source: trunk/essentials/dev-lang/perl/ext/Digest/MD5/Changes

2005-11-30   Gisle Aas <[email protected]>

   Release 2.36

   Fix documentation typo.



2005-11-26   Gisle Aas <[email protected]>

   Release 2.35

   Forgot to incorporate fixes already applied to bleadperl :-(
     - doc typo
     - consting
     - unused my_na
     - USE_HEAP_INSTEAD_OF_STACK for Symbian



2005-11-26   Gisle Aas <[email protected]>

   Release 2.34

   Document that it is now easy to generate different messages that produce the
   same MD5 digest.

   Use XSLoader; perl-5.6 is now required.

   Tweaks to the processing of $? after running the test program.



2003-12-07   Gisle Aas <[email protected]>

   Release 2.33
   
   Enable explicit context passing for slight performance
   improvement in threaded perls.
   
   Tweaks to the Makefile.PL so that it is suitable both for
   core and CPAN use.



2003-12-05   Gisle Aas <[email protected]>

   Release 2.32

   Don't run u32align test program on HP-UX 10.20 as it
   will hang.  Patch by H.Merijn Brand <[email protected]>.

   Fixed documentation typo.



2003-11-28   Gisle Aas <[email protected]>

   Release 2.31

   Inherit add_bits() from Digest::base if available.



2003-10-09   Gisle Aas <[email protected]>

   Release 2.30

   Some tweaks to make the module build on perl-5.004 and
   perl-5.005 again.



2003-10-06   Gisle Aas <[email protected]>

   Release 2.29

   Another try.  Forgot to update the test checksums.



2003-10-06   Gisle Aas <[email protected]>

   Release 2.28

   Fix minor documentation typo.



2003-08-04   Gisle Aas <[email protected]>

   Release 2.27

   Avoid having to figure how to compile alignment test program
   on MS-Windows by just assume free alignment as for i386.  Source
   builds on Windows was apparently broken.



2003-07-21   Gisle Aas <[email protected]>

   Release 2.26

   Don't assume PerlIO_read() works like fread() even though
   it was documented like that for perl 5.6.  It returns negative
   on read failure.

   Kill test #3 in t/badfile.t.  I don't know a reliable way
   to test read failures on a file handle.  Seems better not to
   test than to make many worry.



2003-07-04   Gisle Aas <[email protected]>

   Release 2.25

   The $md5->addfile method now croaks if it discovers
   errors on the handle after reading from it.  This should
   make it more difficult to end up with the wrong digest
   just because you are to lazy to check the error status
   on your file handles after reading from them.

   Improved documentation.

   Sync up with bleadperl; even safer patchlevel include.



2003-03-09   Gisle Aas <[email protected]>

   Release 2.24

   Don't let the $^W test get confused by lexical warnings.

   Sync up with bleadperl; safer patchlevel include.



2003-01-18   Gisle Aas <[email protected]>

   Release 2.23

   Override INSTALLDIRS for 5.8 as suggested by
   Guido Ostkamp <[email protected]>.



2003-01-04   Gisle Aas <[email protected]>

   Release 2.22.

   Added clone method.
   Contributed by Holger Smolinski <[email protected]>



2002-12-27   Gisle Aas <[email protected]>

   Release 2.21

   Minor tweaks sync up with bleadperl:
     - VMS optimizer tweaks to the Makefile.PL
     - MacOS support
     - Added alignment test

   Added example to the MD5 POD that shows how to calculate the
   digest of Unicode strings.



2002-05-05   Gisle Aas <[email protected]>

   Release 2.20

   More synchronization with tweaks Jarkko have done to the
   bleadperl test suite. This time various EBCDIC hacks.

   Outside PERL_CORE the md5-aaa.t test loaded the wrong version of
   the module (and would fail if no previous Digest::MD5 was installed).
   Patch by Mike Stok <[email protected]>



2002-05-01   Gisle Aas <[email protected]>

   Release 2.19

   One more test suite update from Jarkko to sync it
   even better with bleadperl.



2002-05-01   Gisle Aas <[email protected]>

   Release 2.18

   Changes #12954 and #16173 from bleadperl.  Documentation typo fix
   and some signed/unsigned mismatches that Microsoft's C compiler
   complained about.

   The EBCDIC-aware md5-aaa.t from bleadperl.



2002-04-25   Gisle Aas <[email protected]>

   Release 2.17

   The SvPVbyte in perl-5.6.1 is buggy.  Use the one from 5.7.3
   instead.

   Give warning if the function interface is used as instance
   methods:  $md5->md5_hex().



2001-09-07   Gisle Aas <[email protected]>

   Release 2.16

   Sync up with the bleadperl version:
      - use SvPVbyte() if available
      - fixes to make the code 'gcc -Wall'-clean



2001-08-27   Gisle Aas <[email protected]>

   Release 2.15

   Avoid exit() in Makefile.PL and bleadperl redefinition of printf
   in the alignment test program.
   Patch by Doug MacEachern <[email protected]>.



2001-07-18   Gisle Aas <[email protected]>

   Release 2.14

   Try to warn if the functional interface is used as methods,
   i.e. Digest::MD5->md5_hex("foo") will make noise if -w is
   enabled.

   Document the missing padding for the base64 digests.

   If both XS bootstrap and locating Digest::Perl::MD5 fails
   re-raise the original XS bootstrap exception.



2001-03-13   Gisle Aas <[email protected]>

   Release 2.13

   Moved all other Digest:: modules out of the Digest-MD5 dist.



2000-09-18   Gisle Aas <[email protected]>

   Release 2.12

   Avoid pointer cast warning for machines with bigger ints
   than pointers.  Patch by Robin Barker <[email protected]>.



2000-08-19   Gisle Aas <[email protected]>

   Release 2.11
   
   The fallback code introduced in 2.10 did only work for
   perl-5.6.0.  It should now for for perl5.004 and 5.005
   as well.  Patch by Ville Skyttä <[email protected]>.



2000-08-18   Gisle Aas <[email protected]>

   Release 2.10

   Digest::MD5 will now try to fallback to the pure perl
   implementation of Digest::Perl::MD5 if bootstrap fails.

   Added a bit internal paranoia about casting the IV
   in the Digest::MD5 object to the MD5_CTX* pointer.



1999-08-06   Gisle Aas <[email protected]>

   Release 2.09

   Documentation update.



1999-07-28   Gisle Aas <[email protected]>

   Release 2.08

   The addfile() methods could trigger a core dump when passed
   a filehandle that had failed to open.



1999-04-26   Gisle Aas <[email protected]>

   Release 2.07

   The Digest::SHA1 module failed on some 64-bit systems, because I
   assumed there was a correspondence between the U32 size and
   BYTEORDER.  This version use 'unsigned long' as Uwe's original
   SHA module did.

   The module should now work better when liked statically with perl,
   because we now use a safer module-loaded test in Digest.pm.

   Assume we know the outcome of the alignment test on VMS. Patch by
   Chuck Lane <[email protected]>



1999-03-26   Gisle Aas <[email protected]>

   Release 2.06

   Avoid LONG and BYTE types in SHA.xs as they was in conflict
   with similar definitions in <winnt.h>.

   Patch by Marko Asplund <[email protected]> to make the the alignment
   test program link successfully with sfio-perl.

   Fixed a typo in MD5.xs that might have affected 64-bit systems.
   Spotted by Nick Ing-Simmons



1999-03-15   Gisle Aas <[email protected]>

   Release 2.05

   Included Digest::SHA1 based on Uwe Hollerbach's SHA module.



1999-03-05   Gisle Aas <[email protected]>

   Release 2.04

   Avoid the -o option when compiling alignment test program
   for Win32 as suggested by Gurusamy Sarathy.

   DEC Compiler bug workaround.  Contributed by D Roland Walker
   <[email protected]>

   Having references to a local variable called "na" was not
   very safe either.  Some older versions of Perl can apparently