Changes for libdmtx

version 0.7.2 [September 2009]
  Added initial macro decoding support (thanks Marlon!)
  Fast quad fill for dmtxDecodeMatrixRegion() (thanks Mackenzie!)
  Fixed capacity bug with rectangle requests
  New Vala wrapper (thanks Evan!)
  Wrapper integration in build system (thanks Evan!)
  Add as source package option
  Add as binary package option
  Add "project" directory to EXTRA_DIST in

version 0.7.0 [March 2009]
  New Java wrapper (thanks Pete and Dikran!)
  New .NET wrapper (thanks Vali and Joe!)
  Added solution and project files for MS Visual Studio 9.0
  Reader support for FNC1 and upper shift (thanks Robin!)
  Support for byte-padded rows through dmtxImageSetProp()
  New no-copy image handling with configurable pixel packing
  Moved image scaling factors to DmtxDecode
  Moved scan cache from DmtxImage to DmtxDecode
  Eliminate types DmtxRgb, DmtxColor3, and DmtxGradient
  API updates for consistent naming and conventions
  Added dmtxEncodeSetProp() and dmtxEncodeGetProp()
  Option to print extended ASCII as UTF-8 (thanks Robin!)
  Fixed diagnostic image output
  Fixed misrepresented capacity in verbose mode
  True vector SVG output bypassing ImageMagick (thanks Richard!)
  Use ImageMagick to write images in major raster formats
  Fixed several bugs and compiler warnings

version 0.6.0 [November 2008]
  dmtxread now scans all major image formats [Olivier]
  New encoding/decoding Ruby wrapper [Romain]
  Reduced memory footprint
  Will scan multiple barcodes per image
  Various platform-specific improvements
  Initial work preparing for custom pixel packing in future
  Begin static analysis cleanup with splint
  New --disable-dmtxread and --disable-dmtxwrite [Romain]
  Ability to specify max/min expected barcode sizes
  New edge neighbor tracking (Hough + 2 way edge cache)
  Info cache to track scan progress and avoid rescanning pixels
  Major reduction in floating point operations
  New informative return codes (found, not found, error)
  Read input from STDIN
  Diagnostic images display trail left by scanning logic
  Option to write output to STDOUT [Olivier]
  PHP wrapper now compiles with recent libdmtx
  Dedicated instructions for specific platforms

version 0.5.2 [September 2008]
  Move SetRangeLimit and SetScanRegion into library
  Add edge threshold filtering
  Add encoding support for 144x144 barcodes
  Fixed encoding case when message starts with two digits
  Fixed bug in range limit option
  Add dynamic image shrinking (pixel skipping)
  Add step-by-step diagnostic image dump (debug build)
  Fixed bug in minimum scan gap setting
  Removed y-flip from internal pixel storage
  Added strict border tests to eliminate false positives
  Added squareness deviation filter
  Implement simplified Hough transform for locating first edge
  Several behind-the-scenes performance enhancements
  Python wrapper update; 15x faster (thanks Jonathan!)
  New PHP wrapper code added
  Various improvements when building for OS X and FreeBSD

version 0.5.1 [July 2008]
  Fixed Extended ASCII encoding bug
  Fixed error correction bug related to multiple interleaved blocks
  Added timeout condition for region detection
  Allow partial and complete disabling of error correction
  Replaced DmtxPixel struct with DmtxRgb for safe pixel copies
  Tighter integration with libfec
  Conditional build logic for libtiff
  Added placeholder for new utility, dmtxquery
  Added unit test program for testing libdmtx internals
  Include local copies of getopt1.c, getopt.c, and getopt.h
  Various things to help compiling in MS VC++
  Lots of holes filled in comments (Doxygen)
  Fixed experimental Data Mosaic decoding
  New Cocoa/iPhone wrapper (thanks Stefan!)

version 0.5 [April 2008]
  Error correction using libfec (thanks Florian!)
  Rework encoding and decoding API for consistency and intuitiveness
  Handle region detection and region decoding as separate tasks
  Pass found regions back to calling app before attempting decode
  Image mask approach (for performance and multi-barcode scans)
  Fix TestForEndOfSymbolEdifact() to handle special cases correctly
  Roll scan pattern into core library (inward breadth-first cross)
  Replace dmtxScanLine() with dmtxScanPixel()
  Implement 4-direction weighted module decisions (eliminates thresholds)
  Add ability to scan portion of image
  Add --diagnose option that dumps image with embedded scan infomation
  Added Z rotation angle (in degrees) to verbose output
  Move ASCII and codeword output to separate --preview option
  Added -R option for setting image print resolution in dpi (PNG only)
  Remove gltest and simpletest from default build target
  Update Subversion to be keyword friendly ($Id$)
  Updated documentation to reflect API and option changes

version 0.4 [December 2007]
  Remove arbitrary sz scaling (100.0) since it doesn't matter anyway
  Fix 4 bottom-right modules in sizes where they are not used (thanks Matthias R.!)
  Replace callback references with preprocessor macros
  Implement remaining encodation schemes for encoding (X12, Base 256, etc...)
  Implement remaining encodation schemes for decoding (X12, Base 256, etc...)
  Implement --auto-best option for best possible encoding efficiency
  Implement multi-region symbols
  Read and write rectangle barcodes
  Use GNU autotools (autoconf, automake, libtool)
  New region detection overhaul
  Include initial version of Python bindings from pydmtx project (thanks Dan!)
  Add decoding functionality through Python
  Add marathon images to project (thanks!)
  Fix dmtxread crash when same barcode is found more than 16 times
  Verbose messages describing traits of detected barcodes
  --codewords option to print codewords instead of decoded message
  --newline option to insert newline character at end of output
  Additional output formats (PNG, ASCII, codewords)
  'make test' executes regression tests for encodation

version 0.3 [October 2006]
  Several high-level changes: build process, naming consistency, file layout
  Added new automatic style checks in script directory ("make style")
  Implemented remaining encodation schemes for decode (X12, Base 256, etc...)
  Fixed instability/regressions that were introduced in v0.2 release
  Color sampling now averaged from multiple pixel locations
  Size calibration accuracy improved with new approach
  dmtxread: increased default scanline count, added multi-page TIFF format
  dmtxwrite: bug fixes, implemented -o option
  Improved documentation: deps listed in INSTALL, new man page for dmtxwrite

version 0.2 [June 2006]
  Cleaned up program structure surrounding decoding process
  Improved API for real-world use (no longer just dumping results to STDOUT)
  Added "dmtxread" command line utility
  Added "dmtxwrite" command line utility
  Implemented Reed-Solomon error detection
  Created "simpletest.c" for full circle processing test
  Now using libpng(3) in library to read Data Matrix images
  Improved documentation (somewhat)

version 0.1 [April 2006]
  Initial release