01-24" "pdf read/write library"
+.TH pdfio 3 "pdf read/write library" "2024-10-25" "pdf read/write library"
 .SH NAME
 pdfio \- pdf read/write library
 .SH Introduction
@@ -138,6 +138,121 @@ PDFio also provides PDF content helper functions for producing PDF content that
 
     #include <pdfio\-content.h>
 .fi
+.SS Understanding PDF Files
+.PP
+A PDF file provides data and commands for displaying pages of graphics and text, and is structured in a way that allows it to be displayed in the same way across multiple devices and platforms. The following is a PDF which shows "Hello, World!" on one page:
+.nf
+
+    %PDF\-1.0                        % Header starts here
+    %âãÏÓ
+    1 0 obj                         % Body starts here
+    <<
+    /Kids [2 0 R]
+    /Count 1
+    /Type /Pages
+.fi
+.PP
+> endobj 2 0 obj <
+/Rotate 0
+/Parent 1 0 R
+/Resources 3 0 R
+/MediaBox [0 0 612 792]
+/Contents [4 0 R]/Type /Page
+ endobj 3 0 obj <
+/Font
+<<
+/F0
+<<
+/BaseFont /Times\-Italic
+/Subtype /Type1
+/Type /Font
+ > > endobj 4 0 obj <
+/Length 65
+ stream
+
+.IP \(bu 5
+.PP
+0. 0. 1. 50. 700. cm BT /F0 36. Tf (Hello, World!) Tj ET endstream endobj 5 0 obj << /Pages 1 0 R /Type /Catalog
+
+
+.PP
+> endobj xref % Cross\-reference table starts here 0 6 0000000000 65535 f 0000000015 00000 n 0000000074 00000 n 0000000192 00000 n 0000000291 00000 n 0000000409 00000 n trailer % Trailer starts here << /Root 5 0 R /Size 6 > startxref 459 %%EOF
+.nf
+
+.fi
+
+.PP
+Header
+.PP
+The header is the first line of a PDF file that specifies the version of the PDF format that has been used, for example %PDF\-1.0\.
+.PP
+Since PDF files almost always contain binary data, they can become corrupted if line endings are changed. For example, if the file is transferred using FTP in text mode or is edited in Notepad on Windows. To allow legacy file transfer programs to determine that the file is binary, the PDF standard recommends including some bytes with character codes higher than 127 in the header, for example:
+.nf
+
+    %âãÏÓ
+.fi
+.PP
+The percent sign indicates a comment line while the other few bytes are arbitrary character codes in excess of 127. So, the whole header in our example is:
+.nf
+
+    %PDF\-1.0
+    %âãÏÓ
+.fi
+.PP
+Body
+.PP
+The file body consists of a sequence of objects, each preceded by an object number, generation number, and the obj keyword on one line, and followed by the endobj keyword on another. For example:
+.nf
+
+    1 0 obj
+    <<
+    /Kids [2 0 R]
+    /Count 1
+    /Type /Pages
+.fi
+.PP
+> endobj
+.nf
+
+.fi
+
+.PP
+In this example, the object number is 1 and the generation number is 0, meaning it is the first version of the object. The content for object 1 is between the initial 1 0 obj and trailing endobj lines. In this case, the content is the dictionary <</Kids [2 0 R] /Count 1 /Type /Pages>>\.
+.PP
+Cross\-Reference Table
+.PP
+The cross\-reference table lists the byte offset of each object in the file body. This allows random access to objects, meaning they don't have to be read in order. Objects that are not used are never read, making the process efficient. Operations like counting the number of pages in a PDF document are fast, even in large files.
+.PP
+Each object has an object number and a generation number. Generation numbers are used when a cross\-reference table entry is reused. For simplicity, we will assume generation numbers to be always zero and ignore them. The cross\-reference table consists of a header line that indicates the number of entries, a free entry line for object 0, and a line for each of the objects in the file body. For example:
+.nf
+
+    0 6                             % Six entries in table, starting at 0
+    0000000000 65535 f              % Free entry for object 0
+    0000000015 00000 n              % Object 1 is at byte offset 15
+    0000000074 00000 n              % Object 2 is at byte offset 74
+    0000000192 00000 n              % etc...
+    0000000291 00000 n
+    0000000409 00000 n              % Object 5 is at byte offset 409
+.fi
+.PP
+Trailer
+.PP
+The first line of the trailer is just the trailer keyword. This is followed by the trailer dictionary which contains at least the /Size entry specifying the number of entries in the cross\-reference table and the /Root entry which references the object for the document catalog which is the root element of the graph of objects in the body.
+.PP
+There follows a line with just the startxref keyword, a line with a single number specifying the byte offset of the start of the cross\-reference table within the file, and then the line %%EOF which signals the end of the PDF file.
+.nf
+
+    trailer                         % Trailer keyword
+    <<                              % The trailer dictinonary
+    /Root 5 0 R
+    /Size 6
+.fi
+.PP
+> startxref % startxref keyword 459 % Byte offset of cross\-reference table %%EOF % End\-of\-file marker
+.nf
+
+.fi
+
 .SH API Overview
 .PP
 PDFio exposes several types:
@@ -218,7 +333,90 @@ Each PDF file contains one or more pages. The pdfioFileGetNumPages function retu
     }
 .fi
 .PP
-Each page is represented by a "page tree" object (what pdfioFileGetPage returns) that specifies information about the page and one or more "content" objects that contain the images, fonts, text, and graphics that appear on the page. Use the pdfioPageGetNumStreams and pdfioPageOpenStream functions to access the content streams for each page.
+Each page is represented by a "page tree" object (what pdfioFileGetPage returns) that specifies information about the page and one or more "content" objects that contain the images, fonts, text, and graphics that appear on the page. Use the pdfioPageGetNumStreams and pdfioPageOpenStream functions to access the content streams for each page, and pdfioObjGetDict to get the associated page object dictionary. For example, if you want to display the media and crop boxes for a given page:
+.nf
+
+    pdfio_file_t  *pdf;             // PDF file
+    size_t        i;                // Looping var
+    size_t        count;            // Number of pages
+    pdfio_obj_t   *page;            // Current page
+    pdfio_dict_t  *dict;            // Current page dictionary
+    pdfio_array_t *media_box;       // MediaBox array
+    double        media_values[4];  // MediaBox values
+    pdfio_array_t *crop_box;        // CropBox array
+    double        crop_values[4];   // CropBox values
+    
+    // Iterate the pages in the PDF file
+    for (i = 0, count = pdfioFileGetNumPages(pdf); i < count; i ++)
+    {
+      page = pdfioFileGetPage(pdf, i);
+      dict = pdfioObjGetDict(page);
+    
+      media_box       = pdfioDictGetArray(dict, "MediaBox");
+      media_values[0] = pdfioArrayGetNumber(media_box, 0);
+      media_values[1] = pdfioArrayGetNumber(media_box, 1);
+      media_values[2] = pdfioArrayGetNumber(media_box, 2);
+      media_values[3] = pdfioArrayGetNumber(media_box, 3);
+    
+      crop_box       = pdfioDictGetArray(dict, "CropBox");
+      crop_values[0] = pdfioArrayGetNumber(crop_box, 0);
+      crop_values[1] = pdfioArrayGetNumber(crop_box, 1);
+      crop_values[2] = pdfioArrayGetNumber(crop_box, 2);
+      crop_values[3] = pdfioArrayGetNumber(crop_box, 3);
+    
+      printf("Page %u: MediaBox=[%g %g %g %g], CropBox=[%g %g %g %g]\\n",
+             (unsigned)(i + 1),
+             media_values[0], media_values[1], media_values[2], media_values[3],
+             crop_values[0], crop_values[1], crop_values[2], crop_values[3]);
+    }
+.fi
+.PP
+Page object dictionaries have several (mostly optional) key/value pairs, including:
+.IP \(bu 5
+.PP
+"Annots": An array of annotation dictionaries for the page; use pdfioDictGetArray to get the array
+
+.IP \(bu 5
+.PP
+"CropBox": The crop box as an array of four numbers for the left, bottom, right, and top coordinates of the target media; use pdfioDictGetArray to get a pointer to the array of numbers
+
+.IP \(bu 5
+.PP
+"Dur": The number of seconds the page should be displayed; use pdfioDictGetNumber to get the page duration value
+
+.IP \(bu 5
+.PP
+"Group": The dictionary of transparency group values for the page; use pdfioDictGetDict to get a pointer to the resources dictionary
+
+.IP \(bu 5
+.PP
+"LastModified": The date and time when this page was last modified; use pdfioDictGetDate to get the Unix time_t value
+
+.IP \(bu 5
+.PP
+"Parent": The parent page tree node object for this page; use pdfioDictGetObj to get a pointer to the object
+
+.IP \(bu 5
+.PP
+"MediaBox": The media box as an array of four numbers for the left, bottom, right, and top coordinates of the target media; use pdfioDictGetArray to get a pointer to the array of numbers
+
+.IP \(bu 5
+.PP
+"Resources": The dictionary of resources for the page; use pdfioDictGetDict to get a pointer to the resources dictionary
+
+.IP \(bu 5
+.PP
+"Rotate": A number indicating the number of degrees of counter\-clockwise rotation to apply to the page when viewing; use pdfioDictGetNumber to get the rotation angle
+
+.IP \(bu 5
+.PP
+"Thumb": A thumbnail image object for the page; use pdfioDictGetObj to get a pointer to the thumbnail image object
+
+.IP \(bu 5
+.PP
+"Trans": The page transition dictionary; use pdfioDictGetDict to get a pointer to the dictionary
+
+
 .PP
 The pdfioFileClose function closes a PDF file and frees all memory that was used for it:
 .nf
@@ -361,7 +559,7 @@ pdfioStreamWrite writes a buffer of data to the stream
 .PP
 The PDF content helper functions provide additional functions for writing specific PDF page stream commands.
 .PP
-When you are done writing the stream, call pdfioStreamCLose to close both the stream and the object.
+When you are done writing the stream, call pdfioStreamClose to close both the stream and the object.
 .SS PDF Content Helper Functions
 .PP
 PDFio includes many helper functions for embedding or writing specific kinds of content to a PDF file. These functions can be roughly grouped into five categories:
@@ -787,6 +985,125 @@ pdfioContentTextShowf draws a formatted string in a text block
 pdfioContentTextShowJustified draws an array of literal strings with offsets between them
 
 
+.SH Examples
+.SS Read PDF Metadata
+.PP
+The following example function will open a PDF file and print the title, author, creation date, and number of pages:
+.nf
+
+    #include <pdfio.h>
+    #include <time.h>
+    
+    
+    void
+    show_pdf_info(const char *filename)
+    {
+      pdfio_file_t *pdf;
+      time_t       creation_date;
+      struct tm    *creation_tm;
+      char         creation_text[256];
+    
+    
+      // Open the PDF file with the default callbacks...
+      pdf = pdfioFileOpen(filename, /*password_cb*/NULL, /*password_cbdata*/NULL, /*error_cb*/NULL, /*error_cbdata*/NULL);
+      if (pdf == NULL)
+        return;
+    
+      // Get the creation date and convert to a string...
+      creation_date = pdfioFileGetCreationDate(pdf);
+      creation_tm   = localtime(&creation_date);
+      strftime(creation_text, sizeof(creation_text), "%c", &creation_tm);
+    
+      // Print file information to stdout...
+      printf("%s:\\n", filename);
+      printf("         Title: %s\\n", pdfioFileGetTitle(pdf));
+      printf("        Author: %s\\n", pdfioFileGetAuthor(pdf));
+      printf("    Created On: %s\\n", creation_text);
+      printf("  Number Pages: %u\\n", (unsigned)pdfioFileGetNumPages(pdf));
+    
+      // Close the PDF file...
+      pdfioFileClose(pdf);
+    }
+.fi
+.SS Create PDF File With Text and Image
+.PP
+The following example function will create a PDF file, embed a base font and the named JPEG or PNG image file, and then creates a page with the image centered on the page with the text centered below:
+.nf
+
+    #include <pdfio.h>
+    #include <pdfio\-content.h>
+    #include <string.h>
+    
+    
+    void
+    create_pdf_image_file(const char *pdfname, const char *imagename, const char *caption)
+    {
+      pdfio_file_t   *pdf;
+      pdfio_obj_t    *font;
+      pdfio_obj_t    *image;
+      pdfio_dict_t   *dict;
+      pdfio_stream_t *page;
+      double         width, height;
+      double         swidth, sheight;
+      double         tx, ty;
+    
+    
+      // Create the PDF file...
+      pdf = pdfioFileCreate(pdfname, /*version*/NULL, /*media_box*/NULL, /*crop_box*/NULL, /*error_cb*/NULL, /*error_cbdata*/NULL);
+    
+      // Create a Courier base font for the caption
+      font = pdfioFileCreateFontObjFromBase(pdf, "Courier");
+    
+      // Create an image object from the JPEG/PNG image file...
+      image = pdfioFileCreateImageObjFromFile(pdf, imagename, true);
+    
+      // Create a page dictionary with the font and image...
+      dict = pdfioDictCreate(pdf);
+      pdfioPageDictAddFont(dict, "F1", font);
+      pdfioPageDictAddImage(dict, "IM1", image);
+    
+      // Create the page and its content stream...
+      page = pdfioFileCreatePage(pdf, dict);
+    
+      // Position and scale the image on the page...
+      width  = pdfioImageGetWidth(image);
+      height = pdfioImageGetHeight(image);
+    
+      // Default media_box is "universal" 595.28x792 points (8.27x11in or 210x279mm)
+      // Use margins of 36 points (0.5in or 12.7mm) with another 36 points for the
+      // caption underneath...
+      swidth  = 595.28 \- 72.0;
+      sheight = swidth * height / width;
+      if (sheight > (792.0 \- 36.0 \- 72.0))
+      {
+        sheight = 792.0 \- 36.0 \- 72.0;
+        swidth  = sheight * width / height;
+      }
+    
+      tx = 0.5 * (595.28 \- swidth);
+      ty = 0.5 * (792 \- 36 \- sheight);
+    
+      pdfioContentDrawImage(page, "IM1", tx, ty + 36.0, swidth, sheight);
+    
+      // Draw the caption in black...
+      pdfioContentSetFillColorDeviceGray(page, 0.0);
+    
+      // Compute the starting point for the text \- Courier is monospaced with a
+      // nominal width of 0.6 times the text height...
+      tx = 0.5 * (595.28 \- 18.0 * 0.6 * strlen(caption));
+    
+      // Position and draw the caption underneath...
+      pdfioContentTextBegin(page);
+      pdfioContentSetTextFont(page, "F1", 18.0);
+      pdfioContentTextMoveTo(page, tx, ty);
+      pdfioContentTextShow(page, /*unicode*/false, caption);
+      pdfioContentTextEnd(page);
+    
+      // Close the page stream and the PDF file...
+      pdfioStreamClose(page);
+      pdfioFileClose(pdf);
+    }
+.fi
 
 .SH ENUMERATIONS
 .SS pdfio_cs_e
@@ -1285,6 +1602,15 @@ pdfio_valtype_t  pdfioArrayGetType (
     size_t n
 );
 .fi
+.SS pdfioArrayRemove
+Remove an array entry.
+.PP
+.nf
+bool  pdfioArrayRemove (
+    pdfio_array_t *a,
+    size_t n
+);
+.fi
 .SS pdfioContentClip
 Clip output to the current path.
 .PP
@@ -1865,6 +2191,15 @@ bool  pdfioContentTextShowf (
     ...
 );
 .fi
+.SS pdfioDictClear
+Remove a key/value pair from a dictionary.
+.PP
+.nf
+bool  pdfioDictClear (
+    pdfio_dict_t *dict,
+    const char *key
+);
+.fi
 .SS pdfioDictCopy
 Copy a dictionary to a PDF file.
 .PP
@@ -1928,6 +2263,15 @@ pdfio_dict_t * pdfioDictGetDict (
     const char *key
 );
 .fi
+.SS pdfioDictGetKey
+Get the key for the specified pair.
+.PP
+.nf
+const char * pdfioDictGetKey (
+    pdfio_dict_t *dict,
+    size_t n
+);
+.fi
 .SS pdfioDictGetName
 Get a key name value from a dictionary.
 .PP
@@ -1937,6 +2281,14 @@ const char * pdfioDictGetName (
     const char *key
 );
 .fi
+.SS pdfioDictGetNumPairs
+Get the number of key/value pairs in a dictionary.
+.PP
+.nf
+size_t  pdfioDictGetNumPairs (
+    pdfio_dict_t *dict
+);
+.fi
 .SS pdfioDictGetNumber
 Get a key number value from a dictionary.
 .PP
@@ -2148,7 +2500,7 @@ pdfio_file_t * pdfioFileCreate (
     pdfio_rect_t *media_box,
     pdfio_rect_t *crop_box,
     pdfio_error_cb_t error_cb,
-    void *error_data
+    void *error_cbdata
 );
 .fi
 .PP
@@ -2162,7 +2514,7 @@ The "media_box" and "crop_box" arguments specify the default MediaBox and
 CropBox for pages in the PDF file - if \fBNULL\fR then a default "Universal" size
 of 8.27x11in (the intersection of US Letter and ISO A4) is used.
 .PP
-The "error_cb" and "error_data" arguments specify an error handler callback
+The "error_cb" and "error_cbdata" arguments specify an error handler callback
 and its data pointer - if \fBNULL\fR the default error handler is used that
 writes error messages to \fBstderr\fR.
 .SS pdfioFileCreateArrayObj
@@ -2298,6 +2650,18 @@ Note: Currently PNG support is limited to grayscale, RGB, or indexed files
 without interlacing or alpha.  Transparency (masking) based on color/index
 .IP 5
 is supported.
+.SS pdfioFileCreateNameObj
+Create a new object in a PDF file containing a name.
+.PP
+.nf
+pdfio_obj_t * pdfioFileCreateNameObj (
+    pdfio_file_t *pdf,
+    const char *name
+);
+.fi
+.PP
+This function creates a new object with a name value in a PDF file.
+You must call \fIpdfioObjClose\fR to write the object to the file.
 .SS pdfioFileCreateNumberObj
 Create a new object in a PDF file containing a number.
 .PP
@@ -2325,23 +2689,23 @@ Create a PDF file through an output callback.
 .nf
 pdfio_file_t * pdfioFileCreateOutput (
     pdfio_output_cb_t output_cb,
-    void *output_ctx,
+    void *output_cbdata,
     const char *version,
     pdfio_rect_t *media_box,
     pdfio_rect_t *crop_box,
     pdfio_error_cb_t error_cb,
-    void *error_data
+    void *error_cbdata
 );
 .fi
 .PP
 This function creates a new PDF file that is streamed though an output
-callback.  The "output_cb" and "output_ctx" arguments specify the output
-callback and its context pointer which is called whenever data needs to be
+callback.  The "output_cb" and "output_cbdata" arguments specify the output
+callback and its data pointer which is called whenever data needs to be
 written:
 .PP
 .nf
     ssize_t
-    output_cb(void *output_ctx, const void *buffer, size_t bytes)
+    output_cb(void *output_cbdata, const void *buffer, size_t bytes)
     {
       // Write buffer to output and return the number of bytes written
     }
@@ -2355,7 +2719,7 @@ The "media_box" and "crop_box" arguments specify the default MediaBox and
 CropBox for pages in the PDF file - if \fBNULL\fR then a default "Universal" size
 of 8.27x11in (the intersection of US Letter and ISO A4) is used.
 .PP
-The "error_cb" and "error_data" arguments specify an error handler callback
+The "error_cb" and "error_cbdata" arguments specify an error handler callback
 and its data pointer - if \fBNULL\fR the default error handler is used that
 writes error messages to \fBstderr\fR.
 .PP
@@ -2397,7 +2761,7 @@ pdfio_file_t * pdfioFileCreateTemporary (
     pdfio_rect_t *media_box,
     pdfio_rect_t *crop_box,
     pdfio_error_cb_t error_cb,
-    void *error_data
+    void *error_cbdata
 );
 .fi
 .SS pdfioFileFindObj
@@ -2420,6 +2784,16 @@ const char * pdfioFileGetAuthor (
     pdfio_file_t *pdf
 );
 .fi
+.SS pdfioFileGetCatalog
+Get the document catalog dictionary.
+.PP
+.nf
+pdfio_dict_t * pdfioFileGetCatalog (
+    pdfio_file_t *pdf
+);
+.fi
+.PP
+
 .SS pdfioFileGetCreationDate
 Get the creation date for a PDF file.
 .PP
@@ -2545,22 +2919,22 @@ Open a PDF file for reading.
 pdfio_file_t * pdfioFileOpen (
     const char *filename,
     pdfio_password_cb_t password_cb,
-    void *password_data,
+    void *password_cbdata,
     pdfio_error_cb_t error_cb,
-    void *error_data
+    void *error_cbdata
 );
 .fi
 .PP
 This function opens an existing PDF file.  The "filename" argument specifies
 the name of the PDF file to create.
 .PP
-The "password_cb" and "password_data" arguments specify a password callback
+The "password_cb" and "password_cbdata" arguments specify a password callback
 and its data pointer for PDF files that use one of the standard Adobe
 "security" handlers.  The callback returns a password string or \fBNULL\fR to
 cancel the open.  If \fBNULL\fR is specified for the callback function and the
 PDF file requires a password, the open will always fail.
 .PP
-The "error_cb" and "error_data" arguments specify an error handler callback
+The "error_cb" and "error_cbdata" arguments specify an error handler callback
 and its data pointer - if \fBNULL\fR the default error handler is used that
 writes error messages to \fBstderr\fR.
 .SS pdfioFileSetAuthor
@@ -2724,6 +3098,14 @@ size_t  pdfioObjGetLength (
     pdfio_obj_t *obj
 );
 .fi
+.SS pdfioObjGetName
+Get the name value associated with an object.
+.PP
+.nf
+const char * pdfioObjGetName (
+    pdfio_obj_t *obj
+);
+.fi
 .SS pdfioObjGetNumber
 Get the object's number.
 .PP
@@ -2740,6 +3122,29 @@ const char * pdfioObjGetSubtype (
     pdfio_obj_t *obj
 );
 .fi
+.PP
+This function returns an object's PDF subtype name, if any.  Common subtype
+names include:
+.PP
+.IP \(bu 5
+"CIDFontType0": A CID Type0 font
+.IP \(bu 5
+"CIDFontType2": A CID TrueType font
+.IP \(bu 5
+"Image": An image or image mask
+.IP \(bu 5
+"Form": A fillable form
+.IP \(bu 5
+"OpenType": An OpenType font
+.IP \(bu 5
+"Type0": A composite font
+.IP \(bu 5
+"Type1": A PostScript Type1 font
+.IP \(bu 5
+"Type3": A PDF Type3 font
+.IP \(bu 5
+"TrueType": A TrueType font</li>
+</ul>
 .SS pdfioObjGetType
 Get an object's type.
 .PP
@@ -2748,6 +3153,27 @@ const char * pdfioObjGetType (
     pdfio_obj_t *obj
 );
 .fi
+.PP
+This function returns an object's PDF type name, if any. Common type names
+include:
+.PP
+.IP \(bu 5
+"CMap": A character map for composite fonts
+.IP \(bu 5
+"Font": An embedded font (\fIpdfioObjGetSubtype\fR will tell you the
+  font format)
+.IP \(bu 5
+"FontDescriptor": A font descriptor
+.IP \(bu 5
+"Page": A (visible) page
+.IP \(bu 5
+"Pages": A page tree node
+.IP \(bu 5
+"Template": An invisible template page
+.IP \(bu 5
+"XObject": An image, image mask, or form (\fIpdfioObjGetSubtype\fR will
+  tell you which)</li>
+</ul>
 .SS pdfioObjOpenStream
 Open an object's (data) stream for reading.
 .PP

doc/pdfio.md

@@ -120,6 +120,182 @@ that are defined in a separate header file:
 ```
 
 
+Understanding PDF Files
+-----------------------
+
+A PDF file provides data and commands for displaying pages of graphics and text,
+and is structured in a way that allows it to be displayed in the same way across
+multiple devices and platforms.  The following is a PDF which shows "Hello,
+World!" on one page:
+
+```
+%PDF-1.0                        % Header starts here
+%âãÏÓ
+1 0 obj                         % Body starts here
+<<
+/Kids [2 0 R]
+/Count 1
+/Type /Pages
+>>
+endobj
+2 0 obj
+<<
+/Rotate 0
+/Parent 1 0 R
+/Resources 3 0 R
+/MediaBox [0 0 612 792]
+/Contents [4 0 R]/Type /Page
+>>
+endobj
+3 0 obj
+<<
+/Font
+<<
+/F0
+<<
+/BaseFont /Times-Italic
+/Subtype /Type1
+/Type /Font
+>>
+>>
+>>
+endobj
+4 0 obj
+<<
+/Length 65
+>>
+stream
+1. 0. 0. 1. 50. 700. cm
+BT
+  /F0 36. Tf
+  (Hello, World!) Tj
+ET
+endstream
+endobj
+5 0 obj
+<<
+/Pages 1 0 R
+/Type /Catalog
+>>
+endobj
+xref                            % Cross-reference table starts here
+0 6
+0000000000 65535 f
+0000000015 00000 n
+0000000074 00000 n
+0000000192 00000 n
+0000000291 00000 n
+0000000409 00000 n
+trailer                         % Trailer starts here
+<<
+/Root 5 0 R
+/Size 6
+>>
+startxref
+459
+%%EOF
+```
+
+
+### Header
+
+The header is the first line of a PDF file that specifies the version of the PDF
+format that has been used, for example `%PDF-1.0`.
+
+Since PDF files almost always contain binary data, they can become corrupted if
+line endings are changed.  For example, if the file is transferred using FTP in
+text mode or is edited in Notepad on Windows.  To allow legacy file transfer
+programs to determine that the file is binary, the PDF standard recommends
+including some bytes with character codes higher than 127 in the header, for
+example:
+
+```
+%âãÏÓ
+```
+
+The percent sign indicates a comment line while the other few bytes are
+arbitrary character codes in excess of 127.  So, the whole header in our example
+is:
+
+```
+%PDF-1.0
+%âãÏÓ
+```
+
+
+### Body
+
+The file body consists of a sequence of objects, each preceded by an object
+number, generation number, and the obj keyword on one line, and followed by the
+endobj keyword on another.  For example:
+
+```
+1 0 obj
+<<
+/Kids [2 0 R]
+/Count 1
+/Type /Pages
+>>
+endobj
+```
+
+In this example, the object number is 1 and the generation number is 0, meaning
+it is the first version of the object.  The content for object 1 is between the
+initial `1 0 obj` and trailing `endobj` lines.  In this case, the content is the
+dictionary `<</Kids [2 0 R] /Count 1 /Type /Pages>>`.
+
+
+### Cross-Reference Table
+
+The cross-reference table lists the byte offset of each object in the file body.
+This allows random access to objects, meaning they don't have to be read in
+order.  Objects that are not used are never read, making the process efficient.
+Operations like counting the number of pages in a PDF document are fast, even in
+large files.
+
+Each object has an object number and a generation number.  Generation numbers
+are used when a cross-reference table entry is reused.  For simplicity, we will
+assume generation numbers to be always zero and ignore them.  The
+cross-reference table consists of a header line that indicates the number of
+entries, a free entry line for object 0, and a line for each of the objects in
+the file body.  For example:
+
+```
+0 6                             % Six entries in table, starting at 0
+0000000000 65535 f              % Free entry for object 0
+0000000015 00000 n              % Object 1 is at byte offset 15
+0000000074 00000 n              % Object 2 is at byte offset 74
+0000000192 00000 n              % etc...
+0000000291 00000 n
+0000000409 00000 n              % Object 5 is at byte offset 409
+```
+
+
+### Trailer
+
+The first line of the trailer is just the `trailer` keyword.  This is followed
+by the trailer dictionary which contains at least the `/Size` entry specifying
+the number of entries in the cross-reference table and the `/Root` entry which
+references the object for the document catalog which is the root element of the
+graph of objects in the body.
+
+There follows a line with just the `startxref` keyword, a line with a single
+number specifying the byte offset of the start of the cross-reference table
+within the file, and then the line `%%EOF` which signals the end of the PDF
+file.
+
+```
+trailer                         % Trailer keyword
+<<                              % The trailer dictinonary
+/Root 5 0 R
+/Size 6
+>>
+startxref                       % startxref keyword
+459                             % Byte offset of cross-reference table
+%%EOF                           % End-of-file marker
+```
+
+
 API Overview
 ============
 
@@ -132,6 +308,7 @@ PDFio exposes several types:
 - `pdfio_stream_t`: An object stream
 
 
+
 Reading PDF Files
 -----------------
 
@@ -202,7 +379,74 @@ Each page is represented by a "page tree" object (what [`pdfioFileGetPage`](@@)
 returns) that specifies information about the page and one or more "content"
 objects that contain the images, fonts, text, and graphics that appear on the
 page.  Use the [`pdfioPageGetNumStreams`](@@) and [`pdfioPageOpenStream`](@@)
-functions to access the content streams for each page.
+functions to access the content streams for each page, and
+[`pdfioObjGetDict`](@@) to get the associated page object dictionary.  For
+example, if you want to display the media and crop boxes for a given page:
+
+```c
+pdfio_file_t  *pdf;             // PDF file
+size_t        i;                // Looping var
+size_t        count;            // Number of pages
+pdfio_obj_t   *page;            // Current page
+pdfio_dict_t  *dict;            // Current page dictionary
+pdfio_array_t *media_box;       // MediaBox array
+double        media_values[4];  // MediaBox values
+pdfio_array_t *crop_box;        // CropBox array
+double        crop_values[4];   // CropBox values
+
+// Iterate the pages in the PDF file
+for (i = 0, count = pdfioFileGetNumPages(pdf); i < count; i ++)
+{
+  page = pdfioFileGetPage(pdf, i);
+  dict = pdfioObjGetDict(page);
+
+  media_box       = pdfioDictGetArray(dict, "MediaBox");
+  media_values[0] = pdfioArrayGetNumber(media_box, 0);
+  media_values[1] = pdfioArrayGetNumber(media_box, 1);
+  media_values[2] = pdfioArrayGetNumber(media_box, 2);
+  media_values[3] = pdfioArrayGetNumber(media_box, 3);
+
+  crop_box       = pdfioDictGetArray(dict, "CropBox");
+  crop_values[0] = pdfioArrayGetNumber(crop_box, 0);
+  crop_values[1] = pdfioArrayGetNumber(crop_box, 1);
+  crop_values[2] = pdfioArrayGetNumber(crop_box, 2);
+  crop_values[3] = pdfioArrayGetNumber(crop_box, 3);
+
+  printf("Page %u: MediaBox=[%g %g %g %g], CropBox=[%g %g %g %g]\n",
+         (unsigned)(i + 1),
+         media_values[0], media_values[1], media_values[2], media_values[3],
+         crop_values[0], crop_values[1], crop_values[2], crop_values[3]);
+}
+```
+
+Page object dictionaries have several (mostly optional) key/value pairs,
+including:
+
+- "Annots": An array of annotation dictionaries for the page; use
+  [`pdfioDictGetArray`](@@) to get the array
+- "CropBox": The crop box as an array of four numbers for the left, bottom,
+  right, and top coordinates of the target media; use [`pdfioDictGetArray`](@@)
+  to get a pointer to the array of numbers
+- "Dur": The number of seconds the page should be displayed; use
+  [`pdfioDictGetNumber`](@@) to get the page duration value
+- "Group": The dictionary of transparency group values for the page; use
+  [`pdfioDictGetDict`](@@) to get a pointer to the resources dictionary
+- "LastModified": The date and time when this page was last modified; use
+  [`pdfioDictGetDate`](@@) to get the Unix `time_t` value
+- "Parent": The parent page tree node object for this page; use
+  [`pdfioDictGetObj`](@@) to get a pointer to the object
+- "MediaBox": The media box as an array of four numbers for the left, bottom,
+  right, and top coordinates of the target media; use [`pdfioDictGetArray`](@@)
+  to get a pointer to the array of numbers
+- "Resources": The dictionary of resources for the page; use
+  [`pdfioDictGetDict`](@@) to get a pointer to the resources dictionary
+- "Rotate": A number indicating the number of degrees of counter-clockwise
+  rotation to apply to the page when viewing; use [`pdfioDictGetNumber`](@@)
+  to get the rotation angle
+- "Thumb": A thumbnail image object for the page; use [`pdfioDictGetObj`](@@)
+  to get a pointer to the thumbnail image object
+- "Trans": The page transition dictionary; use [`pdfioDictGetDict`](@@) to get
+  a pointer to the dictionary
 
 The [`pdfioFileClose`](@@) function closes a PDF file and frees all memory that
 was used for it:
@@ -345,7 +589,7 @@ to the stream:
 The [PDF content helper functions](@) provide additional functions for writing
 specific PDF page stream commands.
 
-When you are done writing the stream, call [`pdfioStreamCLose`](@@) to close
+When you are done writing the stream, call [`pdfioStreamClose`](@@) to close
 both the stream and the object.
 
 
@@ -586,3 +830,133 @@ escaping, as needed:
 - [`pdfioContentTextShowf`](@@) draws a formatted string in a text block
 - [`pdfioContentTextShowJustified`](@@) draws an array of literal strings with
   offsets between them
+
+
+Examples
+========
+
+Read PDF Metadata
+-----------------
+
+The following example function will open a PDF file and print the title, author,
+creation date, and number of pages:
+
+```c
+#include <pdfio.h>
+#include <time.h>
+
+
+void
+show_pdf_info(const char *filename)
+{
+  pdfio_file_t *pdf;
+  time_t       creation_date;
+  struct tm    *creation_tm;
+  char         creation_text[256];
+
+
+  // Open the PDF file with the default callbacks...
+  pdf = pdfioFileOpen(filename, /*password_cb*/NULL, /*password_cbdata*/NULL, /*error_cb*/NULL, /*error_cbdata*/NULL);
+  if (pdf == NULL)
+    return;
+
+  // Get the creation date and convert to a string...
+  creation_date = pdfioFileGetCreationDate(pdf);
+  creation_tm   = localtime(&creation_date);
+  strftime(creation_text, sizeof(creation_text), "%c", &creation_tm);
+
+  // Print file information to stdout...
+  printf("%s:\n", filename);
+  printf("         Title: %s\n", pdfioFileGetTitle(pdf));
+  printf("        Author: %s\n", pdfioFileGetAuthor(pdf));
+  printf("    Created On: %s\n", creation_text);
+  printf("  Number Pages: %u\n", (unsigned)pdfioFileGetNumPages(pdf));
+
+  // Close the PDF file...
+  pdfioFileClose(pdf);
+}
+```
+
+
+Create PDF File With Text and Image
+-----------------------------------
+
+The following example function will create a PDF file, embed a base font and the
+named JPEG or PNG image file, and then creates a page with the image centered on
+the page with the text centered below:
+
+```c
+#include <pdfio.h>
+#include <pdfio-content.h>
+#include <string.h>
+
+
+void
+create_pdf_image_file(const char *pdfname, const char *imagename, const char *caption)
+{
+  pdfio_file_t   *pdf;
+  pdfio_obj_t    *font;
+  pdfio_obj_t    *image;
+  pdfio_dict_t   *dict;
+  pdfio_stream_t *page;
+  double         width, height;
+  double         swidth, sheight;
+  double         tx, ty;
+
+
+  // Create the PDF file...
+  pdf = pdfioFileCreate(pdfname, /*version*/NULL, /*media_box*/NULL, /*crop_box*/NULL, /*error_cb*/NULL, /*error_cbdata*/NULL);
+
+  // Create a Courier base font for the caption
+  font = pdfioFileCreateFontObjFromBase(pdf, "Courier");
+
+  // Create an image object from the JPEG/PNG image file...
+  image = pdfioFileCreateImageObjFromFile(pdf, imagename, true);
+
+  // Create a page dictionary with the font and image...
+  dict = pdfioDictCreate(pdf);
+  pdfioPageDictAddFont(dict, "F1", font);
+  pdfioPageDictAddImage(dict, "IM1", image);
+
+  // Create the page and its content stream...
+  page = pdfioFileCreatePage(pdf, dict);
+
+  // Position and scale the image on the page...
+  width  = pdfioImageGetWidth(image);
+  height = pdfioImageGetHeight(image);
+
+  // Default media_box is "universal" 595.28x792 points (8.27x11in or 210x279mm)
+  // Use margins of 36 points (0.5in or 12.7mm) with another 36 points for the
+  // caption underneath...
+  swidth  = 595.28 - 72.0;
+  sheight = swidth * height / width;
+  if (sheight > (792.0 - 36.0 - 72.0))
+  {
+    sheight = 792.0 - 36.0 - 72.0;
+    swidth  = sheight * width / height;
+  }
+
+  tx = 0.5 * (595.28 - swidth);
+  ty = 0.5 * (792 - 36 - sheight);
+
+  pdfioContentDrawImage(page, "IM1", tx, ty + 36.0, swidth, sheight);
+
+  // Draw the caption in black...
+  pdfioContentSetFillColorDeviceGray(page, 0.0);
+
+  // Compute the starting point for the text - Courier is monospaced with a
+  // nominal width of 0.6 times the text height...
+  tx = 0.5 * (595.28 - 18.0 * 0.6 * strlen(caption));
+
+  // Position and draw the caption underneath...
+  pdfioContentTextBegin(page);
+  pdfioContentSetTextFont(page, "F1", 18.0);
+  pdfioContentTextMoveTo(page, tx, ty);
+  pdfioContentTextShow(page, /*unicode*/false, caption);
+  pdfioContentTextEnd(page);
+
+  // Close the page stream and the PDF file...
+  pdfioStreamClose(page);
+  pdfioFileClose(pdf);
+}
+```

examples/Makefile

@@ -0,0 +1,46 @@
+#
+# Makefile for PDFio examples.
+#
+# Copyright © 2024 by Michael R Sweet.
+#
+# Licensed under Apache License v2.0.  See the file "LICENSE" for more
+# information.
+#
+
+# POSIX makefile
+.POSIX:
+
+
+# Common options
+CFLAGS		=	-g $(CPPFLAGS)
+CPPFLAGS	=	-I..
+LIBS		=	-L.. -lpdfio -lz
+
+
+# Targets
+TARGETS		=	\
+			code128 \
+			md2pdf
+
+
+# Make everything
+all:	$(TARGETS)
+
+
+# Clean everything
+clean:
+	rm -f $(TARGETS)
+
+
+# code128
+code128:	code128.c
+	$(CC) $(CFLAGS) -o $@ code128.c $(LIBS)
+
+
+# md2pdf
+md2pdf:		md2pdf.c mmd.c mmd.h
+	$(CC) $(CFLAGS) -o $@ md2pdf.c mmd.c $(LIBS)
+
+
+# Common dependencies...
+$(TARGETS):	Makefile ../pdfio.h ../pdfio-content.h

examples/code128.c

@@ -0,0 +1,209 @@
+//
+// Code 128 barcode example for PDFio.
+//
+// Copyright © 2024 by Michael R Sweet.
+//
+// Licensed under Apache License v2.0.  See the file "LICENSE" for more
+// information.
+//
+// Usage:
+//
+//   ./code128 "BARCODE" ["TEXT"] >FILENAME.pdf
+//
+
+#include <pdfio.h>
+#include <pdfio-content.h>
+
+
+//
+// 'make_code128()' - Make a Code 128 barcode string.
+//
+// This function produces a Code B (printable ASCII) representation of the
+// source string and doesn't try to optimize using Code C.  Non-printable and
+// extended characters are ignored in the source string.
+//
+
+
+static char *				// O - Output string
+make_code128(char       *dst,		// I - Destination buffer
+             const char *src,		// I - Source string
+             size_t     dstsize)	// I - Size of destination buffer
+{
+  char		*dstptr,		// Pointer into destination buffer
+		*dstend;		// End of destination buffer
+  int		sum;			// Weighted sum
+  static const char *code128_chars =	// Code 128 characters
+		" !\"#$%&'()*+,-./0123456789:;<=>?"
+		"@ABCDEFGHIJKLMNOPQRSTUVWXYZ[\\]^_"
+		"`abcdefghijklmnopqrstuvwxyz{|}~\303"
+		"\304\305\306\307\310\311\312";
+  static const char code128_fnc_3 = '\304';
+					// FNC 3
+  static const char code128_fnc_2 = '\305';
+					// FNC 2
+  static const char code128_shift_b = '\306';
+					// Shift B (for lowercase)
+  static const char code128_code_c = '\307';
+					// Code C
+  static const char code128_code_b = '\310';
+					// Code B
+  static const char code128_fnc_4 = '\311';
+					// FNC 4
+  static const char code128_fnc_1 = '\312';
+					// FNC 1
+  static const char code128_start_code_a = '\313';
+					// Start code A
+  static const char code128_start_code_b = '\314';
+					// Start code A
+  static const char code128_start_code_c = '\315';
+					// Start code A
+  static const char code128_stop = '\316';
+					// Stop pattern
+
+
+  // Start a Code B barcode...
+  dstptr = dst;
+  dstend = dst + dstsize - 3;
+
+  *dstptr++ = code128_start_code_b;
+  sum       = code128_start_code_b - 100;
+
+  while (*src && dstptr < dstend)
+  {
+    if (*src >= ' ' && *src < 0x7f)
+    {
+      sum       += (dstptr - dst) * (*src - ' ');
+      *dstptr++ = *src;
+    }
+
+    src ++;
+  }
+
+  // Add the weighted sum modulo 103
+  *dstptr++ = code128_chars[sum % 103];
+
+  // Add the stop pattern and return...
+  *dstptr++ = code128_stop;
+  *dstptr   = '\0';
+
+  return (dst);
+}
+
+
+//
+// 'output_cb()' - Write PDF data to the standard output...
+//
+
+static ssize_t				// O - Number of bytes written
+output_cb(void       *output_cbdata,	// I - Callback data (not used)
+          const void *buffer,		// I - Buffer to write
+          size_t     bytes)		// I - Number of bytes to write
+{
+  (void)output_cbdata;
+
+  return ((ssize_t)fwrite(buffer, 1, bytes, stdout));
+}
+
+
+//
+// 'main()' - Produce a single-page barcode file.
+//
+
+int					// O - Exit status
+main(int  argc,				// I - Number of command-line arguments
+     char *argv[])			// I - Command-line arguments
+{
+  const char	*barcode,		// Barcode to show
+		*text;			// Text to display under barcode
+  pdfio_file_t	*pdf;			// Output PDF file
+  pdfio_obj_t	*barcode_font;		// Barcode font object
+  pdfio_obj_t	*text_font = NULL;	// Text font object
+  pdfio_dict_t	*page_dict;		// Page dictionary
+  pdfio_rect_t	media_box;		// Media/CropBox for page
+  pdfio_stream_t *page_st;		// Page stream
+  char		barcode_temp[256];	// Barcode buffer
+  double	barcode_height = 36.0,	// Height of barcode
+		barcode_width,		// Width of barcode
+		text_height = 0.0,	// Height of text
+		text_width = 0.0;	// Width of text
+
+
+  // Get the barcode and optional text from the command-line...
+  if (argc < 2 || argc > 3)
+  {
+    fputs("Usage: code128 \"BARCODE\" [\"TEXT\"] >FILENAME.pdf\n", stderr);
+    return (1);
+  }
+
+  barcode = argv[1];
+  text    = argv[2];
+
+  // Output a PDF file to the standard output...
+#ifdef _WIN32
+  setmode(1, O_BINARY);			// Force binary output on Windows
+#endif // _WIN32
+
+  if ((pdf = pdfioFileCreateOutput(output_cb, /*output_cbdata*/NULL, /*version*/NULL, /*media_box*/NULL, /*crop_box*/NULL, /*error_cb*/NULL, /*error_data*/NULL)) == NULL)
+    return (1);
+
+  // Load fonts...
+  barcode_font = pdfioFileCreateFontObjFromFile(pdf, "code128.ttf", /*unicode*/false);
+  if (text)
+    text_font = pdfioFileCreateFontObjFromFile(pdf, "../testfiles/OpenSans-Regular.ttf", /*unicode*/true);
+
+  // Generate Code128 characters for the desired barcode...
+  if (!(barcode[0] & 0x80))
+    barcode = make_code128(barcode_temp, barcode, sizeof(barcode_temp));
+
+  // Compute sizes of the text...
+  barcode_width = pdfioContentTextMeasure(barcode_font, barcode, barcode_height);
+  if (text && text_font)
+  {
+    text_height = 9.0;
+    text_width  = pdfioContentTextMeasure(text_font, text, text_height);
+  }
+
+  // Compute the size of the PDF page...
+  media_box.x1 = 0.0;
+  media_box.y1 = 0.0;
+  media_box.x2 = (barcode_width > text_width ? barcode_width : text_width) + 18.0;
+  media_box.y2 = barcode_height + text_height + 18.0;
+
+  // Start a page for the barcode...
+  page_dict = pdfioDictCreate(pdf);
+
+  pdfioDictSetRect(page_dict, "MediaBox", &media_box);
+  pdfioDictSetRect(page_dict, "CropBox", &media_box);
+
+  pdfioPageDictAddFont(page_dict, "B128", barcode_font);
+  if (text_font)
+    pdfioPageDictAddFont(page_dict, "TEXT", text_font);
+
+  page_st = pdfioFileCreatePage(pdf, page_dict);
+
+  // Draw the page...
+  pdfioContentSetStrokeColorGray(page_st, 0.0);
+
+  pdfioContentSetTextFont(page_st, "B128", barcode_height);
+  pdfioContentTextBegin(page_st);
+  pdfioContentTextMoveTo(page_st, 0.5 * (media_box.x2 - barcode_width), 9.0 + text_height);
+  pdfioContentTextShow(page_st, /*unicode*/false, barcode);
+  pdfioContentTextEnd(page_st);
+
+  if (text && text_font)
+  {
+    pdfioContentSetTextFont(page_st, "TEXT", text_height);
+    pdfioContentTextBegin(page_st);
+    pdfioContentTextMoveTo(page_st, 0.5 * (media_box.x2 - text_width), 9.0);
+    pdfioContentTextShow(page_st, /*unicode*/true, text);
+    pdfioContentTextEnd(page_st);
+  }
+
+  pdfioStreamClose(page_st);
+
+  // Close and return...
+  pdfioFileClose(pdf);
+
+  return (0);
+}
+

examples/md2pdf.c

@@ -0,0 +1,645 @@
+//
+// Simple markdown to PDF converter example for PDFio.
+//
+// Copyright © 2024 by Michael R Sweet.
+//
+// Licensed under Apache License v2.0.  See the file "LICENSE" for more
+// information.
+//
+// Usage:
+//
+//   ./md2pdf FILENAME.md >FILENAME.pdf
+//
+// The generated PDF file is formatted for a "universal" paper size (8.27x11",
+// the intersection of US Letter and ISO A4) with 1" top and bottom margins and
+// 0.5" side margins.  The document title (if present) is centered at the top
+// of the second and subsequent pages while the current heading and page number
+// are provided at the bottom of each page.
+//
+
+#include <stdio.h>
+#include <stdlib.h>
+#include <string.h>
+#include "mmd.h"
+#include <pdfio.h>
+#include <pdfio-content.h>
+
+
+//
+// Types...
+//
+
+typedef enum doccolor_e			// Document color enumeration
+{
+  DOCCOLOR_BLACK,			// #000
+  DOCCOLOR_RED,				// #900
+  DOCCOLOR_GREEN,			// #090
+  DOCCOLOR_BLUE,			// #00C
+  DOCCOLOR_GRAY				// #555
+} doccolor_t;
+
+typedef enum docfont_e			// Document font enumeration
+{
+  DOCFONT_REGULAR,			// Roboto-Regular
+  DOCFONT_BOLD,				// Roboto-Bold
+  DOCFONT_ITALIC,			// Roboto-Italic
+  DOCFONT_MONOSPACE,			// RobotoMono-Regular
+  DOCFONT_MAX				// Number of fonts
+} docfont_t;
+
+typedef struct docimage_s		// Document image info
+{
+  const char	*url;			// Reference URL
+  pdfio_obj_t	*obj;			// Image object
+} docimage_t;
+
+#define DOCIMAGE_MAX	1000		// Maximum number of images
+
+typedef struct docdata_s		// Document formatting data
+{
+  pdfio_file_t	*pdf;			// PDF file
+  pdfio_rect_t	media_box;		// Media (page) box
+  pdfio_rect_t	crop_box;		// Crop box (for margins)
+  pdfio_rect_t	art_box;		// Art box (for markdown content)
+  pdfio_obj_t	*fonts[DOCFONT_MAX];	// Embedded fonts
+  size_t	num_images;		// Number of embedded images
+  docimage_t	images[DOCIMAGE_MAX];	// Embedded images
+  const char	*title;			// Document title
+  char		*heading;		// Current document heading
+  pdfio_stream_t *st;			// Current page stream
+  double	y;			// Current position on page
+  doccolor_t	color;			// Current color
+} docdata_t;
+
+
+//
+// Macros...
+//
+
+#define in2pt(in) (in * 72.0)
+#define mm2pt(mm) (mm * 72.0 / 25.4)
+
+
+//
+// Constants...
+//
+
+static const char * const docfont_filenames[] =
+{
+  "Roboto-Regular.ttf",
+  "Roboto-Bold.ttf",
+  "Roboto-Italic.ttf",
+  "RobotoMono-Regular.ttf"
+};
+
+static const char * const docfont_names[] =
+{
+  "FR",
+  "FB",
+  "FI",
+  "FM"
+};
+
+#define LINE_HEIGHT	1.4		// Multiplier for line height
+
+#define SIZE_BODY	11.0		// Size of body text (points)
+#define SIZE_CODEBLOCK	10.0		// Size of code block text (points)
+#define SIZE_HEADFOOT	9.0		// Size of header/footer text (points)
+#define SIZE_HEADING_1	18.0		// Size of first level heading (points)
+#define SIZE_HEADING_2	16.0		// Size of top level heading (points)
+#define SIZE_HEADING_3	15.0		// Size of top level heading (points)
+#define SIZE_HEADING_4	14.0		// Size of top level heading (points)
+#define SIZE_HEADING_5	13.0		// Size of top level heading (points)
+#define SIZE_HEADING_6	12.0		// Size of top level heading (points)
+#define SIZE_TABLE	10.0		// Size of table text (points)
+
+#define PAGE_WIDTH	mm2pt(210)	// Page width in points
+#define PAGE_LENGTH	in2pt(11)	// Page length in points
+#define PAGE_LEFT	in2pt(0.5)	// Left margin in points
+#define PAGE_RIGHT	(PAGE_WIDTH - in2pt(0.5))
+					// Right margin in points
+#define PAGE_BOTTOM	in2pt(1.0)	// Bottom margin in points
+#define PAGE_TOP	(PAGE_LENGTH - in2pt(1.0))
+					// Top margin in points
+#define PAGE_HEADER	(PAGE_LENGTH - in2pt(0.5))
+					// Vertical position of header
+#define PAGE_FOOTER	in2pt(0.5)	// Vertical position of footer
+
+#define UNICODE_VALUE	true		// `true` for Unicode text, `false` for ISO-8859-1
+
+
+//
+// 'set_color()' - Set the stroke and fill color.
+//
+
+static void
+set_color(docdata_t  *dd,		// I - Document data
+          doccolor_t color)		// I - Document color
+{
+  switch (color)
+  {
+    case DOCCOLOR_BLACK :
+        pdfioContentSetFillColorDeviceGray(dd->st, 0.0);
+        pdfioContentSetStrokeColorDeviceGray(dd->st, 0.0);
+        break;
+    case DOCCOLOR_RED :
+        pdfioContentSetFillColorDeviceRGB(dd->st, 0.6, 0.0, 0.0);
+        pdfioContentSetStrokeColorDeviceRGB(dd->st, 0.6, 0.0, 0.0);
+        break;
+    case DOCCOLOR_GREEN :
+        pdfioContentSetFillColorDeviceRGB(dd->st, 0.0, 0.6, 0.0);
+        pdfioContentSetStrokeColorDeviceRGB(dd->st, 0.0, 0.6, 0.0);
+        break;
+    case DOCCOLOR_BLUE :
+        pdfioContentSetFillColorDeviceRGB(dd->st, 0.0, 0.0, 0.8);
+        pdfioContentSetStrokeColorDeviceRGB(dd->st, 0.0, 0.0, 0.8);
+        break;
+    case DOCCOLOR_GRAY :
+        pdfioContentSetFillColorDeviceGray(dd->st, 0.333);
+        pdfioContentSetStrokeColorDeviceGray(dd->st, 0.333);
+        break;
+  }
+
+  dd->color = color;
+}
+
+
+//
+// 'new_page()' - Start a new page.
+//
+
+static void
+new_page(docdata_t *dd)			// I - Document data
+{
+  pdfio_dict_t	*page_dict;		// Page dictionary
+  docfont_t	fontface;		// Current font face
+  size_t	i;			// Looping var
+  char		temp[32];		// Temporary string
+  double	width;			// Width of fragment
+
+
+  // Close the current page...
+  pdfioStreamClose(dd->st);
+
+  // Prep the new page...
+  page_dict = pdfioDictCreate(dd->pdf);
+
+  pdfioDictSetRect(page_dict, "MediaBox", &dd->media_box);
+//  pdfioDictSetRect(page_dict, "CropBox", &dd->crop_box);
+  pdfioDictSetRect(page_dict, "ArtBox", &dd->art_box);
+
+  for (fontface = DOCFONT_REGULAR; fontface < DOCFONT_MAX; fontface ++)
+    pdfioPageDictAddFont(page_dict, docfont_names[fontface], dd->fonts[fontface]);
+
+  for (i = 0; i < dd->num_images; i ++)
+  {
+    snprintf(temp, sizeof(temp), "I%u", (unsigned)i);
+    pdfioPageDictAddImage(page_dict, temp, dd->images[i].obj);
+  }
+
+  dd->st = pdfioFileCreatePage(dd->pdf, page_dict);
+
+  // Add header/footer text
+  set_color(dd, DOCCOLOR_GRAY);
+  pdfioContentSetTextFont(dd->st, docfont_names[DOCFONT_REGULAR], SIZE_HEADFOOT);
+
+  if (pdfioFileGetNumPages(dd->pdf) > 1 && dd->title)
+  {
+    // Show title in header...
+    width = pdfioContentTextMeasure(dd->fonts[DOCFONT_REGULAR], dd->title, SIZE_HEADFOOT);
+
+    pdfioContentTextBegin(dd->st);
+    pdfioContentTextMoveTo(dd->st, dd->crop_box.x1 + 0.5 * (dd->crop_box.x2 - dd->crop_box.x1 - width), dd->crop_box.y2 - SIZE_HEADFOOT);
+    pdfioContentTextShow(dd->st, UNICODE_VALUE, dd->title);
+    pdfioContentTextEnd(dd->st);
+
+    pdfioContentPathMoveTo(dd->st, dd->crop_box.x1, dd->crop_box.y2 - 2 * SIZE_HEADFOOT * LINE_HEIGHT + SIZE_HEADFOOT);
+    pdfioContentPathLineTo(dd->st, dd->crop_box.x2, dd->crop_box.y2 - 2 * SIZE_HEADFOOT * LINE_HEIGHT + SIZE_HEADFOOT);
+    pdfioContentStroke(dd->st);
+  }
+
+  // Show page number and current heading...
+  pdfioContentPathMoveTo(dd->st, dd->crop_box.x1, dd->crop_box.y1 + SIZE_HEADFOOT * LINE_HEIGHT);
+  pdfioContentPathLineTo(dd->st, dd->crop_box.x2, dd->crop_box.y1 + SIZE_HEADFOOT * LINE_HEIGHT);
+  pdfioContentStroke(dd->st);
+
+  pdfioContentTextBegin(dd->st);
+  snprintf(temp, sizeof(temp), "%u", (unsigned)pdfioFileGetNumPages(dd->pdf));
+  if (pdfioFileGetNumPages(dd->pdf) & 1)
+  {
+    // Page number on right...
+    width = pdfioContentTextMeasure(dd->fonts[DOCFONT_REGULAR], temp, SIZE_HEADFOOT);
+    pdfioContentTextMoveTo(dd->st, dd->crop_box.x2 - width, dd->crop_box.y1);
+  }
+  else
+  {
+    // Page number on left...
+    pdfioContentTextMoveTo(dd->st, dd->crop_box.x1, dd->crop_box.y1);
+  }
+
+  pdfioContentTextShow(dd->st, UNICODE_VALUE, temp);
+  pdfioContentTextEnd(dd->st);
+
+  if (dd->heading)
+  {
+    pdfioContentTextBegin(dd->st);
+
+    if (pdfioFileGetNumPages(dd->pdf) & 1)
+    {
+      // Current heading on left...
+      pdfioContentTextMoveTo(dd->st, dd->crop_box.x1, dd->crop_box.y1);
+    }
+    else
+    {
+      width = pdfioContentTextMeasure(dd->fonts[DOCFONT_REGULAR], dd->heading, SIZE_HEADFOOT);
+      pdfioContentTextMoveTo(dd->st, dd->crop_box.x2 - width, dd->crop_box.y1);
+    }
+
+    pdfioContentTextShow(dd->st, UNICODE_VALUE, dd->heading);
+    pdfioContentTextEnd(dd->st);
+  }
+
+  // The rest of the text will be full black...
+  set_color(dd, DOCCOLOR_BLACK);
+
+  dd->y = dd->art_box.y2;
+}
+
+
+//
+// 'format_block()' - Format a block of text
+//
+
+static void
+format_block(docdata_t  *dd,		// I - Document data
+             mmd_t      *block,		// I - Block to format
+             docfont_t  fontface,	// I - Default font
+             double     fontsize,	// I - Size of font
+             double     left,		// I - Left margin
+             double     right,		// I - Right margin
+             const char *leader)	// I - Leader text on the first line
+{
+  mmd_type_t	blocktype;		// Block type
+  mmd_t		*current,		// Current node
+		*next;			// Next node
+  mmd_type_t	curtype;		// Current node type
+  const char	*curtext,		// Current text
+		*cururl;		// Current URL, if any
+  bool		curws;			// Current whitespace
+  docfont_t	curface,		// Current font face
+		prevface;		// Previous font face
+  double	x, y;			// Current position
+  double	width,			// Width of current fragment
+		lwidth,			// Leader width
+		wswidth;		// Width of whitespace
+  doccolor_t	color;			// Color of text
+
+
+  blocktype = mmdGetType(block);
+
+  if (!dd->st)
+    new_page(dd);
+
+  if ((y = dd->y - 2.0 * fontsize * LINE_HEIGHT) < dd->art_box.y1)
+  {
+    new_page(dd);
+    y = dd->y - fontsize;
+  }
+
+  if (leader)
+  {
+    // Add leader text on first line...
+    pdfioContentSetTextFont(dd->st, docfont_names[prevface = fontface], fontsize);
+
+    lwidth = pdfioContentTextMeasure(dd->fonts[fontface], leader, fontsize);
+
+    pdfioContentTextBegin(dd->st);
+    pdfioContentTextMoveTo(dd->st, left - lwidth, y);
+    pdfioContentTextShow(dd->st, UNICODE_VALUE, leader);
+  }
+  else
+  {
+    // No leader text...
+    prevface = DOCFONT_MAX;
+    lwidth   = 0.0;
+  }
+
+  for (current = mmdGetFirstChild(block), x = left; current; current = next)
+  {
+    // Get information about the current node...
+    curtype = mmdGetType(current);
+    curtext = mmdGetText(current);
+    cururl  = mmdGetURL(current);
+    curws   = mmdGetWhitespace(current);
+
+//    fprintf(stderr, "current=%p, curtype=%d, curtext=\"%s\", cururl=\"%s\", curws=%s\n", (void *)current, curtype, curtext, cururl, curws ? "true" : "false");
+
+    // Figure out the next node under this block...
+    if ((next = mmdGetFirstChild(current)) == NULL)
+    {
+      if ((next = mmdGetNextSibling(current)) == NULL)
+      {
+        mmd_t *parent;			// Parent node
+
+        if ((parent = mmdGetParent(current)) != block)
+        {
+          while ((next = mmdGetNextSibling(parent)) == NULL)
+          {
+            if ((parent = mmdGetParent(parent)) == block)
+              break;
+          }
+        }
+      }
+    }
+
+    // Process the node...
+    if (!curtext)
+      continue;
+
+    if (curtype == MMD_TYPE_EMPHASIZED_TEXT)
+      curface = DOCFONT_ITALIC;
+    else if (curtype == MMD_TYPE_STRONG_TEXT)
+      curface = DOCFONT_BOLD;
+    else  if (curtype == MMD_TYPE_CODE_TEXT)
+      curface = DOCFONT_MONOSPACE;
+    else
+      curface = fontface;
+
+    if (curtype == MMD_TYPE_CODE_TEXT)
+      color = DOCCOLOR_RED;
+    else if (curtype == MMD_TYPE_LINKED_TEXT)
+      color = DOCCOLOR_BLUE;
+    else
+      color = DOCCOLOR_BLACK;
+
+    width = pdfioContentTextMeasure(dd->fonts[curface], curtext, fontsize);
+    if (curws)
+      wswidth = pdfioContentTextMeasure(dd->fonts[curface], " ", fontsize);
+    else
+      wswidth = 0.0;
+
+    if (x > left && (x + width + wswidth) >= right)
+    {
+      // New line...
+      x = left;
+      y -= fontsize * LINE_HEIGHT;
+
+      if (y < dd->art_box.y1)
+      {
+        // New page...
+	if (prevface != DOCFONT_MAX)
+	{
+	  pdfioContentTextEnd(dd->st);
+	  prevface = DOCFONT_MAX;
+	}
+
+        new_page(dd);
+
+        y = dd->y - fontsize * LINE_HEIGHT;
+      }
+      else
+      {
+	pdfioContentTextMoveTo(dd->st, lwidth, -fontsize * LINE_HEIGHT);
+	lwidth = 0.0;
+      }
+    }
+
+    if (curface != prevface)
+    {
+      if (prevface == DOCFONT_MAX)
+      {
+	pdfioContentTextBegin(dd->st);
+	pdfioContentTextMoveTo(dd->st, x, y);
+      }
+
+      pdfioContentSetTextFont(dd->st, docfont_names[prevface = curface], fontsize);
+    }
+
+    if (color != dd->color)
+      set_color(dd, color);
+
+    if (x > left && curws)
+    {
+      pdfioContentTextShowf(dd->st, UNICODE_VALUE, " %s", curtext);
+      x += width + wswidth;
+    }
+    else
+    {
+      pdfioContentTextShow(dd->st, UNICODE_VALUE, curtext);
+      x += width;
+    }
+
+    if (blocktype == MMD_TYPE_CODE_BLOCK)
+    {
+      // Force a new line...
+      x = left;
+      y -= fontsize * LINE_HEIGHT;
+
+      if (y < dd->art_box.y1)
+      {
+        // New page...
+	if (prevface != DOCFONT_MAX)
+	{
+	  pdfioContentTextEnd(dd->st);
+	  prevface = DOCFONT_MAX;
+	}
+
+        new_page(dd);
+
+        y = dd->y - fontsize * LINE_HEIGHT;
+      }
+      else
+      {
+	pdfioContentTextMoveTo(dd->st, lwidth, -fontsize * LINE_HEIGHT);
+	lwidth = 0.0;
+      }
+    }
+  }
+
+  // End the current text block and save out position on the page...
+  if (prevface != DOCFONT_MAX)
+    pdfioContentTextEnd(dd->st);
+
+  dd->y = y;
+}
+
+
+//
+// 'format_doc()' - Format a document.
+//
+
+static void
+format_doc(docdata_t *dd,		// I - Document data
+           mmd_t     *doc,		// I - Document node to format
+           double    left,		// I - Left margin
+           double    right)		// I - Right margin
+{
+  int		i;			// Child number
+  mmd_type_t	doctype;		// Document node type
+  mmd_t		*current;		// Current node
+  mmd_type_t	curtype;		// Current node type
+  char		leader[32];		// Leader
+  static const double heading_sizes[] =	// Heading font sizes
+  {
+    SIZE_HEADING_1,
+    SIZE_HEADING_2,
+    SIZE_HEADING_3,
+    SIZE_HEADING_4,
+    SIZE_HEADING_5,
+    SIZE_HEADING_6
+  };
+
+
+  doctype = mmdGetType(doc);
+
+  for (i = 1, current = mmdGetFirstChild(doc); current; i ++, current = mmdGetNextSibling(current))
+  {
+    switch (curtype = mmdGetType(current))
+    {
+      default :
+          break;
+
+      case MMD_TYPE_BLOCK_QUOTE :
+          format_doc(dd, current, left + 36.0, right - 36.0);
+          break;
+
+      case MMD_TYPE_ORDERED_LIST :
+      case MMD_TYPE_UNORDERED_LIST :
+          format_doc(dd, current, left + 36.0, right);
+          break;
+
+      case MMD_TYPE_LIST_ITEM :
+          if (doctype == MMD_TYPE_ORDERED_LIST)
+          {
+            snprintf(leader, sizeof(leader), "%d. ", i);
+            format_block(dd, current, DOCFONT_REGULAR, SIZE_BODY, left, right, leader);
+          }
+          else
+          {
+            format_block(dd, current, DOCFONT_REGULAR, SIZE_BODY, left, right, /*leader*/"• ");
+          }
+          break;
+
+      case MMD_TYPE_HEADING_1 :
+      case MMD_TYPE_HEADING_2 :
+      case MMD_TYPE_HEADING_3 :
+      case MMD_TYPE_HEADING_4 :
+      case MMD_TYPE_HEADING_5 :
+      case MMD_TYPE_HEADING_6 :
+          if (dd->heading)
+            free(dd->heading);
+
+          dd->heading = mmdCopyAllText(current);
+
+          format_block(dd, current, DOCFONT_BOLD, heading_sizes[curtype - MMD_TYPE_HEADING_1], left, right, /*leader*/NULL);
+          break;
+
+      case MMD_TYPE_PARAGRAPH :
+          format_block(dd, current, DOCFONT_REGULAR, SIZE_BODY, left, right, /*leader*/NULL);
+          break;
+
+      case MMD_TYPE_CODE_BLOCK :
+          format_block(dd, current, DOCFONT_MONOSPACE, SIZE_BODY, left + 36.0, right, /*leader*/NULL);
+          break;
+    }
+  }
+}
+
+
+//
+// 'output_cb()' - Write PDF data to the standard output...
+//
+
+static ssize_t				// O - Number of bytes written
+output_cb(void       *output_cbdata,	// I - Callback data (not used)
+          const void *buffer,		// I - Buffer to write
+          size_t     bytes)		// I - Number of bytes to write
+{
+  (void)output_cbdata;
+
+  return ((ssize_t)fwrite(buffer, 1, bytes, stdout));
+}
+
+
+//
+// 'main()' - Convert markdown to PDF.
+//
+
+int					// O - Exit status
+main(int  argc,				// I - Number of command-line arguments
+     char *argv[])			// I - Command-line arguments
+{
+  docdata_t	dd;			// Document data
+  docfont_t	fontface;		// Current font
+  mmd_t		*doc;			// Markdown document
+  const char	*value;			// Metadata value
+
+
+  // Get the markdown file from the command-line...
+  if (argc != 2)
+  {
+    fputs("Usage: md2pdf FILENANE.md >FILENAME.pdf\n", stderr);
+    return (1);
+  }
+
+  if ((doc = mmdLoad(/*root*/NULL, argv[1])) == NULL)
+    return (1);
+
+  // Initialize the document data
+  memset(&dd, 0, sizeof(dd));
+
+  dd.media_box.x2 = PAGE_WIDTH;
+  dd.media_box.y2 = PAGE_LENGTH;
+
+  dd.crop_box.x1 = PAGE_LEFT;
+  dd.crop_box.y1 = PAGE_FOOTER;
+  dd.crop_box.x2 = PAGE_RIGHT;
+  dd.crop_box.y2 = PAGE_HEADER;
+
+  dd.art_box.x1  = PAGE_LEFT;
+  dd.art_box.y1  = PAGE_BOTTOM;
+  dd.art_box.x2  = PAGE_RIGHT;
+  dd.art_box.y2  = PAGE_TOP;
+
+  dd.title       = mmdGetMetadata(doc, "title");
+
+  // Output a PDF file to the standard output...
+#ifdef _WIN32
+  setmode(1, O_BINARY);			// Force binary output on Windows
+#endif // _WIN32
+
+  if ((dd.pdf = pdfioFileCreateOutput(output_cb, /*output_cbdata*/NULL, /*version*/NULL, /*media_box*/NULL, /*crop_box*/NULL, /*error_cb*/NULL, /*error_data*/NULL)) == NULL)
+    return (1);
+
+  if ((value = mmdGetMetadata(doc, "author")) != NULL)
+    pdfioFileSetAuthor(dd.pdf, value);
+
+  if ((value = mmdGetMetadata(doc, "keywords")) != NULL)
+    pdfioFileSetKeywords(dd.pdf, value);
+
+  if ((value = mmdGetMetadata(doc, "subject")) != NULL)
+    pdfioFileSetSubject(dd.pdf, value);
+  else if ((value = mmdGetMetadata(doc, "copyright")) != NULL)
+    pdfioFileSetSubject(dd.pdf, value);
+
+  if (dd.title)
+    pdfioFileSetTitle(dd.pdf, dd.title);
+
+  // Add fonts...
+  for (fontface = DOCFONT_REGULAR; fontface < DOCFONT_MAX; fontface ++)
+  {
+    if ((dd.fonts[fontface] = pdfioFileCreateFontObjFromFile(dd.pdf, docfont_filenames[fontface], UNICODE_VALUE)) == NULL)
+      return (1);
+  }
+
+  // Parse the markdown document...
+  format_doc(&dd, doc, dd.art_box.x1, dd.art_box.x2);
+
+  // Close the PDF and return...
+  pdfioStreamClose(dd.st);
+  pdfioFileClose(dd.pdf);
+
+  mmdFree(doc);
+
+  return (0);
+}

examples/mmd.h

@@ -0,0 +1,112 @@
+//
+// Header file for miniature markdown library.
+//
+//     https://www.msweet.org/mmd
+//
+// Copyright © 2017-2024 by Michael R Sweet.
+//
+// Licensed under Apache License v2.0.  See the file "LICENSE" for more
+// information.
+//
+
+#ifndef MMD_H
+#  define MMD_H
+#  include <stdio.h>
+#  include <stdbool.h>
+#  ifdef __cplusplus
+extern "C" {
+#  endif // __cplusplus
+
+
+//
+// Constants...
+//
+
+enum mmd_option_e
+{
+  MMD_OPTION_NONE = 0x00,		// No markdown extensions
+  MMD_OPTION_METADATA = 0x01,		// Jekyll metadata extension
+  MMD_OPTION_TABLES = 0x02,		// Github table extension
+  MMD_OPTION_TASKS = 0x04,		// Github task item extension (check boxes)
+  MMD_OPTION_ALL = 0x07			// All supported markdown extensions
+};
+typedef unsigned mmd_option_t;
+
+typedef enum mmd_type_e
+{
+  MMD_TYPE_NONE = -1,
+  MMD_TYPE_DOCUMENT,			// The document root
+  MMD_TYPE_METADATA,			// Document metadata
+  MMD_TYPE_BLOCK_QUOTE,			// <blockquote>
+  MMD_TYPE_ORDERED_LIST,		// <ol>
+  MMD_TYPE_UNORDERED_LIST,		// <ul>
+  MMD_TYPE_LIST_ITEM,			// <li>
+  MMD_TYPE_TABLE,			// <table>
+  MMD_TYPE_TABLE_HEADER,		// <thead>
+  MMD_TYPE_TABLE_BODY,			// <tbody>
+  MMD_TYPE_TABLE_ROW,			// <tr>
+  MMD_TYPE_HEADING_1 = 10,		// <h1>
+  MMD_TYPE_HEADING_2,			// <h2>
+  MMD_TYPE_HEADING_3,			// <h3>
+  MMD_TYPE_HEADING_4,			// <h4>
+  MMD_TYPE_HEADING_5,			// <h5>
+  MMD_TYPE_HEADING_6,			// <h6>
+  MMD_TYPE_PARAGRAPH,			// <p>
+  MMD_TYPE_CODE_BLOCK,			// <pre><code>
+  MMD_TYPE_THEMATIC_BREAK,		// <hr />
+  MMD_TYPE_TABLE_HEADER_CELL,		// <th>
+  MMD_TYPE_TABLE_BODY_CELL_LEFT,	// <td align="left">
+  MMD_TYPE_TABLE_BODY_CELL_CENTER,	// <td align="center">
+  MMD_TYPE_TABLE_BODY_CELL_RIGHT,	// <td align="right">
+  MMD_TYPE_NORMAL_TEXT = 100,		// Normal text
+  MMD_TYPE_EMPHASIZED_TEXT,		// <em>text</em>
+  MMD_TYPE_STRONG_TEXT,			// <strong>text</strong>
+  MMD_TYPE_STRUCK_TEXT,			// <del>text</del>
+  MMD_TYPE_LINKED_TEXT,			// <a href="link">text</a>
+  MMD_TYPE_CODE_TEXT,			// <code>text</code>
+  MMD_TYPE_IMAGE,			// <img src="link" />
+  MMD_TYPE_HARD_BREAK,			// <br />
+  MMD_TYPE_SOFT_BREAK,			// <wbr />
+  MMD_TYPE_METADATA_TEXT,		// name: value
+  MMD_TYPE_CHECKBOX			// [ ] or [x]
+} mmd_type_t;
+
+
+//
+// Types...
+//
+
+typedef struct _mmd_s mmd_t;		// Markdown node
+typedef size_t (*mmd_iocb_t)(void *cbdata, char *buffer, size_t bytes);
+					// mmdLoadIO callback function
+
+//
+// Functions...
+//
+
+extern char         *mmdCopyAllText(mmd_t *node);
+extern void         mmdFree(mmd_t *node);
+extern const char   *mmdGetExtra(mmd_t *node);
+extern mmd_t        *mmdGetFirstChild(mmd_t *node);
+extern mmd_t        *mmdGetLastChild(mmd_t *node);
+extern const char   *mmdGetMetadata(mmd_t *doc, const char *keyword);
+extern mmd_t        *mmdGetNextSibling(mmd_t *node);
+extern mmd_option_t mmdGetOptions(void);
+extern mmd_t        *mmdGetParent(mmd_t *node);
+extern mmd_t        *mmdGetPrevSibling(mmd_t *node);
+extern const char   *mmdGetText(mmd_t *node);
+extern mmd_type_t   mmdGetType(mmd_t *node);
+extern const char   *mmdGetURL(mmd_t *node);
+extern bool         mmdGetWhitespace(mmd_t *node);
+extern bool         mmdIsBlock(mmd_t *node);
+extern mmd_t        *mmdLoad(mmd_t *root, const char *filename);
+extern mmd_t        *mmdLoadFile(mmd_t *root, FILE *fp);
+extern mmd_t        *mmdLoadIO(mmd_t *root, mmd_iocb_t cb, void *cbdata);
+extern mmd_t        *mmdLoadString(mmd_t *root, const char *s);
+extern void         mmdSetOptions(mmd_option_t options);
+
+
+#  ifdef __cplusplus
+}
+#  endif // __cplusplus
+#endif // !MMD_H

makesrcdist

@@ -2,18 +2,61 @@
 #
 # makesrcdist - make a source distribution of pdfio.
 #
+# Usage:
+#
+#   ./makesrcdist [--snapshot] VERSION
+#
 
+# Support "--snapshot" option...
+if test "$1" == "--snapshot"; then
+    shift
+    snapshot=1
+else
+    snapshot=0
+fi
+
+# Get version...
 if test $# != 1; then
-	echo "Usage: ./makesrcdist version"
-	exit 1
+    echo "Usage: ./makesrcdist [--snapshot] VERSION"
+    exit 1
 fi
 
 version=$1
 
-echo Creating tag for release...
-git tag -m "Tag $version" v$version
-git push origin v$version
+# Check that version number has been updated everywhere...
+if test $(grep AC_INIT configure.ac | awk '{print $2}') != "[$version],"; then
+    echo "Still need to update AC_INIT version in 'configure.ac'."
+    exit 1
+fi
 
+if test $(grep PDFIO_VERSION= configure | awk -F \" '{print $2}') != "$version"; then
+    echo "Still need to run 'autoconf -f'."
+    exit 1
+fi
+
+if test $(grep '<version>' pdfio_native.nuspec | sed -E -e '1,$s/^.*<version>([0-9.]+).*$/\1/') != "$version"; then
+    echo "Still need to update version in 'pdfio_native.nuspec'."
+    exit 1
+fi
+
+if test $(grep '<version>' pdfio_native.redist.nuspec | sed -E -e '1,$s/^.*<version>([0-9.]+).*$/\1/') != "$version"; then
+    echo "Still need to update version in 'pdfio_native.redist.nuspec'."
+    exit 1
+fi
+
+if test $(grep PDFIO_VERSION pdfio.h | awk -F \" '{print $2}') != "$version"; then
+    echo "Still need to update PDFIO_VERSION in 'pdfio.h'."
+    exit 1
+fi
+
+# Tag release...
+if test $snapshot = 0; then
+    echo Creating tag for release...
+    git tag -m "Tag $version" v$version
+    git push origin v$version
+fi
+
+# Make source archives...
 echo Creating pdfio-$version.tar.gz...
 git archive --format tar --prefix=pdfio-$version/ HEAD | gzip -v9 >pdfio-$version.tar.gz
 gpg --detach-sign pdfio-$version.tar.gz

pdfio-array.c

@@ -1,7 +1,7 @@
 //
 // PDF array functions for PDFio.
 //
-// Copyright © 2021 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -363,6 +363,9 @@ _pdfioArrayDebug(pdfio_array_t *a,	// I - Array
   _pdfio_value_t	*v;		// Current value
 
 
+  if (!a)
+    return;
+
   putc('[', fp);
   for (i = a->num_values, v = a->values; i > 0; i --, v ++)
     _pdfioValueDebug(v, fp);
@@ -634,6 +637,28 @@ _pdfioArrayRead(pdfio_file_t   *pdf,	// I - PDF file
 }
 
 
+//
+// 'pdfioArrayRemove()' - Remove an array entry.
+//
+
+bool					// O - `true` on success, `false` otherwise
+pdfioArrayRemove(pdfio_array_t *a,	// I - Array
+                 size_t        n)	// I - Index
+{
+  if (!a || n >= a->num_values)
+    return (false);
+
+  if (a->values[n].type == PDFIO_VALTYPE_BINARY)
+    free(a->values[n].value.binary.data);
+
+  a->num_values --;
+  if (n < a->num_values)
+    memmove(a->values + n, a->values + n + 1, (a->num_values - n) * sizeof(_pdfio_value_t));
+
+  return (true);
+}
+
+
 //
 // '_pdfioArrayWrite()' - Write an array to a PDF file.
 //

pdfio-common.c

@@ -1,7 +1,7 @@
 //
 // Common support functions for pdfio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -261,7 +261,7 @@ _pdfioFilePrintf(pdfio_file_t *pdf,	// I - PDF file
 
   // Format the string...
   va_start(ap, format);
-  vsnprintf(buffer, sizeof(buffer), format, ap);
+  _pdfio_vsnprintf(pdf, buffer, sizeof(buffer), format, ap);
   va_end(ap);
 
   // Write it...

pdfio-content.c

@@ -2349,7 +2349,7 @@ pdfioPageDictAddColorSpace(
 bool					// O - `true` on success, `false` on failure
 pdfioPageDictAddFont(
     pdfio_dict_t *dict,			// I - Page dictionary
-    const char   *name,			// I - Font name
+    const char   *name,			// I - Font name; must not contain spaces
     pdfio_obj_t  *obj)			// I - Font object
 {
   pdfio_dict_t	*resources;		// Resource dictionary

pdfio-dict.c

@@ -1,7 +1,7 @@
 //
 // PDF dictionary functions for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -18,19 +18,22 @@ static int	compare_pairs(_pdfio_pair_t *a, _pdfio_pair_t *b);
 
 
 //
-// '_pdfioDictClear()' - Remove a key/value pair from a dictionary.
+// 'pdfioDictClear()' - Remove a key/value pair from a dictionary.
 //
 
-void
-_pdfioDictClear(pdfio_dict_t *dict,	// I - Dictionary
-                const char   *key)	// I - Key
+bool					// O - `true` if cleared, `false` otherwise
+pdfioDictClear(pdfio_dict_t *dict,	// I - Dictionary
+               const char   *key)	// I - Key
 {
   size_t	idx;			// Index into pairs
   _pdfio_pair_t	*pair,			// Current pair
 		pkey;			// Search key
 
 
-  PDFIO_DEBUG("_pdfioDictClear(dict=%p, key=\"%s\")\n", dict, key);
+  PDFIO_DEBUG("pdfioDictClear(dict=%p, key=\"%s\")\n", dict, key);
+
+  if (!dict || !key)
+    return (false);
 
   // See if the key is already set...
   if (dict->num_pairs > 0)
@@ -48,8 +51,12 @@ _pdfioDictClear(pdfio_dict_t *dict,	// I - Dictionary
 
       if (idx < dict->num_pairs)
         memmove(pair, pair + 1, (dict->num_pairs - idx) * sizeof(_pdfio_pair_t));
+
+      return (true);
     }
   }
+
+  return (false);
 }
 
 
@@ -194,6 +201,9 @@ _pdfioDictDebug(pdfio_dict_t *dict,	// I - Dictionary
   _pdfio_pair_t	*pair;			// Current pair
 
 
+  if (!dict)
+    return;
+
   for (i = dict->num_pairs, pair = dict->pairs; i > 0; i --, pair ++)
   {
     fprintf(fp, "/%s", pair->key);
@@ -332,6 +342,18 @@ pdfioDictGetDict(pdfio_dict_t *dict,	// I - Dictionary
 }
 
 
+//
+// 'pdfioDictGetKey()' - Get the key for the specified pair.
+//
+
+const char *				// O - Key for specified pair
+pdfioDictGetKey(pdfio_dict_t *dict,	// I - Dictionary
+                size_t       n)		// I - Pair index (`0`-based)
+{
+  return ((dict && n < dict->num_pairs) ? dict->pairs[n].key : NULL);
+}
+
+
 //
 // 'pdfioDictGetName()' - Get a key name value from a dictionary.
 //
@@ -350,6 +372,17 @@ pdfioDictGetName(pdfio_dict_t *dict,	// I - Dictionary
 }
 
 
+//
+// 'pdfioDictGetNumPairs()' - Get the number of key/value pairs in a dictionary.
+//
+
+size_t					// O - Number of pairs
+pdfioDictGetNumPairs(pdfio_dict_t *dict)// I - Dictionary
+{
+  return (dict ? dict->num_pairs : 0);
+}
+
+
 //
 // 'pdfioDictGetNumber()' - Get a key number value from a dictionary.
 //
@@ -426,9 +459,27 @@ pdfioDictGetString(pdfio_dict_t *dict,	// I - Dictionary
 
 
   if (value && value->type == PDFIO_VALTYPE_STRING)
+  {
     return (value->value.string);
+  }
+  else if (value && value->type == PDFIO_VALTYPE_BINARY && value->value.binary.datalen < 4096)
+  {
+    // Convert binary string to regular string...
+    char	temp[4096];		// Temporary string
+
+    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;
+    value->value.string = pdfioStringCreate(dict->pdf, temp);
+
+    return (value->value.string);
+  }
   else
+  {
     return (NULL);
+  }
 }
 
 

pdfio-file.c

@@ -1,7 +1,7 @@
 //
 // PDF file functions for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -19,11 +19,12 @@
 
 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 pdfio_file_t	*create_common(const char *filename, int fd, pdfio_output_cb_t output_cb, void *output_cbdata, const char *version, pdfio_rect_t *media_box, pdfio_rect_t *crop_box, pdfio_error_cb_t error_cb, void *error_cbdata);
 static const char	*get_info_string(pdfio_file_t *pdf, const char *key);
+static struct lconv	*get_lconv(void);
 static bool		load_obj_stream(pdfio_obj_t *obj);
 static bool		load_pages(pdfio_file_t *pdf, pdfio_obj_t *obj, size_t depth);
 static bool		load_xref(pdfio_file_t *pdf, off_t xref_offset, pdfio_password_cb_t password_cb, void *password_data);
-static bool		write_catalog(pdfio_file_t *pdf);
 static bool		write_pages(pdfio_file_t *pdf);
 static bool		write_trailer(pdfio_file_t *pdf);
 
@@ -119,11 +120,8 @@ pdfioFileClose(pdfio_file_t *pdf)	// I - PDF file
   {
     ret = false;
 
-    if (pdfioObjClose(pdf->info_obj))
-      if (write_pages(pdf))
-	if (write_catalog(pdf))
-	  if (write_trailer(pdf))
-	    ret = _pdfioFileFlush(pdf);
+    if (pdfioObjClose(pdf->info_obj) && write_pages(pdf) && pdfioObjClose(pdf->root_obj) && write_trailer(pdf))
+      ret = _pdfioFileFlush(pdf);
   }
 
   if (pdf->fd >= 0 && close(pdf->fd) < 0)
@@ -172,7 +170,7 @@ pdfioFileClose(pdfio_file_t *pdf)	// I - PDF file
 // CropBox for pages in the PDF file - if `NULL` then a default "Universal" size
 // of 8.27x11in (the intersection of US Letter and ISO A4) is used.
 //
-// The "error_cb" and "error_data" arguments specify an error handler callback
+// The "error_cb" and "error_cbdata" arguments specify an error handler callback
 // and its data pointer - if `NULL` the default error handler is used that
 // writes error messages to `stderr`.
 //
@@ -184,122 +182,39 @@ pdfioFileCreate(
     pdfio_rect_t     *media_box,	// I - Default MediaBox for pages
     pdfio_rect_t     *crop_box,		// I - Default CropBox for pages
     pdfio_error_cb_t error_cb,		// I - Error callback or `NULL` for default
-    void             *error_data)	// I - Error callback data, if any
+    void             *error_cbdata)	// I - Error callback data, if any
 {
   pdfio_file_t	*pdf;			// PDF file
-  pdfio_dict_t	*dict;			// Dictionary for pages object
-  pdfio_dict_t	*info_dict;		// Dictionary for information object
-  unsigned char	id_value[16];		// File ID value
+  int		fd;			// File descriptor
 
 
+  PDFIO_DEBUG("pdfioFileCreate(filename=\"%s\", version=\"%s\", media_box=%p, crop_box=%p, error_cb=%p, error_cbdata=%p)\n", filename, version, (void *)media_box, (void *)crop_box, (void *)error_cb, (void *)error_cbdata);
+
   // Range check input...
   if (!filename)
     return (NULL);
 
-  if (!version)
-    version = "2.0";
-
-  if (!error_cb)
-  {
-    error_cb   = _pdfioFileDefaultError;
-    error_data = NULL;
-  }
-
-  // Allocate a PDF file structure...
-  if ((pdf = (pdfio_file_t *)calloc(1, sizeof(pdfio_file_t))) == NULL)
+  // Create the file...
+  if ((fd = open(filename, O_WRONLY | O_BINARY | O_CREAT | O_TRUNC, 0666)) < 0)
   {
     pdfio_file_t temp;			// Dummy file
     char	message[8192];		// Message string
 
     temp.filename = (char *)filename;
-    snprintf(message, sizeof(message), "Unable to allocate memory for PDF file - %s", strerror(errno));
-    (error_cb)(&temp, message, error_data);
+    snprintf(message, sizeof(message), "Unable to create '%s': %s", filename, strerror(errno));
+    (error_cb)(&temp, message, error_cbdata);
+
     return (NULL);
   }
 
-  pdf->filename    = strdup(filename);
-  pdf->version     = strdup(version);
-  pdf->mode        = _PDFIO_MODE_WRITE;
-  pdf->error_cb    = error_cb;
-  pdf->error_data  = error_data;
-  pdf->permissions = PDFIO_PERMISSION_ALL;
-  pdf->bufptr      = pdf->buffer;
-  pdf->bufend      = pdf->buffer + sizeof(pdf->buffer);
-
-  if (media_box)
+  if ((pdf = create_common(filename, fd, /*output_cb*/NULL, /*output_cbdata*/NULL, version, media_box, crop_box, error_cb, error_cbdata)) == NULL)
   {
-    pdf->media_box = *media_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->media_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->media_box.y2 = 11.0f * 72.0f;
-  }
-
-  if (crop_box)
-  {
-    pdf->crop_box = *crop_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->crop_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->crop_box.y2 = 11.0f * 72.0f;
-  }
-
-  // Create the file...
-  if ((pdf->fd = open(filename, O_WRONLY | O_BINARY | O_CREAT | O_TRUNC, 0666)) < 0)
-  {
-    _pdfioFileError(pdf, "Unable to create file - %s", strerror(errno));
-    free(pdf->filename);
-    free(pdf->version);
-    free(pdf);
-    return (NULL);
-  }
-
-  // Write a standard PDF header...
-  if (!_pdfioFilePrintf(pdf, "%%PDF-%s\n%%\342\343\317\323\n", version))
-    goto error;
-
-  // Create the pages object...
-  if ((dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetName(dict, "Type", "Pages");
-
-  if ((pdf->pages_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
-    goto error;
-
-  // Create the info object...
-  if ((info_dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetDate(info_dict, "CreationDate", time(NULL));
-  pdfioDictSetString(info_dict, "Producer", "pdfio/" PDFIO_VERSION);
-
-  if ((pdf->info_obj = pdfioFileCreateObj(pdf, info_dict)) == NULL)
-    goto error;
-
-  // Create random file ID values...
-  _pdfioCryptoMakeRandom(id_value, sizeof(id_value));
-
-  if ((pdf->id_array = pdfioArrayCreate(pdf)) != NULL)
-  {
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
+    // Remove the newly created file if we can't create the PDF file object...
+    close(fd);
+    unlink(filename);
   }
 
   return (pdf);
-
-  // Common error handling code...
-  error:
-
-  pdfioFileClose(pdf);
-
-  unlink(filename);
-
-  return (NULL);
 }
 
 
@@ -329,6 +244,35 @@ pdfioFileCreateArrayObj(
 }
 
 
+//
+// 'pdfioFileCreateNameObj()' - Create a new object in a PDF file containing a name.
+//
+// This function creates a new object with a name value in a PDF file.
+// You must call @link pdfioObjClose@ to write the object to the file.
+//
+
+pdfio_obj_t *				// O - New object
+pdfioFileCreateNameObj(
+    pdfio_file_t *pdf,			// I - PDF file
+    const char   *name)			// I - Name value
+{
+  _pdfio_value_t	value;		// Object value
+
+
+  // Range check input...
+  if (!pdf || !name)
+    return (NULL);
+
+  value.type         = PDFIO_VALTYPE_NAME;
+  value.value.name   = pdfioStringCreate(pdf, name);
+
+  if (!value.value.name)
+    return (NULL);
+
+  return (_pdfioFileCreateObj(pdf, NULL, &value));
+}
+
+
 //
 // 'pdfioFileCreateNumberObj()' - Create a new object in a PDF file containing a number.
 //
@@ -439,13 +383,13 @@ _pdfioFileCreateObj(
 // 'pdfioFileCreateOutput()' - Create a PDF file through an output callback.
 //
 // This function creates a new PDF file that is streamed though an output
-// callback.  The "output_cb" and "output_ctx" arguments specify the output
-// callback and its context pointer which is called whenever data needs to be
+// callback.  The "output_cb" and "output_cbdata" arguments specify the output
+// callback and its data pointer which is called whenever data needs to be
 // written:
 //
 // ```
 // ssize_t
-// output_cb(void *output_ctx, const void *buffer, size_t bytes)
+// output_cb(void *output_cbdata, const void *buffer, size_t bytes)
 // {
 //   // Write buffer to output and return the number of bytes written
 // }
@@ -458,7 +402,7 @@ _pdfioFileCreateObj(
 // CropBox for pages in the PDF file - if `NULL` then a default "Universal" size
 // of 8.27x11in (the intersection of US Letter and ISO A4) is used.
 //
-// The "error_cb" and "error_data" arguments specify an error handler callback
+// The "error_cb" and "error_cbdata" arguments specify an error handler callback
 // and its data pointer - if `NULL` the default error handler is used that
 // writes error messages to `stderr`.
 //
@@ -469,121 +413,17 @@ _pdfioFileCreateObj(
 
 pdfio_file_t *				// O - PDF file or `NULL` on error
 pdfioFileCreateOutput(
-    pdfio_output_cb_t output_cb,	// I - Output callback
-    void              *output_ctx,	// I - Output context
+    pdfio_output_cb_t output_cb,	// I - Output callback function
+    void              *output_cbdata,	// I - Output callback data
     const char        *version,		// I - PDF version number or `NULL` for default (2.0)
     pdfio_rect_t      *media_box,	// I - Default MediaBox for pages
     pdfio_rect_t      *crop_box,	// I - Default CropBox for pages
     pdfio_error_cb_t  error_cb,		// I - Error callback or `NULL` for default
-    void              *error_data)	// I - Error callback data, if any
+    void              *error_cbdata)	// I - Error callback data, if any
 {
-  pdfio_file_t	*pdf;			// PDF file
-  pdfio_dict_t	*dict;			// Dictionary for pages object
-  pdfio_dict_t	*info_dict;		// Dictionary for information object
-  unsigned char	id_value[16];		// File ID value
+  PDFIO_DEBUG("pdfioFileCreate(output_cb=%p, output_cbdata=%p, version=\"%s\", media_box=%p, crop_box=%p, error_cb=%p, error_cbdata=%p)\n", (void *)output_cb, (void *)output_cbdata, version, (void *)media_box, (void *)crop_box, (void *)error_cb, (void *)error_cbdata);
 
-
-  // Range check input...
-  if (!output_cb)
-    return (NULL);
-
-  if (!version)
-    version = "2.0";
-
-  if (!error_cb)
-  {
-    error_cb   = _pdfioFileDefaultError;
-    error_data = NULL;
-  }
-
-  // Allocate a PDF file structure...
-  if ((pdf = (pdfio_file_t *)calloc(1, sizeof(pdfio_file_t))) == NULL)
-  {
-    pdfio_file_t temp;			// Dummy file
-    char	message[8192];		// Message string
-
-    temp.filename = (char *)"output.pdf";
-    snprintf(message, sizeof(message), "Unable to allocate memory for PDF file - %s", strerror(errno));
-    (error_cb)(&temp, message, error_data);
-    return (NULL);
-  }
-
-  pdf->filename    = strdup("output.pdf");
-  pdf->version     = strdup(version);
-  pdf->mode        = _PDFIO_MODE_WRITE;
-  pdf->error_cb    = error_cb;
-  pdf->error_data  = error_data;
-  pdf->permissions = PDFIO_PERMISSION_ALL;
-  pdf->bufptr      = pdf->buffer;
-  pdf->bufend      = pdf->buffer + sizeof(pdf->buffer);
-
-  if (media_box)
-  {
-    pdf->media_box = *media_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->media_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->media_box.y2 = 11.0f * 72.0f;
-  }
-
-  if (crop_box)
-  {
-    pdf->crop_box = *crop_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->crop_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->crop_box.y2 = 11.0f * 72.0f;
-  }
-
-  // Save output callback...
-  pdf->fd         = -1;
-  pdf->output_cb  = output_cb;
-  pdf->output_ctx = output_ctx;
-
-  // Write a standard PDF header...
-  if (!_pdfioFilePrintf(pdf, "%%PDF-%s\n%%\342\343\317\323\n", version))
-    goto error;
-
-  // Create the pages object...
-  if ((dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetName(dict, "Type", "Pages");
-
-  if ((pdf->pages_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
-    goto error;
-
-  // Create the info object...
-  if ((info_dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetDate(info_dict, "CreationDate", time(NULL));
-  pdfioDictSetString(info_dict, "Producer", "pdfio/" PDFIO_VERSION);
-
-  if ((pdf->info_obj = pdfioFileCreateObj(pdf, info_dict)) == NULL)
-    goto error;
-
-  // Create random file ID values...
-  _pdfioCryptoMakeRandom(id_value, sizeof(id_value));
-
-  if ((pdf->id_array = pdfioArrayCreate(pdf)) != NULL)
-  {
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
-  }
-
-  return (pdf);
-
-  // Common error handling code...
-  error:
-
-  pdfioFileClose(pdf);
-
-  return (NULL);
+  return (create_common("output.pdf", /*fd*/-1, output_cb, output_cbdata, version, media_box, crop_box, error_cb, error_cbdata));
 }
 
 
@@ -705,13 +545,11 @@ pdfioFileCreateTemporary(
     pdfio_rect_t     *media_box,	// I - Default MediaBox for pages
     pdfio_rect_t     *crop_box,		// I - Default CropBox for pages
     pdfio_error_cb_t error_cb,		// I - Error callback or `NULL` for default
-    void             *error_data)	// I - Error callback data, if any
+    void             *error_cbdata)	// I - Error callback data, if any
 {
   pdfio_file_t	*pdf;			// PDF file
-  pdfio_dict_t	*dict;			// Dictionary for pages object
-  pdfio_dict_t	*info_dict;		// Dictionary for information object
-  unsigned char	id_value[16];		// File ID value
-  int		i;			// Looping var
+  int		i,			// Looping var
+		fd;			// File descriptor
   const char	*tmpdir;		// Temporary directory
 #if _WIN32 || defined(__APPLE__)
   char		tmppath[256];		// Temporary directory path
@@ -719,6 +557,8 @@ pdfioFileCreateTemporary(
   unsigned	tmpnum;			// Temporary filename number
 
 
+  PDFIO_DEBUG("pdfioFileCreate(buffer=%p, bufsize=%lu, version=\"%s\", media_box=%p, crop_box=%p, error_cb=%p, error_cbdata=%p)\n", (void *)buffer, (unsigned long)bufsize, version, (void *)media_box, (void *)crop_box, (void *)error_cb, (void *)error_cbdata);
+
   // Range check input...
   if (!buffer || bufsize < 32)
   {
@@ -727,31 +567,7 @@ pdfioFileCreateTemporary(
     return (NULL);
   }
 
-  if (!version)
-    version = "2.0";
-
-  if (!error_cb)
-  {
-    error_cb   = _pdfioFileDefaultError;
-    error_data = NULL;
-  }
-
-  // Allocate a PDF file structure...
-  if ((pdf = (pdfio_file_t *)calloc(1, sizeof(pdfio_file_t))) == NULL)
-  {
-    pdfio_file_t temp;			// Dummy file
-    char	message[8192];		// Message string
-
-    temp.filename = (char *)"temporary.pdf";
-    snprintf(message, sizeof(message), "Unable to allocate memory for PDF file - %s", strerror(errno));
-    (error_cb)(&temp, message, error_data);
-
-    *buffer = '\0';
-
-    return (NULL);
-  }
-
-  // Create the file...
+  // Create the temporary PDF file...
 #if _WIN32
   if ((tmpdir = getenv("TEMP")) == NULL)
   {
@@ -779,98 +595,35 @@ pdfioFileCreateTemporary(
     tmpdir = "/tmp";
 #endif // _WIN32
 
-  for (i = 0; i < 1000; i ++)
+  for (i = 0, fd = -1; i < 1000; i ++)
   {
     _pdfioCryptoMakeRandom((uint8_t *)&tmpnum, sizeof(tmpnum));
     snprintf(buffer, bufsize, "%s/%08x.pdf", tmpdir, tmpnum);
-    if ((pdf->fd = open(buffer, O_WRONLY | O_BINARY | O_CREAT | O_TRUNC | O_EXCL, 0666)) >= 0)
+    if ((fd = open(buffer, O_WRONLY | O_BINARY | O_CREAT | O_TRUNC | O_EXCL, 0666)) >= 0)
       break;
   }
 
-  pdf->filename = strdup(buffer);
-
-  if (i >= 1000)
+  if (fd < 0)
   {
-    _pdfioFileError(pdf, "Unable to create file - %s", strerror(errno));
-    free(pdf->filename);
-    free(pdf);
-    *buffer = '\0';
+    pdfio_file_t temp;			// Dummy file
+    char	message[8192];		// Message string
+
+    temp.filename = (char *)"<temporary>";
+    snprintf(message, sizeof(message), "Unable to create temporary PDF file: %s", strerror(errno));
+    (error_cb)(&temp, message, error_cbdata);
+
     return (NULL);
   }
 
-  pdf->version     = strdup(version);
-  pdf->mode        = _PDFIO_MODE_WRITE;
-  pdf->error_cb    = error_cb;
-  pdf->error_data  = error_data;
-  pdf->permissions = PDFIO_PERMISSION_ALL;
-  pdf->bufptr      = pdf->buffer;
-  pdf->bufend      = pdf->buffer + sizeof(pdf->buffer);
-
-  if (media_box)
+  if ((pdf = create_common(buffer, fd, /*output_cb*/NULL, /*output_cbdata*/NULL, version, media_box, crop_box, error_cb, error_cbdata)) == NULL)
   {
-    pdf->media_box = *media_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->media_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->media_box.y2 = 11.0f * 72.0f;
-  }
-
-  if (crop_box)
-  {
-    pdf->crop_box = *crop_box;
-  }
-  else
-  {
-    // Default to "universal" size (intersection of A4 and US Letter)
-    pdf->crop_box.x2 = 210.0 * 72.0f / 25.4f;
-    pdf->crop_box.y2 = 11.0f * 72.0f;
-  }
-
-  // Write a standard PDF header...
-  if (!_pdfioFilePrintf(pdf, "%%PDF-%s\n%%\342\343\317\323\n", version))
-    goto error;
-
-  // Create the pages object...
-  if ((dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetName(dict, "Type", "Pages");
-
-  if ((pdf->pages_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
-    goto error;
-
-  // Create the info object...
-  if ((info_dict = pdfioDictCreate(pdf)) == NULL)
-    goto error;
-
-  pdfioDictSetDate(info_dict, "CreationDate", time(NULL));
-  pdfioDictSetString(info_dict, "Producer", "pdfio/" PDFIO_VERSION);
-
-  if ((pdf->info_obj = pdfioFileCreateObj(pdf, info_dict)) == NULL)
-    goto error;
-
-  // Create random file ID values...
-  _pdfioCryptoMakeRandom(id_value, sizeof(id_value));
-
-  if ((pdf->id_array = pdfioArrayCreate(pdf)) != NULL)
-  {
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
-    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
+    // Remove the temporary file if we can't create the PDF file object...
+    close(fd);
+    unlink(buffer);
+    *buffer = '\0';
   }
 
   return (pdf);
-
-  // Common error handling code...
-  error:
-
-  pdfioFileClose(pdf);
-
-  unlink(buffer);
-  *buffer = '\0';
-
-  return (NULL);
 }
 
 
@@ -930,11 +683,12 @@ pdfioFileFindObj(
   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]);
+  PDFIO_DEBUG("pdfioFileFindObj: objs[current=%lu]=%p(%lu)\n", (unsigned long)current, (void *)pdf->objs[current], (unsigned long)(pdf->objs[current] ? pdf->objs[current]->number : 0));
 
   if (number == pdf->objs[current]->number)
   {
     // Fast match...
+    PDFIO_DEBUG("pdfioFileFindObj: Returning %lu (%p)\n", (unsigned long)current, pdf->objs[current]);
     return (pdf->objs[current]);
   }
   else if (number < pdf->objs[current]->number)
@@ -961,11 +715,20 @@ pdfioFileFindObj(
   }
 
   if (number == pdf->objs[left]->number)
+  {
+    PDFIO_DEBUG("pdfioFileFindObj: Returning %lu (%p)\n", (unsigned long)left, pdf->objs[left]);
     return (pdf->objs[left]);
+  }
   else if (number == pdf->objs[right]->number)
+  {
+    PDFIO_DEBUG("pdfioFileFindObj: Returning %lu (%p)\n", (unsigned long)right, pdf->objs[right]);
     return (pdf->objs[right]);
+  }
   else
+  {
+    PDFIO_DEBUG("pdfioFileFindObj: Returning NULL\n");
     return (NULL);
+  }
 }
 
 
@@ -980,6 +743,19 @@ pdfioFileGetAuthor(pdfio_file_t *pdf)	// I - PDF file
 }
 
 
+//
+// 'pdfioFileGetCatalog()' - Get the document catalog dictionary.
+//
+// @since PDFio 1.3@
+//
+
+pdfio_dict_t *				// O - Catalog dictionary
+pdfioFileGetCatalog(pdfio_file_t *pdf)	// I - PDF file
+{
+  return (pdf ? pdfioObjGetDict(pdf->root_obj) : NULL);
+}
+
+
 //
 // 'pdfioFileGetCreationDate()' - Get the creation date for a PDF file.
 //
@@ -1169,13 +945,13 @@ pdfioFileGetVersion(
 // This function opens an existing PDF file.  The "filename" argument specifies
 // the name of the PDF file to create.
 //
-// The "password_cb" and "password_data" arguments specify a password callback
+// The "password_cb" and "password_cbdata" arguments specify a password callback
 // and its data pointer for PDF files that use one of the standard Adobe
 // "security" handlers.  The callback returns a password string or `NULL` to
 // cancel the open.  If `NULL` is specified for the callback function and the
 // PDF file requires a password, the open will always fail.
 //
-// The "error_cb" and "error_data" arguments specify an error handler callback
+// The "error_cb" and "error_cbdata" arguments specify an error handler callback
 // and its data pointer - if `NULL` the default error handler is used that
 // writes error messages to `stderr`.
 //
@@ -1184,9 +960,10 @@ pdfio_file_t *				// O - PDF file
 pdfioFileOpen(
     const char          *filename,	// I - Filename
     pdfio_password_cb_t password_cb,	// I - Password callback or `NULL` for none
-    void                *password_data,	// I - Password callback data, if any
+    void                *password_cbdata,
+					// I - Password callback data, if any
     pdfio_error_cb_t    error_cb,	// I - Error callback or `NULL` for default
-    void                *error_data)	// I - Error callback data, if any
+    void                *error_cbdata)	// I - Error callback data, if any
 {
   pdfio_file_t	*pdf;			// PDF file
   char		line[1025],		// Line from file
@@ -1196,14 +973,16 @@ pdfioFileOpen(
   off_t		xref_offset;		// Offset to xref table
 
 
+  PDFIO_DEBUG("pdfioFileOpen(filename=\"%s\", password_cb=%p, password_cbdata=%p, error_cb=%p, error_cbdata=%p)\n", filename, (void *)password_cb, (void *)password_cbdata, (void *)error_cb, (void *)error_cbdata);
+
   // Range check input...
   if (!filename)
     return (NULL);
 
   if (!error_cb)
   {
-    error_cb   = _pdfioFileDefaultError;
-    error_data = NULL;
+    error_cb     = _pdfioFileDefaultError;
+    error_cbdata = NULL;
   }
 
   // Allocate a PDF file structure...
@@ -1214,14 +993,15 @@ pdfioFileOpen(
 
     temp.filename = (char *)filename;
     snprintf(message, sizeof(message), "Unable to allocate memory for PDF file - %s", strerror(errno));
-    (error_cb)(&temp, message, error_data);
+    (error_cb)(&temp, message, error_cbdata);
     return (NULL);
   }
 
+  pdf->loc         = get_lconv();
   pdf->filename    = strdup(filename);
   pdf->mode        = _PDFIO_MODE_READ;
   pdf->error_cb    = error_cb;
-  pdf->error_data  = error_data;
+  pdf->error_data  = error_cbdata;
   pdf->permissions = PDFIO_PERMISSION_ALL;
 
   // Open the file...
@@ -1277,7 +1057,7 @@ pdfioFileOpen(
 
   xref_offset = (off_t)strtol(ptr + 9, NULL, 10);
 
-  if (!load_xref(pdf, xref_offset, password_cb, password_data))
+  if (!load_xref(pdf, xref_offset, password_cb, password_cbdata))
     goto error;
 
   return (pdf);
@@ -1369,7 +1149,7 @@ pdfioFileSetPermissions(
   if (!pdf)
     return (false);
 
-  if (pdf->num_objs > 2)		// First two objects are pages and info
+  if (pdf->num_objs > 3)		// First three objects are pages, info, and root
   {
     _pdfioFileError(pdf, "You must call pdfioFileSetPermissions before adding any objects.");
     return (false);
@@ -1531,6 +1311,144 @@ compare_objmaps(_pdfio_objmap_t *a,	// I - First object map
 }
 
 
+//
+// 'create_common()' - Allocate and initialize a pdfio_file_t object for writing.
+//
+
+static pdfio_file_t *			// O - New PDF file
+create_common(
+    const char        *filename,	// I - Filename
+    int               fd,		// I - File descriptor, if any
+    pdfio_output_cb_t output_cb,	// I - Output callback function, if any
+    void              *output_cbdata,	// I - Output callback data, if any
+    const char        *version,		// I - PDF version
+    pdfio_rect_t      *media_box,	// I - Media box or `NULL` for default
+    pdfio_rect_t      *crop_box,	// I - Crop box of `NULL` for default
+    pdfio_error_cb_t  error_cb,		// I - Error callback function
+    void              *error_cbdata)	// I - Error callback data
+{
+  pdfio_file_t	*pdf;			// New PDF file
+  pdfio_dict_t	*dict;			// Dictionary
+  unsigned char	id_value[16];		// File ID value
+
+
+  PDFIO_DEBUG("create_common(filename=\"%s\", fd=%d, output_cb=%p, output_cbdata=%p, version=\"%s\", media_box=%p, crop_box=%p, error_cb=%p, error_cbdata=%p)\n", filename, fd, (void *)output_cb, (void *)output_cbdata, version, (void *)media_box, (void *)crop_box, (void *)error_cb, (void *)error_cbdata);
+
+  // Range check input...
+  if (!filename || (fd < 0 && !output_cb))
+    return (NULL);
+
+  if (!version)
+    version = "2.0";
+
+  if (!error_cb)
+  {
+    error_cb     = _pdfioFileDefaultError;
+    error_cbdata = NULL;
+  }
+
+  // Allocate a PDF file structure...
+  if ((pdf = (pdfio_file_t *)calloc(1, sizeof(pdfio_file_t))) == NULL)
+  {
+    pdfio_file_t temp;			// Dummy file
+    char	message[8192];		// Message string
+
+    temp.filename = (char *)filename;
+    snprintf(message, sizeof(message), "Unable to allocate memory for PDF file: %s", strerror(errno));
+    (error_cb)(&temp, message, error_cbdata);
+
+    return (NULL);
+  }
+
+  // Initialize PDF object...
+  pdf->loc         = get_lconv();
+  pdf->fd          = fd;
+  pdf->output_cb   = output_cb;
+  pdf->output_ctx  = output_cbdata;
+  pdf->filename    = strdup(filename);
+  pdf->version     = strdup(version);
+  pdf->mode        = _PDFIO_MODE_WRITE;
+  pdf->error_cb    = error_cb;
+  pdf->error_data  = error_cbdata;
+  pdf->permissions = PDFIO_PERMISSION_ALL;
+  pdf->bufptr      = pdf->buffer;
+  pdf->bufend      = pdf->buffer + sizeof(pdf->buffer);
+
+  if (media_box)
+  {
+    pdf->media_box = *media_box;
+  }
+  else
+  {
+    // Default to "universal" size (intersection of A4 and US Letter)
+    pdf->media_box.x2 = 210.0 * 72.0f / 25.4f;
+    pdf->media_box.y2 = 11.0f * 72.0f;
+  }
+
+  if (crop_box)
+  {
+    pdf->crop_box = *crop_box;
+  }
+  else
+  {
+    // Default to "universal" size (intersection of A4 and US Letter)
+    pdf->crop_box.x2 = 210.0 * 72.0f / 25.4f;
+    pdf->crop_box.y2 = 11.0f * 72.0f;
+  }
+
+  // Write a standard PDF header...
+  if (!_pdfioFilePrintf(pdf, "%%PDF-%s\n%%\342\343\317\323\n", version))
+    goto error;
+
+  // Create the pages object...
+  if ((dict = pdfioDictCreate(pdf)) == NULL)
+    goto error;
+
+  pdfioDictSetName(dict, "Type", "Pages");
+
+  if ((pdf->pages_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
+    goto error;
+
+  // Create the info object...
+  if ((dict = pdfioDictCreate(pdf)) == NULL)
+    goto error;
+
+  pdfioDictSetDate(dict, "CreationDate", time(NULL));
+  pdfioDictSetString(dict, "Producer", "pdfio/" PDFIO_VERSION);
+
+  if ((pdf->info_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
+    goto error;
+
+  // Create the root object...
+  if ((dict = pdfioDictCreate(pdf)) == NULL)
+    goto error;
+
+  pdfioDictSetName(dict, "Type", "Catalog");
+  pdfioDictSetObj(dict, "Pages", pdf->pages_obj);
+
+  if ((pdf->root_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
+    goto error;
+
+  // Create random file ID values...
+  _pdfioCryptoMakeRandom(id_value, sizeof(id_value));
+
+  if ((pdf->id_array = pdfioArrayCreate(pdf)) != NULL)
+  {
+    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
+    pdfioArrayAppendBinary(pdf->id_array, id_value, sizeof(id_value));
+  }
+
+  return (pdf);
+
+  // Common error handling code...
+  error:
+
+  pdfioFileClose(pdf);
+
+  return (NULL);
+}
+
+
 //
 // 'get_info_string()' - Get a string value from the Info dictionary.
 //
@@ -1543,36 +1461,35 @@ get_info_string(pdfio_file_t *pdf,	// I - PDF file
                 const char   *key)	// I - Dictionary key
 {
   pdfio_dict_t	*dict;			// Info dictionary
-  _pdfio_value_t *value;		// Value
+
 
   // Range check input...
-  if (!pdf || !pdf->info_obj || (dict = pdfioObjGetDict(pdf->info_obj)) == NULL || (value = _pdfioDictGetValue(dict, key)) == NULL)
+  if (!pdf || !pdf->info_obj || (dict = pdfioObjGetDict(pdf->info_obj)) == NULL)
     return (NULL);
-
-  // If we already have a value, return it...
-  if (value->type == PDFIO_VALTYPE_NAME || value->type == PDFIO_VALTYPE_STRING)
-  {
-    return (value->value.string);
-  }
-  else if (value->type == PDFIO_VALTYPE_BINARY && value->value.binary.datalen < 4096)
-  {
-    // Convert binary string to regular string...
-    char	temp[4096];		// Temporary string
-
-    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;
-    value->value.string = pdfioStringCreate(pdf, temp);
-
-    return (value->value.string);
-  }
   else
+    return (pdfioDictGetString(dict, key));
+}
+
+
+//
+// 'get_lconv()' - Get any locale-specific numeric information.
+//
+
+static struct lconv *			// O - Locale information or `NULL`
+get_lconv(void)
+{
+  struct lconv	*loc;			// Locale information
+
+
+  if ((loc = localeconv()) != NULL)
   {
-    // Something else that is not a string...
-    return (NULL);
+    PDFIO_DEBUG("get_lconv: loc=%p, loc->decimal_point=\"%s\"\n", loc, loc->decimal_point);
+
+    if (!loc->decimal_point || !strcmp(loc->decimal_point, "."))
+      loc = NULL;
   }
+
+  return (loc);
 }
 
 
@@ -2060,7 +1977,7 @@ load_xref(
       {
         PDFIO_DEBUG("load_xref: '%s' at offset %lu\n", line, (unsigned long)trailer_offset);
 
-	if (!strncmp(line, "trailer", 7) && (!line[7] || isspace(line[7] & 255)))
+	if (!strncmp(line, "trailer", 7) && (!line[7] || isspace(line[7] & 255) || line[7] == '<'))
 	{
 	  if (line[7])
 	  {
@@ -2218,30 +2135,6 @@ load_xref(
 }
 
 
-//
-// 'write_catalog()' - Write the PDF root object/catalog.
-//
-
-static bool				// O - `true` on success, `false` on failure
-write_catalog(pdfio_file_t *pdf)	// I - PDF file
-{
-  pdfio_dict_t	*dict;			// Dictionary for catalog...
-
-
-  if ((dict = pdfioDictCreate(pdf)) == NULL)
-    return (false);
-
-  pdfioDictSetName(dict, "Type", "Catalog");
-  pdfioDictSetObj(dict, "Pages", pdf->pages_obj);
-  // TODO: Add support for all of the root object dictionary keys
-
-  if ((pdf->root_obj = pdfioFileCreateObj(pdf, dict)) == NULL)
-    return (false);
-  else
-    return (pdfioObjClose(pdf->root_obj));
-}
-
-
 //
 // 'write_pages()' - Write the PDF pages objects.
 //

pdfio-object.c

@@ -1,7 +1,7 @@
 //
 // PDF object functions for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -99,7 +99,7 @@ pdfioObjCopy(pdfio_file_t *pdf,		// I - PDF file
     return (NULL);
 
   if (dstobj->value.type == PDFIO_VALTYPE_DICT)
-    _pdfioDictClear(dstobj->value.value.dict, "Length");
+    pdfioDictClear(dstobj->value.value.dict, "Length");
 
   if (srcobj->stream_offset)
   {
@@ -333,6 +333,26 @@ pdfioObjGetLength(pdfio_obj_t *obj)	// I - Object
 }
 
 
+//
+// 'pdfioObjGetName()' - Get the name value associated with an object.
+//
+
+const char *				// O - Dictionary or `NULL` on error
+pdfioObjGetName(pdfio_obj_t *obj)	// I - Object
+{
+  if (!obj)
+    return (NULL);
+
+  if (obj->value.type == PDFIO_VALTYPE_NONE)
+    _pdfioObjLoad(obj);
+
+  if (obj->value.type == PDFIO_VALTYPE_NAME)
+    return (obj->value.value.name);
+  else
+    return (NULL);
+}
+
+
 //
 // 'pdfioObjGetNumber()' - Get the object's number.
 //
@@ -347,8 +367,21 @@ pdfioObjGetNumber(pdfio_obj_t *obj)	// I - Object
 //
 // 'pdfioObjGetSubtype()' - Get an object's subtype.
 //
+// This function returns an object's PDF subtype name, if any.  Common subtype
+// names include:
+//
+// - "CIDFontType0": A CID Type0 font
+// - "CIDFontType2": A CID TrueType font
+// - "Image": An image or image mask
+// - "Form": A fillable form
+// - "OpenType": An OpenType font
+// - "Type0": A composite font
+// - "Type1": A PostScript Type1 font
+// - "Type3": A PDF Type3 font
+// - "TrueType": A TrueType font
+//
 
-const char *				// O - Object subtype
+const char *				// O - Object subtype name or `NULL` for none
 pdfioObjGetSubtype(pdfio_obj_t *obj)	// I - Object
 {
   pdfio_dict_t	*dict;			// Object dictionary
@@ -364,8 +397,21 @@ pdfioObjGetSubtype(pdfio_obj_t *obj)	// I - Object
 //
 // 'pdfioObjGetType()' - Get an object's type.
 //
+// This function returns an object's PDF type name, if any. Common type names
+// include:
+//
+// - "CMap": A character map for composite fonts
+// - "Font": An embedded font (@link pdfioObjGetSubtype@ will tell you the
+//   font format)
+// - "FontDescriptor": A font descriptor
+// - "Page": A (visible) page
+// - "Pages": A page tree node
+// - "Template": An invisible template page
+// - "XObject": An image, image mask, or form (@link pdfioObjGetSubtype@ will
+//   tell you which)
+//
 
-const char *				// O - Object type
+const char *				// O - Object type name or `NULL` for none
 pdfioObjGetType(pdfio_obj_t *obj)	// I - Object
 {
   pdfio_dict_t	*dict;			// Object dictionary

pdfio-private.h

@@ -1,7 +1,7 @@
 //
 // Private header file for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -20,6 +20,7 @@
 #  include <errno.h>
 #  include <inttypes.h>
 #  include <fcntl.h>
+#  include <locale.h>
 #  ifdef _WIN32
 #    include <io.h>
 #    include <direct.h>
@@ -37,9 +38,11 @@
 #    define unlink	_unlink
 #    define vsnprintf	_vsnprintf
 #    define write	_write
-#    define F_OK	00		// POSIX parameters/flags
-#    define W_OK	02
-#    define R_OK	04
+#    ifndef F_OK
+#      define F_OK	00		// POSIX parameters/flags
+#      define W_OK	02
+#      define R_OK	04
+#    endif // !F_OK
 #    define O_RDONLY	_O_RDONLY	// Map standard POSIX open flags
 #    define O_WRONLY	_O_WRONLY
 #    define O_CREAT	_O_CREAT
@@ -224,6 +227,7 @@ typedef struct _pdfio_objmap_s		// PDF object map
 struct _pdfio_file_s			// PDF file structure
 {
   char		*filename;		// Filename
+  struct lconv	*loc;			// Locale data
   char		*version;		// Version number
   pdfio_rect_t	media_box,		// Default MediaBox value
 		crop_box;		// Default CropBox value
@@ -322,6 +326,9 @@ struct _pdfio_stream_s			// Stream
 // Functions...
 //
 
+extern double		_pdfio_strtod(pdfio_file_t *pdf, const char *s) _PDFIO_INTERNAL;
+extern ssize_t		_pdfio_vsnprintf(pdfio_file_t *pdf, char *buffer, size_t bufsize, const char *format, va_list ap) _PDFIO_INTERNAL;
+
 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;
@@ -346,7 +353,6 @@ extern void		_pdfioCryptoSHA256Init(_pdfio_sha256_t *ctx) _PDFIO_INTERNAL;
 extern void		_pdfioCryptoSHA256Finish(_pdfio_sha256_t *ctx, uint8_t *Message_Digest) _PDFIO_INTERNAL;
 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;

pdfio-stream.c

@@ -1,7 +1,7 @@
 //
 // PDF stream functions for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -408,6 +408,7 @@ _pdfioStreamOpen(pdfio_obj_t *obj,	// I - Object
   pdfio_stream_t	*st;		// Stream
   pdfio_dict_t		*dict = pdfioObjGetDict(obj);
 					// Object dictionary
+  const char		*type;		// Object type
 
 
   PDFIO_DEBUG("_pdfioStreamOpen(obj=%p(%u), decode=%s)\n", obj, (unsigned)obj->number, decode ? "true" : "false");
@@ -434,7 +435,9 @@ _pdfioStreamOpen(pdfio_obj_t *obj,	// I - Object
     return (NULL);
   }
 
-  if (obj->pdf->encryption)
+  type = pdfioObjGetType(obj);
+
+  if (obj->pdf->encryption && (!type || strcmp(type, "XRef")))
   {
     uint8_t	iv[64];			// Initialization vector
     size_t	ivlen;			// Length of initialization vector, if any
@@ -687,7 +690,7 @@ pdfioStreamPrintf(
 
   // Format the string...
   va_start(ap, format);
-  vsnprintf(buffer, sizeof(buffer), format, ap);
+  _pdfio_vsnprintf(st->pdf, buffer, sizeof(buffer), format, ap);
   va_end(ap);
 
   // Write the string...
@@ -1061,19 +1064,11 @@ stream_read(pdfio_stream_t *st,		// I - Stream
       st->flate.next_out  = (Bytef *)buffer;
       st->flate.avail_out = (uInt)bytes;
 
-      avail_in  = st->flate.avail_in;
-      avail_out = st->flate.avail_out;
-
       if ((status = inflate(&(st->flate), Z_NO_FLUSH)) < Z_OK)
       {
 	_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)
-      {
-	_pdfioFileError(st->pdf, "Corrupt stream data.");
-	return (-1);
-      }
 
       return (st->flate.next_out - (Bytef *)buffer);
     }

pdfio-string.c

@@ -1,7 +1,7 @@
 //
-// PDF dictionary functions for PDFio.
+// PDF string functions for PDFio.
 //
-// Copyright © 2021 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -14,7 +14,360 @@
 // Local functions...
 //
 
-static int	compare_strings(char **a, char **b);
+static size_t	find_string(pdfio_file_t *pdf, const char *s, int *rdiff);
+
+
+//
+// '_pdfio_strtod()' - Convert a string to a double value.
+//
+// This function wraps strtod() to avoid locale issues.
+//
+
+double					// O - Double value
+_pdfio_strtod(pdfio_file_t *pdf,	// I - PDF file
+              const char   *s)		// I - String
+{
+  char	temp[64],			// Temporary buffer
+	*tempptr;			// Pointer into temporary buffer
+
+
+  // See if the locale has a special decimal point string...
+  if (!pdf->loc)
+    return (strtod(s, NULL));
+
+  // Copy leading sign, numbers, period, and then numbers...
+  tempptr                = temp;
+  temp[sizeof(temp) - 1] = '\0';
+
+  while (*s && *s != '.')
+  {
+    if (tempptr < (temp + sizeof(temp) - 1))
+      *tempptr++ = *s++;
+    else
+      return (0.0);
+  }
+
+  if (*s == '.')
+  {
+    // Convert decimal point to locale equivalent...
+    size_t	declen = strlen(pdf->loc->decimal_point);
+					// Length of decimal point
+    s ++;
+
+    if (declen <= (sizeof(temp) - (size_t)(tempptr - temp)))
+    {
+      memcpy(tempptr, pdf->loc->decimal_point, declen);
+      tempptr += declen;
+    }
+    else
+    {
+      return (0.0);
+    }
+  }
+
+  // Copy any remaining characters...
+  while (*s)
+  {
+    if (tempptr < (temp + sizeof(temp) - 1))
+      *tempptr++ = *s++;
+    else
+      return (0.0);
+  }
+
+  // Nul-terminate the temporary string and convert the string...
+  *tempptr = '\0';
+
+  return (strtod(temp, NULL));
+}
+
+
+//
+// '_pdfio_vsnprintf()' - Format a string.
+//
+// This function emulates vsnprintf() to avoid locale issues.
+//
+
+ssize_t					// O - Number of bytes
+_pdfio_vsnprintf(pdfio_file_t *pdf,	// I - PDF file
+                 char         *buffer,	// I - Output buffer
+                 size_t       bufsize,	// I - Size of output buffer
+                 const char   *format,	// I - printf-style format string
+                 va_list      ap)	// I - Pointer to additional arguments
+{
+  char		*bufptr,		// Pointer to position in buffer
+		*bufend,		// Pointer to end of buffer
+		size,			// Size character (h, l, L)
+		type;			// Format type character
+  int		width,			// Width of field
+		prec;			// Number of characters of precision
+  char		tformat[100],		// Temporary format string for snprintf()
+		*tptr,			// Pointer into temporary format
+		temp[1024],		// Buffer for formatted numbers
+		*tempptr;		// Pointer into buffer
+  char		*s;			// Pointer to string
+  ssize_t	bytes;			// Total number of bytes needed
+  const char	*dec = pdf->loc ? pdf->loc->decimal_point : ".";
+					// Decimal point string
+  char		*decptr;		// Pointer to decimal point
+
+
+  // Loop through the format string, formatting as needed...
+  bufptr  = buffer;
+  bufend  = buffer + bufsize - 1;
+  *bufend = '\0';
+  bytes   = 0;
+
+  while (*format)
+  {
+    if (*format == '%')
+    {
+      // Format character...
+      tptr = tformat;
+      *tptr++ = *format++;
+
+      if (*format == '%')
+      {
+        if (bufptr < bufend)
+	  *bufptr++ = *format;
+        bytes ++;
+        format ++;
+	continue;
+      }
+      else if (strchr(" -+#\'", *format))
+      {
+        *tptr++ = *format++;
+      }
+
+      if (*format == '*')
+      {
+        // Get width from argument...
+	format ++;
+	width = va_arg(ap, int);
+
+	snprintf(tptr, sizeof(tformat) - (size_t)(tptr - tformat), "%d", width);
+	tptr += strlen(tptr);
+      }
+      else
+      {
+	width = 0;
+
+	while (isdigit(*format & 255))
+	{
+	  if (tptr < (tformat + sizeof(tformat) - 1))
+	    *tptr++ = *format;
+
+	  width = width * 10 + *format++ - '0';
+	}
+      }
+
+      if (*format == '.')
+      {
+	if (tptr < (tformat + sizeof(tformat) - 1))
+	  *tptr++ = *format;
+
+        format ++;
+
+        if (*format == '*')
+	{
+          // Get precision from argument...
+	  format ++;
+	  prec = va_arg(ap, int);
+
+	  snprintf(tptr, sizeof(tformat) - (size_t)(tptr - tformat), "%d", prec);
+	  tptr += strlen(tptr);
+	}
+	else
+	{
+	  prec = 0;
+
+	  while (isdigit(*format & 255))
+	  {
+	    if (tptr < (tformat + sizeof(tformat) - 1))
+	      *tptr++ = *format;
+
+	    prec = prec * 10 + *format++ - '0';
+	  }
+	}
+      }
+
+      if (*format == 'l' && format[1] == 'l')
+      {
+        size = 'L';
+
+	if (tptr < (tformat + sizeof(tformat) - 2))
+	{
+	  *tptr++ = 'l';
+	  *tptr++ = 'l';
+	}
+
+	format += 2;
+      }
+      else if (*format == 'h' || *format == 'l' || *format == 'L')
+      {
+	if (tptr < (tformat + sizeof(tformat) - 1))
+	  *tptr++ = *format;
+
+        size = *format++;
+      }
+      else
+      {
+        size = 0;
+      }
+
+      if (!*format)
+        break;
+
+      if (tptr < (tformat + sizeof(tformat) - 1))
+        *tptr++ = *format;
+
+      type  = *format++;
+      *tptr = '\0';
+
+      switch (type)
+      {
+	case 'E' : // Floating point formats
+	case 'G' :
+	case 'e' :
+	case 'f' :
+	case 'g' :
+	    if ((size_t)(width + 2) > sizeof(temp))
+	      break;
+
+	    snprintf(temp, sizeof(temp), tformat, va_arg(ap, double));
+
+	    if ((decptr = strstr(temp, dec)) != NULL)
+	    {
+	      // Convert locale decimal point to "."
+	      PDFIO_DEBUG("_pdfio_vsnprintf: Before \"%s\"\n", temp);
+	      tempptr = decptr + strlen(dec);
+	      if (tempptr > (decptr + 1))
+	        memmove(decptr + 1, tempptr, strlen(tempptr) + 1);
+	      *decptr = '.';
+
+	      // Strip trailing 0's...
+	      for (tempptr = temp + strlen(temp) - 1; tempptr > temp && *tempptr == '0'; tempptr --)
+		*tempptr = '\0';
+
+	      if (*tempptr == '.')
+		*tempptr = '\0';	// Strip trailing decimal point
+
+	      PDFIO_DEBUG("_pdfio_vsnprintf: After \"%s\"\n", temp);
+	    }
+
+            // Copy to the output buffer
+            bytes += (int)strlen(temp);
+
+            if (bufptr < bufend)
+	    {
+	      strncpy(bufptr, temp, (size_t)(bufend - bufptr - 1));
+	      bufptr += strlen(bufptr);
+	    }
+	    break;
+
+        case 'B' : // Integer formats
+	case 'X' :
+	case 'b' :
+        case 'd' :
+	case 'i' :
+	case 'o' :
+	case 'u' :
+	case 'x' :
+	    if ((size_t)(width + 2) > sizeof(temp))
+	      break;
+
+#  ifdef HAVE_LONG_LONG
+            if (size == 'L')
+	      snprintf(temp, sizeof(temp), tformat, va_arg(ap, long long));
+	    else
+#  endif // HAVE_LONG_LONG
+            if (size == 'l')
+	      snprintf(temp, sizeof(temp), tformat, va_arg(ap, long));
+	    else
+	      snprintf(temp, sizeof(temp), tformat, va_arg(ap, int));
+
+            bytes += (int)strlen(temp);
+
+	    if (bufptr < bufend)
+	    {
+	      strncpy(bufptr, temp, (size_t)(bufend - bufptr - 1));
+	      bufptr += strlen(bufptr);
+	    }
+	    break;
+
+	case 'p' : // Pointer value
+	    if ((size_t)(width + 2) > sizeof(temp))
+	      break;
+
+	    snprintf(temp, sizeof(temp), tformat, va_arg(ap, void *));
+
+            bytes += (int)strlen(temp);
+
+	    if (bufptr < bufend)
+	    {
+	      strncpy(bufptr, temp, (size_t)(bufend - bufptr - 1));
+	      bufptr += strlen(bufptr);
+	    }
+	    break;
+
+        case 'c' : // Character or character array
+	    bytes += width;
+
+	    if (bufptr < bufend)
+	    {
+	      if (width <= 1)
+	      {
+	        *bufptr++ = (char)va_arg(ap, int);
+	      }
+	      else
+	      {
+		if ((bufptr + width) > bufend)
+		  width = (int)(bufend - bufptr);
+
+		memcpy(bufptr, va_arg(ap, char *), (size_t)width);
+		bufptr += width;
+	      }
+	    }
+	    break;
+
+	case 's' : // String
+	    if ((s = va_arg(ap, char *)) == NULL)
+	      s = "(null)";
+
+            bytes += strlen(s);
+
+	    if (bufptr < bufend)
+	    {
+	      strncpy(bufptr, s, (size_t)(bufend - bufptr - 1));
+	      bufptr += strlen(bufptr);
+	    }
+	    break;
+
+	case 'n' : // Output number of chars so far
+	    *(va_arg(ap, int *)) = (int)bytes;
+	    break;
+      }
+    }
+    else
+    {
+      // Literal character...
+      bytes ++;
+
+      if (bufptr < bufend)
+        *bufptr++ = *format++;
+    }
+  }
+
+  // Nul-terminate the string and return the number of characters needed.
+  if (bufptr < bufend)
+  {
+    // Everything fit in the buffer...
+    *bufptr = '\0';
+  }
+
+  PDFIO_DEBUG("_pdfio_vsnprintf: Returning %ld \"%s\"\n", (long)bytes, buffer);
+
+  return (bytes);
+}
 
 
 //
@@ -32,8 +385,9 @@ pdfioStringCreate(
     pdfio_file_t *pdf,			// I - PDF file
     const char   *s)			// I - Nul-terminated string
 {
-  char	*news;				// New string
-  char	**match;			// Matching string
+  char		*news;			// New string
+  size_t	idx;			// Index into strings
+  int		diff;			// Different
 
 
   PDFIO_DEBUG("pdfioStringCreate(pdf=%p, s=\"%s\")\n", pdf, s);
@@ -43,8 +397,17 @@ pdfioStringCreate(
     return (NULL);
 
   // See if the string has already been added...
-  if (pdf->num_strings > 0 && (match = (char **)bsearch(&s, pdf->strings, pdf->num_strings, sizeof(char *), (int (*)(const void *, const void *))compare_strings)) != NULL)
-    return (*match);
+  if (pdf->num_strings > 0)
+  {
+    idx = find_string(pdf, s, &diff);
+    if (diff == 0)
+      return (pdf->strings[idx]);
+  }
+  else
+  {
+    idx  = 0;
+    diff = -1;
+  }
 
   // Not already added, so add it...
   if ((news = strdup(s)) == NULL)
@@ -65,11 +428,17 @@ pdfioStringCreate(
     pdf->alloc_strings += 128;
   }
 
-  // TODO: Change to insertion sort as needed...
-  pdf->strings[pdf->num_strings ++] = news;
+  // Insert the string...
+  if (diff > 0)
+    idx ++;
 
-  if (pdf->num_strings > 1)
-    qsort(pdf->strings, pdf->num_strings, sizeof(char *), (int (*)(const void *, const void *))compare_strings);
+  PDFIO_DEBUG("pdfioStringCreate: Inserting \"%s\" at %u\n", news, (unsigned)idx);
+
+  if (idx < pdf->num_strings)
+    memmove(pdf->strings + idx + 1, pdf->strings + idx, (pdf->num_strings - idx) * sizeof(char *));
+
+  pdf->strings[idx] = news;
+  pdf->num_strings ++;
 
   PDFIO_DEBUG("pdfioStringCreate: %lu strings\n", (unsigned long)pdf->num_strings);
 
@@ -120,17 +489,67 @@ _pdfioStringIsAllocated(
     pdfio_file_t *pdf,			// I - PDF file
     const char   *s)			// I - String
 {
-  return (pdf->num_strings > 0 && bsearch(&s, pdf->strings, pdf->num_strings, sizeof(char *), (int (*)(const void *, const void *))compare_strings) != NULL);
+  int	diff;				// Difference
+
+
+  if (pdf->num_strings == 0)
+    return (false);
+
+  find_string(pdf, s, &diff);
+
+  return (diff == 0);
 }
 
 
 //
-// 'compare_strings()' - Compare two strings.
+// 'find_string()' - Find an element in the array.
 //
 
-static int				// O - Result of comparison
-compare_strings(char **a,		// I - First string
-                char **b)		// I - Second string
+static size_t				// O - Index of match
+find_string(pdfio_file_t *pdf,		// I - PDF file
+	    const char   *s,		// I - String to find
+	    int          *rdiff)	// O - Difference of match
 {
-  return (strcmp(*a, *b));
+  size_t	left,			// Left side of search
+		right,			// Right side of search
+		current;		// Current element
+  int		diff;			// Comparison with current element
+
+
+  // Do a binary search for the string...
+  left  = 0;
+  right = pdf->num_strings - 1;
+
+  do
+  {
+    current = (left + right) / 2;
+    diff    = strcmp(s, pdf->strings[current]);
+
+    if (diff == 0)
+      break;
+    else if (diff < 0)
+      right = current;
+    else
+      left = current;
+  }
+  while ((right - left) > 1);
+
+  if (diff != 0)
+  {
+    // Check the last 1 or 2 elements...
+    if ((diff = strcmp(s, pdf->strings[left])) <= 0)
+    {
+      current = left;
+    }
+    else
+    {
+      diff    = strcmp(s, pdf->strings[right]);
+      current = right;
+    }
+  }
+
+  // Return the closest string and the difference...
+  *rdiff = diff;
+
+  return (current);
 }

pdfio-value.c

@@ -1,7 +1,7 @@
 //
 // PDF value functions for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -215,6 +215,9 @@ void
 _pdfioValueDebug(_pdfio_value_t *v,	// I - Value
 		 FILE           *fp)	// I - Output file
 {
+  if (!v)
+    return;
+
   switch (v->type)
   {
     case PDFIO_VALTYPE_ARRAY :
@@ -497,7 +500,7 @@ _pdfioValueRead(pdfio_file_t   *pdf,	// I - PDF file
 
     // If we get here, we have a number...
     v->type         = PDFIO_VALTYPE_NUMBER;
-    v->value.number = (double)strtod(token, NULL);
+    v->value.number = _pdfio_strtod(pdf, token);
   }
   else if (!strcmp(token, "true") || !strcmp(token, "false"))
   {

pdfio.h

@@ -1,7 +1,7 @@
 //
 // Public header file for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 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		"1.2.0"
+#  define PDFIO_VERSION		"1.4.0"
 
 
 //
@@ -151,7 +151,9 @@ extern pdfio_obj_t	*pdfioArrayGetObj(pdfio_array_t *a, size_t n) _PDFIO_PUBLIC;
 extern size_t		pdfioArrayGetSize(pdfio_array_t *a) _PDFIO_PUBLIC;
 extern const char	*pdfioArrayGetString(pdfio_array_t *a, size_t n) _PDFIO_PUBLIC;
 extern pdfio_valtype_t	pdfioArrayGetType(pdfio_array_t *a, size_t n) _PDFIO_PUBLIC;
+extern bool		pdfioArrayRemove(pdfio_array_t *a, size_t n) _PDFIO_PUBLIC;
 
+extern bool		pdfioDictClear(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
 extern pdfio_dict_t	*pdfioDictCopy(pdfio_file_t *pdf, pdfio_dict_t *dict) _PDFIO_PUBLIC;
 extern pdfio_dict_t	*pdfioDictCreate(pdfio_file_t *pdf) _PDFIO_PUBLIC;
 extern pdfio_array_t	*pdfioDictGetArray(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
@@ -159,7 +161,9 @@ extern unsigned char	*pdfioDictGetBinary(pdfio_dict_t *dict, const char *key, si
 extern bool		pdfioDictGetBoolean(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
 extern time_t		pdfioDictGetDate(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
 extern pdfio_dict_t	*pdfioDictGetDict(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
+extern const char	*pdfioDictGetKey(pdfio_dict_t *dict, size_t n) _PDFIO_PUBLIC;
 extern const char	*pdfioDictGetName(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
+extern size_t		pdfioDictGetNumPairs(pdfio_dict_t *dict) _PDFIO_PUBLIC;
 extern double		pdfioDictGetNumber(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
 extern pdfio_obj_t	*pdfioDictGetObj(pdfio_dict_t *dict, const char *key) _PDFIO_PUBLIC;
 extern pdfio_rect_t	*pdfioDictGetRect(pdfio_dict_t *dict, const char *key, pdfio_rect_t *rect) _PDFIO_PUBLIC;
@@ -182,6 +186,7 @@ extern bool		pdfioDictSetStringf(pdfio_dict_t *dict, const char *key, const char
 extern bool		pdfioFileClose(pdfio_file_t *pdf) _PDFIO_PUBLIC;
 extern pdfio_file_t	*pdfioFileCreate(const char *filename, const char *version, pdfio_rect_t *media_box, pdfio_rect_t *crop_box, pdfio_error_cb_t error_cb, void *error_data) _PDFIO_PUBLIC;
 extern pdfio_obj_t	*pdfioFileCreateArrayObj(pdfio_file_t *pdf, pdfio_array_t *array) _PDFIO_PUBLIC;
+extern pdfio_obj_t	*pdfioFileCreateNameObj(pdfio_file_t *pdf, const char *name) _PDFIO_PUBLIC;
 extern pdfio_obj_t	*pdfioFileCreateNumberObj(pdfio_file_t *pdf, double number) _PDFIO_PUBLIC;
 extern pdfio_obj_t	*pdfioFileCreateObj(pdfio_file_t *pdf, pdfio_dict_t *dict) _PDFIO_PUBLIC;
 extern pdfio_file_t	*pdfioFileCreateOutput(pdfio_output_cb_t output_cb, void *output_ctx, const char *version, pdfio_rect_t *media_box, pdfio_rect_t *crop_box, pdfio_error_cb_t error_cb, void *error_data) _PDFIO_PUBLIC;
@@ -191,6 +196,7 @@ extern pdfio_obj_t	*pdfioFileCreateStringObj(pdfio_file_t *pdf, const char *s) _
 extern pdfio_file_t	*pdfioFileCreateTemporary(char *buffer, size_t bufsize, const char *version, pdfio_rect_t *media_box, pdfio_rect_t *crop_box, pdfio_error_cb_t error_cb, void *error_data) _PDFIO_PUBLIC;
 extern pdfio_obj_t	*pdfioFileFindObj(pdfio_file_t *pdf, size_t number) _PDFIO_PUBLIC;
 extern const char	*pdfioFileGetAuthor(pdfio_file_t *pdf) _PDFIO_PUBLIC;
+extern pdfio_dict_t	*pdfioFileGetCatalog(pdfio_file_t *pdf) _PDFIO_PUBLIC;
 extern time_t		pdfioFileGetCreationDate(pdfio_file_t *pdf) _PDFIO_PUBLIC;
 extern const char	*pdfioFileGetCreator(pdfio_file_t *pdf) _PDFIO_PUBLIC;
 extern pdfio_array_t	*pdfioFileGetID(pdfio_file_t *pdf) _PDFIO_PUBLIC;
@@ -221,6 +227,7 @@ extern pdfio_array_t	*pdfioObjGetArray(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern pdfio_dict_t	*pdfioObjGetDict(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern unsigned short	pdfioObjGetGeneration(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern size_t		pdfioObjGetLength(pdfio_obj_t *obj) _PDFIO_PUBLIC;
+extern const char	*pdfioObjGetName(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern size_t		pdfioObjGetNumber(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern const char	*pdfioObjGetSubtype(pdfio_obj_t *obj) _PDFIO_PUBLIC;
 extern const char	*pdfioObjGetType(pdfio_obj_t *obj) _PDFIO_PUBLIC;

pdfio1.def

@@ -1,7 +1,8 @@
 LIBRARY pdfio1
-VERSION 1.2
+VERSION 1.4
 EXPORTS
 _pdfioArrayDebug
+_pdfioArrayDecrypt
 _pdfioArrayDelete
 _pdfioArrayGetValue
 _pdfioArrayRead
@@ -22,8 +23,8 @@ _pdfioCryptoSHA256Append
 _pdfioCryptoSHA256Finish
 _pdfioCryptoSHA256Init
 _pdfioCryptoUnlock
-_pdfioDictClear
 _pdfioDictDebug
+_pdfioDictDecrypt
 _pdfioDictDelete
 _pdfioDictGetValue
 _pdfioDictRead
@@ -61,9 +62,12 @@ _pdfioTokenPush
 _pdfioTokenRead
 _pdfioValueCopy
 _pdfioValueDebug
+_pdfioValueDecrypt
 _pdfioValueDelete
 _pdfioValueRead
 _pdfioValueWrite
+_pdfio_strtod
+_pdfio_vsnprintf
 pdfioArrayAppendArray
 pdfioArrayAppendBinary
 pdfioArrayAppendBoolean
@@ -91,6 +95,7 @@ pdfioArrayGetObj
 pdfioArrayGetSize
 pdfioArrayGetString
 pdfioArrayGetType
+pdfioArrayRemove
 pdfioContentClip
 pdfioContentDrawImage
 pdfioContentFill
@@ -148,6 +153,7 @@ pdfioContentTextNextLine
 pdfioContentTextShow
 pdfioContentTextShowJustified
 pdfioContentTextShowf
+pdfioDictClear
 pdfioDictCopy
 pdfioDictCreate
 pdfioDictGetArray
@@ -155,7 +161,9 @@ pdfioDictGetBinary
 pdfioDictGetBoolean
 pdfioDictGetDate
 pdfioDictGetDict
+pdfioDictGetKey
 pdfioDictGetName
+pdfioDictGetNumPairs
 pdfioDictGetNumber
 pdfioDictGetObj
 pdfioDictGetRect
@@ -182,6 +190,7 @@ pdfioFileCreateFontObjFromFile
 pdfioFileCreateICCObjFromFile
 pdfioFileCreateImageObjFromData
 pdfioFileCreateImageObjFromFile
+pdfioFileCreateNameObj
 pdfioFileCreateNumberObj
 pdfioFileCreateObj
 pdfioFileCreateOutput
@@ -190,6 +199,7 @@ pdfioFileCreateStringObj
 pdfioFileCreateTemporary
 pdfioFileFindObj
 pdfioFileGetAuthor
+pdfioFileGetCatalog
 pdfioFileGetCreationDate
 pdfioFileGetCreator
 pdfioFileGetID
@@ -222,6 +232,7 @@ pdfioObjGetArray
 pdfioObjGetDict
 pdfioObjGetGeneration
 pdfioObjGetLength
+pdfioObjGetName
 pdfioObjGetNumber
 pdfioObjGetSubtype
 pdfioObjGetType

pdfio_native.nuspec

@@ -3,7 +3,7 @@
     <metadata>
         <id>pdfio_native</id>
         <title>PDFio Library for VS2019+</title>
-        <version>1.2.0</version>
+        <version>1.3.2</version>
         <authors>Michael R Sweet</authors>
         <owners>michaelrsweet</owners>
         <projectUrl>https://github.com/michaelrsweet/pappl</projectUrl>
@@ -16,7 +16,7 @@
         <copyright>Copyright © 2019-2024 by Michael R Sweet</copyright>
         <tags>pdf file native</tags>
 	<dependencies>
-	    <dependency id="pdfio_native.redist" version="1.2.0" />
+	    <dependency id="pdfio_native.redist" version="1.3.2" />
 	    <dependency id="zlib_native.redist" version="1.2.11" />
 	</dependencies>
     </metadata>

pdfio_native.redist.nuspec

@@ -3,7 +3,7 @@
     <metadata>
         <id>pdfio_native.redist</id>
         <title>PDFio Library for VS2019+</title>
-        <version>1.2.0</version>
+        <version>1.3.2</version>
         <authors>Michael R Sweet</authors>
         <owners>michaelrsweet</owners>
         <projectUrl>https://github.com/michaelrsweet/pappl</projectUrl>

runtests.bat

@@ -0,0 +1,16 @@
+:: Script to run unit test program
+::
+:: Usage:
+::
+::   .\runtests.bat x64\{Debug|Release}
+::
+
+:: Copy dependent DLLs to the named build directory
+echo Copying DLLs
+copy packages\zlib_native.redist.1.2.11\build\native\bin\x64\Debug\*.dll %1
+copy packages\zlib_native.redist.1.2.11\build\native\bin\x64\Release\*.dll %1
+
+:: Run unit test program
+echo Running %1\testpdfio.exe
+cd %1
+testpdfio.exe

testpdfio.c

@@ -1,7 +1,7 @@
 //
 // Test program for PDFio.
 //
-// Copyright © 2021-2023 by Michael R Sweet.
+// Copyright © 2021-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -16,6 +16,7 @@
 #include "pdfio-private.h"
 #include "pdfio-content.h"
 #include <math.h>
+#include <locale.h>
 #ifndef M_PI
 #  define M_PI	3.14159265358979323846264338327950288
 #endif // M_PI
@@ -26,7 +27,7 @@
 //
 
 static int	do_crypto_tests(void);
-static int	do_test_file(const char *filename, int objnum, bool verbose);
+static int	do_test_file(const char *filename, int objnum, const char *password, bool verbose);
 static int	do_unit_tests(void);
 static int	draw_image(pdfio_stream_t *st, const char *name, double x, double y, double w, double h, const char *label);
 static bool	error_cb(pdfio_file_t *pdf, const char *message, bool *error);
@@ -36,6 +37,7 @@ static const char *password_cb(void *data, const char *filename);
 static int	read_unit_file(const char *filename, size_t num_pages, size_t first_image, bool is_output);
 static ssize_t	token_consume_cb(const char **s, size_t bytes);
 static ssize_t	token_peek_cb(const char **s, char *buffer, size_t bytes);
+static int	usage(FILE *fp);
 static int	verify_image(pdfio_file_t *pdf, size_t number);
 static int	write_alpha_test(pdfio_file_t *pdf, int number, pdfio_obj_t *font);
 static int	write_color_patch(pdfio_stream_t *st, bool device);
@@ -47,7 +49,7 @@ static int	write_images_test(pdfio_file_t *pdf, int number, pdfio_obj_t *font);
 static int	write_jpeg_test(pdfio_file_t *pdf, const char *title, int number, pdfio_obj_t *font, pdfio_obj_t *image);
 static int	write_png_test(pdfio_file_t *pdf, int number, pdfio_obj_t *font);
 static int	write_text_test(pdfio_file_t *pdf, int first_page, pdfio_obj_t *font, const char *filename);
-static int	write_unit_file(pdfio_file_t *inpdf, pdfio_file_t *outpdf, size_t *num_pages, size_t *first_image);
+static int	write_unit_file(pdfio_file_t *inpdf, const char *outname, pdfio_file_t *outpdf, size_t *num_pages, size_t *first_image);
 
 
 //
@@ -58,20 +60,33 @@ int					// O - Exit status
 main(int  argc,				// I - Number of command-line arguments
      char *argv[])			// I - Command-line arguments
 {
-  int	ret = 0;			// Return value
+  int		ret = 0;		// Return value
 
 
   if (argc > 1)
   {
     int		i;			// Looping var
+    const char	*password = NULL;	// Password
     bool	verbose = false;	// Be verbose?
 
     for (i = 1; i < argc; i ++)
     {
       if (!strcmp(argv[i], "--help"))
       {
-        puts("Usage: ./testpdfio [--help] [--verbose] [filename [objnum] ...]");
-        return (0);
+        return (usage(stdout));
+      }
+      else if (!strcmp(argv[i], "--password"))
+      {
+        i ++;
+        if (i < argc)
+        {
+          password = argv[i];
+        }
+        else
+        {
+          fputs("testpdfio: Missing password after '--password'.\n", stderr);
+          return (usage(stderr));
+        }
       }
       else if (!strcmp(argv[i], "--verbose"))
       {
@@ -79,24 +94,27 @@ main(int  argc,				// I - Number of command-line arguments
       }
       else if (argv[i][0] == '-')
       {
-        printf("Unknown option '%s'.\n\n", argv[i]);
-        puts("Usage: ./testpdfio [--help] [--verbose] [filename [objnum] ...]");
-        return (1);
+        fprintf(stderr, "testpdfio: Unknown option '%s'.\n", argv[i]);
+        return (usage(stderr));
       }
       else if ((i + 1) < argc && isdigit(argv[i + 1][0] & 255))
       {
         // filename.pdf object-number
-        if (do_test_file(argv[i], atoi(argv[i + 1]), verbose))
+        if (do_test_file(argv[i], atoi(argv[i + 1]), password, verbose))
 	  ret = 1;
 
 	i ++;
       }
-      else if (do_test_file(argv[i], 0, verbose))
+      else if (do_test_file(argv[i], 0, password, verbose))
+      {
         ret = 1;
+      }
     }
   }
   else
   {
+    fprintf(stderr, "testpdfio: Test locale is \"%s\".\n", setlocale(LC_ALL, getenv("LANG")));
+
 #if _WIN32
     // Windows puts executables in Platform/Configuration subdirs...
     if (!_access("../../testfiles", 0))
@@ -360,6 +378,7 @@ do_crypto_tests(void)
 static int				// O - Exit status
 do_test_file(const char *filename,	// I - PDF filename
              int        objnum,		// I - Object number to dump, if any
+             const char *password,	// I - Password for file
              bool       verbose)	// I - Be verbose?
 {
   bool		error = false;		// Have we shown an error yet?
@@ -378,7 +397,7 @@ do_test_file(const char *filename,	// I - PDF filename
     fflush(stdout);
   }
 
-  if ((pdf = pdfioFileOpen(filename, /*password_cb*/NULL, /*password_data*/NULL, (pdfio_error_cb_t)error_cb, &error)) != NULL)
+  if ((pdf = pdfioFileOpen(filename, password_cb, (void *)password, (pdfio_error_cb_t)error_cb, &error)) != NULL)
   {
     if (objnum)
     {
@@ -1041,7 +1060,7 @@ do_unit_tests(void)
   else
     goto fail;
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-out.pdf", outpdf, &num_pages, &first_image))
     goto fail;
 
   if (read_unit_file("testpdfio-out.pdf", num_pages, first_image, false))
@@ -1060,7 +1079,7 @@ do_unit_tests(void)
   else
     goto fail;
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-out2.pdf", outpdf, &num_pages, &first_image))
     goto fail;
 
   close(outfd);
@@ -1081,7 +1100,7 @@ do_unit_tests(void)
   else
     return (1);
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-rc4.pdf", outpdf, &num_pages, &first_image))
     return (1);
 
   if (read_unit_file("testpdfio-rc4.pdf", num_pages, first_image, false))
@@ -1100,7 +1119,7 @@ do_unit_tests(void)
   else
     return (1);
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-rc4p.pdf", outpdf, &num_pages, &first_image))
     return (1);
 
   if (read_unit_file("testpdfio-rc4p.pdf", num_pages, first_image, false))
@@ -1118,7 +1137,7 @@ do_unit_tests(void)
   else
     return (1);
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-aes.pdf", outpdf, &num_pages, &first_image))
     return (1);
 
   if (read_unit_file("testpdfio-aes.pdf", num_pages, first_image, false))
@@ -1136,7 +1155,7 @@ do_unit_tests(void)
   else
     return (1);
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "testpdfio-aesp.pdf", outpdf, &num_pages, &first_image))
     return (1);
 
   if (read_unit_file("testpdfio-aesp.pdf", num_pages, first_image, false))
@@ -1148,7 +1167,7 @@ do_unit_tests(void)
   else
     return (1);
 
-  if (write_unit_file(inpdf, outpdf, &num_pages, &first_image))
+  if (write_unit_file(inpdf, "<temporary>", outpdf, &num_pages, &first_image))
     return (1);
 
   if (read_unit_file(temppdf, num_pages, first_image, false))
@@ -1305,6 +1324,7 @@ read_unit_file(const char *filename,	// I - File to read
 	       bool       is_output)	// I - File written with output callback?
 {
   pdfio_file_t	*pdf;			// PDF file
+  pdfio_dict_t	*catalog;		// Catalog dictionary
   size_t	i;			// Looping var
   const char	*s;			// String
   bool		error = false;		// Error callback data
@@ -1317,6 +1337,83 @@ read_unit_file(const char *filename,	// I - File to read
   else
     return (1);
 
+  // Get the root object/catalog dictionary
+  fputs("pdfioFileGetCatalog: ", stdout);
+  if ((catalog = pdfioFileGetCatalog(pdf)) != NULL)
+  {
+    puts("PASS");
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected dictionary)");
+    return (1);
+  }
+
+  // Verify some catalog values...
+  fputs("pdfioDictGetName(PageLayout): ", stdout);
+  if ((s = pdfioDictGetName(catalog, "PageLayout")) != NULL && !strcmp(s, "SinglePage"))
+  {
+    puts("PASS");
+  }
+  else if (s)
+  {
+    printf("FAIL (got '%s', expected 'SinglePage')\n", s);
+    return (1);
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected 'SinglePage')");
+    return (1);
+  }
+
+  fputs("pdfioDictGetName(PageLayout): ", stdout);
+  if ((s = pdfioDictGetName(catalog, "PageLayout")) != NULL && !strcmp(s, "SinglePage"))
+  {
+    puts("PASS");
+  }
+  else if (s)
+  {
+    printf("FAIL (got '%s', expected 'SinglePage')\n", s);
+    return (1);
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected 'SinglePage')");
+    return (1);
+  }
+
+  fputs("pdfioDictGetName(PageMode): ", stdout);
+  if ((s = pdfioDictGetName(catalog, "PageMode")) != NULL && !strcmp(s, "UseThumbs"))
+  {
+    puts("PASS");
+  }
+  else if (s)
+  {
+    printf("FAIL (got '%s', expected 'UseThumbs')\n", s);
+    return (1);
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected 'UseThumbs')");
+    return (1);
+  }
+
+  fputs("pdfioDictGetString(Lang): ", stdout);
+  if ((s = pdfioDictGetString(catalog, "Lang")) != NULL && !strcmp(s, "en"))
+  {
+    puts("PASS");
+  }
+  else if (s)
+  {
+    printf("FAIL (got '%s', expected 'en')\n", s);
+    return (1);
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected 'en')");
+    return (1);
+  }
+
   // Verify metadata...
   fputs("pdfioFileGetAuthor: ", stdout);
   if ((s = pdfioFileGetAuthor(pdf)) != NULL && !strcmp(s, "Michael R Sweet"))
@@ -1478,6 +1575,23 @@ token_peek_cb(const char **s,		// IO - Test string
 }
 
 
+//
+// 'usage()' - Show program usage.
+//
+
+static int				// O - Exit status
+usage(FILE *fp)				// I - Output file
+{
+  fputs("Usage: ./testpdfio [OPTIONS] [FILENAME [OBJNUM]] ...\n", fp);
+  fputs("Options:\n", fp);
+  fputs("  --help               Show program help.\n", fp);
+  fputs("  --password PASSWORD  Set PDF password.\n", fp);
+  fputs("  --verbose            Be verbose.\n", fp);
+
+  return (fp != stdout);
+}
+
+
 //
 // 'verify_image()' - Verify an image object.
 //
@@ -3243,6 +3357,7 @@ write_text_test(pdfio_file_t *pdf,		// I - PDF file
 static int				// O - Exit status
 write_unit_file(
     pdfio_file_t *inpdf,		// I - Input PDF file
+    const char   *outname,		// I - Output PDF file name
     pdfio_file_t *outpdf,		// I - Output PDF file
     size_t       *num_pages,		// O - Number of pages
     size_t       *first_image)		// O - First image object
@@ -3252,8 +3367,26 @@ write_unit_file(
 			*gray_jpg,	// gray.jpg image
 			*helvetica,	// Helvetica font
 			*page;		// Page from test PDF file
+  pdfio_dict_t		*catalog;	// Catalog dictionary
 
 
+  // Get the root object/catalog dictionary
+  fputs("pdfioFileGetCatalog: ", stdout);
+  if ((catalog = pdfioFileGetCatalog(outpdf)) != NULL)
+  {
+    puts("PASS");
+  }
+  else
+  {
+    puts("FAIL (got NULL, expected dictionary)");
+    return (1);
+  }
+
+  // Set some catalog values...
+  pdfioDictSetName(catalog, "PageLayout", "SinglePage");
+  pdfioDictSetName(catalog, "PageMode", "UseThumbs");
+  pdfioDictSetString(catalog, "Lang", "en");
+
   // Set info values...
   fputs("pdfioFileGet/SetAuthor: ", stdout);
   pdfioFileSetAuthor(outpdf, "Michael R Sweet");
@@ -3437,7 +3570,7 @@ write_unit_file(
   }
 
   // Close the new PDF file...
-  fputs("pdfioFileClose(...): ", stdout);
+  printf("pdfioFileClose(\"%s\"): ", outname);
   if (pdfioFileClose(outpdf))
     puts("PASS");
   else

testttf.c

@@ -3,7 +3,7 @@
 //
 //     https://github.com/michaelrsweet/ttf
 //
-// Copyright © 2018-2023 by Michael R Sweet.
+// Copyright © 2018-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -120,6 +120,7 @@ test_font(const char *filename)		// I - Font filename
 
 
   printf("ttfCreate(\"%s\"): ", filename);
+  fflush(stdout);
   if ((font = ttfCreate(filename, 0, error_cb, NULL)) != NULL)
     puts("PASS");
   else

ttf.c

@@ -3,7 +3,7 @@
 //
 //     https://github.com/michaelrsweet/ttf
 //
-// Copyright © 2018-2023 by Michael R Sweet.
+// Copyright © 2018-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -62,7 +62,7 @@
 #  define O_CREAT	_O_CREAT
 #  define O_TRUNC	_O_TRUNC
 
-typedef __int64 ssize_t;		// POSIX type not present on Windows...
+typedef __int64 ssize_t;		// POSIX type not present on Windows... @private@
 
 #else
 #  include <unistd.h>
@@ -99,6 +99,8 @@ typedef __int64 ssize_t;		// POSIX type not present on Windows...
 //
 
 #define TTF_FONT_MAX_CHAR	262144	// Maximum number of character values
+#define TTF_FONT_MAX_GROUPS	65536	// Maximum number of sub-groups
+#define TTF_FONT_MAX_NAMES	16777216// Maximum size of names table we support
 
 
 //
@@ -254,7 +256,7 @@ typedef struct _ttf_off_hhea_s		// Horizontal header
 {
   short		ascender,		// Ascender
 		descender;		// Descender
-  int		numberOfHMetrics;	// Number of horizontal metrics
+  unsigned short numberOfHMetrics;	// Number of horizontal metrics
 } _ttf_off_hhea_t;
 
 typedef struct _ttf_off_os_2_s		// OS/2 information
@@ -297,7 +299,28 @@ static unsigned	seek_table(ttf_t *font, unsigned tag, unsigned offset, bool requ
 
 
 //
-// 'ttfCreate()' - Create a new font object for the named font family.
+// 'ttfCreate()' - Create a new font object for the named font file.
+//
+// This function creates a new font object for the named TrueType or OpenType
+// font file or collection.  The "filename" argument specifies the name of the
+// file to read.
+//
+// The "idx" argument specifies the font to load from a collection - the first
+// font is number `0`.  Once created, you can call the @link ttfGetNumFonts@
+// function to determine whether the loaded font file is a collection with more
+// than one font.
+//
+// The "err_cb" and "err_data" arguments specify a callback function and data
+// pointer for receiving error messages.  If `NULL`, errors are sent to the
+// `stderr` file.  The callback function receives the data pointer and a text
+// message string, for example:
+//
+// ```
+// void my_err_cb(void *err_data, const char *message)
+// {
+//   fprintf(stderr, "ERROR: %s\n", message);
+// }
+// ```
 //
 
 ttf_t *					// O - New font object
@@ -550,6 +573,10 @@ ttfGetAscent(ttf_t *font)		// I - Font
 //
 // 'ttfGetBounds()' - Get the bounds of all characters in a font.
 //
+// This function gets the bounds of all characters in a font.  The "bounds"
+// argument is a pointer to a `ttf_rect_t` structure that will be filled with
+// the limits for characters in the font scaled to a 1000x1000 unit square.
+//
 
 ttf_rect_t *				// O - Bounds or `NULL` on error
 ttfGetBounds(ttf_t      *font,		// I - Font
@@ -631,8 +658,11 @@ ttfGetDescent(ttf_t *font)		// I - Font
 //
 // 'ttfGetExtents()' - Get the extents of a UTF-8 string.
 //
-// This function computes the extents of a UTF-8 string when rendered using the
-// specified font and size.
+// This function computes the extents of the UTF-8 string "s" when rendered
+// using the specified font "font" and size "size".  The "extents" argument is
+// a pointer to a `ttf_rect_t` structure that is filled with the extents of a
+// simple rendering of the string with no kerning or rewriting applied.  The
+// values are scaled using the specified font size.
 //
 
 ttf_rect_t *				// O - Pointer to extents or `NULL` on error
@@ -1272,20 +1302,38 @@ read_cmap(ttf_t *font)			// I - Font
           for (i = 0; i < numGlyphIdArray; i ++)
             glyphIdArray[i] = read_ushort(font);
 
-#ifdef DEBUG
-          for (i = 0; i < segCount; i ++)
-            TTF_DEBUG("read_cmap: segment[%d].startCode=%d, endCode=%d, idDelta=%d, idRangeOffset=%d\n", i, segments[i].startCode, segments[i].endCode, segments[i].idDelta, segments[i].idRangeOffset);
+          for (i = 0, segment = segments; i < segCount; i ++, segment ++)
+          {
+            TTF_DEBUG("read_cmap: segment[%d].startCode=%d, endCode=%d, idDelta=%d, idRangeOffset=%d\n", i, segment->startCode, segment->endCode, segment->idDelta, segment->idRangeOffset);
 
+            if (segment->startCode > segment->endCode)
+            {
+	      errorf(font, "Bad cmap table segment %u to %u.", segments->startCode, segment->endCode);
+	      free(segments);
+	      free(glyphIdArray);
+	      return (false);
+            }
+
+            // Based on the end code of the segment table, allocate space for the
+            // uncompressed cmap table...
+            if (segment->endCode >= font->num_cmap)
+	      font->num_cmap = segment->endCode + 1;
+          }
+
+#ifdef DEBUG
           for (i = 0; i < numGlyphIdArray; i ++)
             TTF_DEBUG("read_cmap: glyphIdArray[%d]=%d\n", i, glyphIdArray[i]);
 #endif /* DEBUG */
 
-          // Based on the end code of the segent table, allocate space for the
-          // uncompressed cmap table...
-//          segCount --;			// Last segment is not used (sigh)
+	  if (font->num_cmap == 0 || font->num_cmap > TTF_FONT_MAX_CHAR)
+	  {
+	    errorf(font, "Invalid cmap table with %u characters.", (unsigned)font->num_cmap);
+	    free(segments);
+	    free(glyphIdArray);
+	    return (false);
+	  }
 
-	  font->num_cmap = segments[segCount - 1].endCode + 1;
-	  font->cmap     = cmapptr = (int *)malloc(font->num_cmap * sizeof(int));
+	  font->cmap = cmapptr = (int *)malloc(font->num_cmap * sizeof(int));
 
 	  if (!font->cmap)
           {
@@ -1356,6 +1404,12 @@ read_cmap(ttf_t *font)			// I - Font
 
 	  TTF_DEBUG("read_cmap: nGroups=%u\n", nGroups);
 
+	  if (nGroups > TTF_FONT_MAX_GROUPS)
+	  {
+	    errorf(font, "Invalid cmap table with %u groups.", nGroups);
+	    return (false);
+	  }
+
 	  if ((groups = (_ttf_off_cmap12_t *)calloc(nGroups, sizeof(_ttf_off_cmap12_t))) == NULL)
           {
             errorf(font, "Unable to allocate memory for cmap.");
@@ -1369,6 +1423,13 @@ read_cmap(ttf_t *font)			// I - Font
 	    group->startGlyphID  = read_ulong(font);
 	    TTF_DEBUG("read_cmap: [%u] startCharCode=%u, endCharCode=%u, startGlyphID=%u\n", gidx, group->startCharCode, group->endCharCode, group->startGlyphID);
 
+            if (group->startCharCode > group->endCharCode)
+            {
+	      errorf(font, "Bad cmap table segment %u to %u.", group->startCharCode, group->endCharCode);
+	      free(groups);
+	      return (false);
+            }
+
             if (group->endCharCode >= font->num_cmap)
               font->num_cmap = group->endCharCode + 1;
 	  }
@@ -1376,6 +1437,14 @@ read_cmap(ttf_t *font)			// I - Font
 	  // Based on the end code of the segent table, allocate space for the
 	  // uncompressed cmap table...
           TTF_DEBUG("read_cmap: num_cmap=%u\n", (unsigned)font->num_cmap);
+
+	  if (font->num_cmap == 0 || font->num_cmap > TTF_FONT_MAX_CHAR)
+	  {
+	    errorf(font, "Invalid cmap table with %u characters.", (unsigned)font->num_cmap);
+	    free(groups);
+	    return (false);
+	  }
+
 	  font->cmap = cmapptr = (int *)malloc(font->num_cmap * sizeof(int));
 
 	  if (!font->cmap)
@@ -1426,6 +1495,12 @@ read_cmap(ttf_t *font)			// I - Font
 
 	  TTF_DEBUG("read_cmap: nGroups=%u\n", nGroups);
 
+	  if (nGroups > TTF_FONT_MAX_GROUPS)
+	  {
+	    errorf(font, "Invalid cmap table with %u groups.", nGroups);
+	    return (false);
+	  }
+
 	  if ((groups = (_ttf_off_cmap13_t *)calloc(nGroups, sizeof(_ttf_off_cmap13_t))) == NULL)
 	  {
 	    errorf(font, "Unable to allocate memory for cmap.");
@@ -1439,6 +1514,13 @@ read_cmap(ttf_t *font)			// I - Font
 	    group->glyphID       = read_ulong(font);
 	    TTF_DEBUG("read_cmap: [%u] startCharCode=%u, endCharCode=%u, glyphID=%u\n", gidx, group->startCharCode, group->endCharCode, group->glyphID);
 
+            if (group->startCharCode > group->endCharCode)
+            {
+	      errorf(font, "Bad cmap table segment %u to %u.", group->startCharCode, group->endCharCode);
+	      free(groups);
+	      return (false);
+            }
+
             if (group->endCharCode >= font->num_cmap)
               font->num_cmap = group->endCharCode + 1;
 	  }
@@ -1446,6 +1528,14 @@ read_cmap(ttf_t *font)			// I - Font
 	  // Based on the end code of the segent table, allocate space for the
 	  // uncompressed cmap table...
           TTF_DEBUG("read_cmap: num_cmap=%u\n", (unsigned)font->num_cmap);
+
+	  if (font->num_cmap == 0 || font->num_cmap > TTF_FONT_MAX_CHAR)
+	  {
+	    errorf(font, "Invalid cmap table with %u characters.", (unsigned)font->num_cmap);
+	    free(groups);
+	    return (false);
+	  }
+
 	  font->cmap = cmapptr = (int *)malloc(font->num_cmap * sizeof(int));
 
 	  if (!font->cmap)
@@ -1565,7 +1655,7 @@ read_hmtx(ttf_t           *font,	// I - Font
           _ttf_off_hhea_t *hhea)	// O - hhea table data
 {
   unsigned	length;			// Length of hmtx table
-  int		i;			// Looping var
+  unsigned	i;			// Looping var
   _ttf_metric_t	*widths;		// Glyph metrics array
 
 
@@ -1644,8 +1734,15 @@ read_names(ttf_t *font)			// I - Font
     return (false);
 
   font->names.storage_size = length - (unsigned)offset;
+  if (font->names.storage_size > TTF_FONT_MAX_NAMES)
+  {
+    errorf(font, "Name table too large - %u bytes.", (unsigned)font->names.storage_size);
+    return (false);
+  }
+
   if ((font->names.storage = malloc(font->names.storage_size)) == NULL)
     return (false);
+
   memset(font->names.storage, 'A', font->names.storage_size);
 
   for (i = font->names.num_names, name = font->names.names; i > 0; i --, name ++)

ttf.h

@@ -3,7 +3,7 @@
 //
 //     https://github.com/michaelrsweet/ttf
 //
-// Copyright © 2018-2023 by Michael R Sweet.
+// Copyright © 2018-2024 by Michael R Sweet.
 //
 // Licensed under Apache License v2.0.  See the file "LICENSE" for more
 // information.
@@ -11,6 +11,7 @@
 
 #ifndef TTF_H
 #  define TTF_H
+#  include <stddef.h>
 #  include <stdbool.h>
 #  include <sys/types.h>
 #  ifdef __cplusplus
@@ -22,12 +23,12 @@ extern "C" {
 // Types...
 //
 
-typedef struct _ttf_s ttf_t;	//// Font object
+typedef struct _ttf_s ttf_t;	// Font object
 
 typedef void (*ttf_err_cb_t)(void *data, const char *message);
-				//// Font error callback
+				// Font error callback
 
-typedef enum ttf_stretch_e	//// Font stretch
+typedef enum ttf_stretch_e	// Font stretch
 {
   TTF_STRETCH_NORMAL,		// normal
   TTF_STRETCH_ULTRA_CONDENSED,	// ultra-condensed
@@ -40,20 +41,20 @@ typedef enum ttf_stretch_e	//// Font stretch
   TTF_STRETCH_ULTRA_EXPANDED	// ultra-expanded
 } ttf_stretch_t;
 
-typedef enum ttf_style_e	//// Font style
+typedef enum ttf_style_e	// Font style
 {
   TTF_STYLE_NORMAL,		// Normal font
   TTF_STYLE_ITALIC,		// Italic font
   TTF_STYLE_OBLIQUE		// Oblique (angled) font
 } ttf_style_t;
 
-typedef enum ttf_variant_e	//// Font variant
+typedef enum ttf_variant_e	// Font variant
 {
   TTF_VARIANT_NORMAL,		// Normal font
   TTF_VARIANT_SMALL_CAPS	// Font whose lowercase letters are small capitals
 } ttf_variant_t;
 
-typedef enum ttf_weight_e	//// Font weight
+typedef enum ttf_weight_e	// Font weight
 {
   TTF_WEIGHT_100 = 100,		// Weight 100 (Thin)
   TTF_WEIGHT_200 = 200,		// Weight 200 (Extra/Ultra-Light)
@@ -66,7 +67,7 @@ typedef enum ttf_weight_e	//// Font weight
   TTF_WEIGHT_900 = 900		// Weight 900 (Black/Heavy)
 } ttf_weight_t;
 
-typedef struct ttf_rect_s	//// Bounding rectangle
+typedef struct ttf_rect_s	// Bounding rectangle
 {
   float	left;			// Left offset
   float	top;			// Top offset