1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
|
Release 2.0.5
=============
* Fix SBCL-specific sequence copying implementation to use better
suited internal interface.
Release 2.0.4
=============
* Clarified public domain status for code authored by Pierre R. Mai
through additional CC0 waiver.
Release 2.0.3
=============
* Add support for md5sum-string for other implementations through
through flexi-streams. Patch supplied by Daniel Kochmanski.
Release 2.0.2
=============
* This release adds support for LispWorks 7.0, reworking the fixes
for 64-bit implementations of Lispworks in 2.0.1 and incorrect
use of the sys:simple-int32-vector type. Patch supplied by
Martin Simmons from LispWorks.
* Minor fix to prevent style-warnings for implementations with no
support for md5sum-string.
Release 2.0.1
=============
* This release fixes problems on 64-bit implementations of Lispworks,
where sys:int32 arithmetic can overflow 32-bit values and hence
yield wrong results.
Release 2.0.0
=============
* This release consolidates the various versions of md5 that have
been floating around, including versions with Lispworks
optimizations, the version maintained by Kevin M. Rosenberg, the
SBCL-optimized version, and various patches and fixes.
* This release tries to separate out string-handling from actual
byte-based MD5 hash generation. Hence users who want to generate
hashes from strings should use the new md5sum-string function,
which will go through your implementation's external-format
handling to generate the actual byte-array that is hashed. Usage
of md5sum-sequence for strings and md5sum-stream for character
streams is deprecated, since it only ever worked correctly for 8bit
characters in most implementations.
* This release should work correctly on at least recentish versions
of CMUCL, SBCL, ECL, CCL, Lispworks, AllegroCL and ABCL and
generally most other conforming implementations. It has been
optimized for CMUCL, SBCL and Lispworks, and should work with
adequate performance on most 64bit implementations. Performance on
other 32bit implementations is still sub-optimal if they don't
support unboxed bit-operations on (unsigned-byte 32).
|
