fukasawa e60969
Libpng 1.6.21 - January 15, 2016
fukasawa e60969
fukasawa e60969
This is a public release of libpng, intended for use in production codes.
fukasawa e60969
fukasawa e60969
Files available for download:
fukasawa e60969
fukasawa e60969
Source files with LF line endings (for Unix/Linux) and with a
fukasawa e60969
"configure" script
fukasawa e60969
fukasawa e60969
   libpng-1.6.21.tar.xz (LZMA-compressed, recommended)
fukasawa e60969
   libpng-1.6.21.tar.gz
fukasawa e60969
fukasawa e60969
Source files with CRLF line endings (for Windows), without the
fukasawa e60969
"configure" script
fukasawa e60969
fukasawa e60969
   /scratch/glennrp/Libpng16/lpng1621.7z  (LZMA-compressed, recommended)
fukasawa e60969
   /scratch/glennrp/Libpng16/lpng1621.zip
fukasawa e60969
fukasawa e60969
Other information:
fukasawa e60969
fukasawa e60969
   libpng-1.6.21-README.txt
fukasawa e60969
   libpng-1.6.21-LICENSE.txt
fukasawa e60969
   libpng-1.6.21-*.asc (armored detached GPG signatures)
fukasawa e60969
fukasawa e60969
Changes since the last public release (1.6.20):
fukasawa e60969
fukasawa e60969
  Fixed syntax "$(command)" in tests/pngstest that some shells other than
fukasawa e60969
    bash could not parse (Bug report by Nelson Beebe). Use `command` instead.
fukasawa e60969
  Moved png_check_keyword() from pngwutil.c to pngset.c
fukasawa e60969
  Removed LE/BE dependencies in pngvalid, to 'fix' the current problem
fukasawa e60969
    in the BigEndian tests by not testing it, making the BE code the same 
fukasawa e60969
    as the LE version.
fukasawa e60969
  Fixes to pngvalid for various reduced build configurations (eliminate unused
fukasawa e60969
    statics) and a fix for the case in rgb_to_gray when the digitize option
fukasawa e60969
    reduces graylo to 0, producing a large error.
fukasawa e60969
  Widened the 'limit' check on the internally calculated error limits in
fukasawa e60969
    the 'DIGITIZE' case (the code used prior to 1.7 for rgb_to_gray error
fukasawa e60969
    checks) and changed the check to only operate in non-release builds
fukasawa e60969
   (base build type not RC or RELEASE.)
fukasawa e60969
  Fixed undefined behavior in pngvalid.c, undefined because
fukasawa e60969
    (png_byte) << shift is undefined if it changes the signed bit
fukasawa e60969
    (because png_byte is promoted to int). The libpng exported functions
fukasawa e60969
    png_get_uint_32 and png_get_uint_16 handle this. (Bug reported by
fukasawa e60969
    David Drysdale as a result of reports from UBSAN in clang 3.8).
fukasawa e60969
  This changes pngvalid to use BE random numbers; this used to produce
fukasawa e60969
    errors but these should not be fixed as a result of the previous changes.
fukasawa e60969
  In projects/vstudio, combined readme.txt and WARNING into README.txt
fukasawa e60969
  Relocated assert() in contrib/tools/pngfix.c, bug found by American
fukasawa e60969
    Fuzzy Lop, reported by Brian Carpenter.
fukasawa e60969
  Marked 'limit' UNUSED in transform_range_check().  This only affects
fukasawa e60969
    release builds.
fukasawa e60969
  Worked around a false-positive Coverity issue in pngvalid.c.
fukasawa e60969
fukasawa e60969
Send comments/corrections/commendations to png-mng-implement at lists.sf.net
fukasawa e60969
(subscription required; visit
fukasawa e60969
https://lists.sourceforge.net/lists/listinfo/png-mng-implement
fukasawa e60969
to subscribe)
fukasawa e60969
or to glennrp at users.sourceforge.net
fukasawa e60969
fukasawa e60969
Glenn R-P