exa/codecrypt
1
0
Fork 0
Code Issues Pull requests Actions Packages Projects Releases Wiki Activity
326 commits 1 branch 0 tags 20 MiB
Commit graph

198 commits

Author SHA1 Message Date
Mirek Kratochvil 0cdd83dbc4 mce_qd: fix unused variable warn 2014-02-26 12:51:44 +01:00
Mirek Kratochvil 7a71ca74f0 mce_qd: faster alternant check matrix computation 
Save a lot of log/antilog lookups. On my machine, this is 4-6x faster.
 2014-02-08 09:48:10 +01:00
Mirek Kratochvil 0978a40372 str_match: fix icase matching 2014-02-02 17:50:30 +01:00
Mirek Kratochvil 6fd26d4c79 gf2m: save some log lookups with combined ops 2014-02-02 17:47:15 +01:00
Mirek Kratochvil b566beaa38 permutation: avoid needlessly precomputed hpermInv 2014-02-02 17:18:48 +01:00
Mirek Kratochvil 9316509223 keyring: remove obsolete code 2014-02-02 17:18:12 +01:00
Mirek Kratochvil e66e82b9a8 str_match: keyspec matches are ignorecase 2014-02-02 17:04:49 +01:00
Mirek Kratochvil 54d5662d55 fmtseq: be more verbose about signatures left 2014-01-25 17:49:11 +01:00
Mirek Kratochvil 812ef1eac7 keyring: do backups and save some writes 2014-01-25 17:27:02 +01:00
Mirek Kratochvil f3ede079e9 keyring: prevent data corruption from signal 2014-01-25 12:35:41 +01:00
Mirek Kratochvil fee6b431c2 fmtseq: privkey internals checking 
Simple size checks that prevent some segfaults from working with mangled
privkeys.
 2014-01-25 10:34:33 +01:00
Mirek Kratochvil a0a9263a42 main: happy new year. 2014-01-24 10:39:36 +01:00
Mirek Kratochvil 1ccd8e8b3a actions: simplify key deletion code a bit 2014-01-24 10:35:56 +01:00
Mirek Kratochvil f2d1ba9365 actions: be helpful on missing ascii-armor option 
Refs #4 on github
 2014-01-24 10:30:13 +01:00
Mirek Kratochvil 06378a826a keyring: compile and work on MinGW 
Thanks for suggestion&test to Jens-Uwe Rammelt <jens-uwer@web.de>.
 2014-01-24 10:08:37 +01:00
Mirek Kratochvil 594e3c0ebc algos_enc: clarify padding checks 2013-10-06 09:18:08 +02:00
Mirek Kratochvil e795744eef mce_qd: remove unnecessary root check 2013-10-06 09:09:08 +02:00
Mirek Kratochvil 5b77413a7f keyring: updating the version was not needed 2013-09-30 12:14:02 +02:00
Mirek Kratochvil d7192caad0 serialization: bump message version strings 2013-09-30 12:06:19 +02:00
Mirek Kratochvil 5333b6e1f8 algos_sig: fix min() macro name collision with STL 2013-09-30 12:00:06 +02:00
Mirek Kratochvil 3deffb0ebe build: crypto++ support is now optional 
possibly also evading "traditional" hash functions. Cubehash is good.
 2013-09-30 11:49:42 +02:00
Mirek Kratochvil 1488927e73 keyring: use cubehash for KeyID 
This creates incompatible keyrings, so keyring identifiers now have a
version string appended to them.
 2013-09-30 11:48:04 +02:00
Mirek Kratochvil 04bbfb49ce cubehash: some basic optimization 2013-09-30 07:21:36 +02:00
Mirek Kratochvil 72bcc1080a algos_sig: new cubehash-based algorithms 2013-09-29 20:53:24 +02:00
Mirek Kratochvil c65557724e algos_enc: new cubehash-based algorithms 2013-09-29 20:48:59 +02:00
Mirek Kratochvil d48665750a cubehash: remove forgotten debugging 2013-09-29 20:34:05 +02:00
Mirek Kratochvil 20b73496d7 cubehash: new hash algorithm 2013-09-29 19:52:46 +02:00
Mirek Kratochvil fae3342e18 main: remove now unnecessary licensing information 2013-09-15 15:58:01 +02:00
Mirek Kratochvil 414d203d18 tiger_hash: use crypto++ lib instead of bundle 2013-09-15 15:50:25 +02:00
Mirek Kratochvil 3af363faf3 rmd_hash: use crypto++ lib instead of bundle 2013-09-15 15:35:47 +02:00
Mirek Kratochvil 23e741c73e sha_hash: remove unnecessary stdint include 2013-09-15 15:35:15 +02:00
Mirek Kratochvil 47ccc8dcf0 algos_enc: forgotten stdint include 2013-09-15 15:33:47 +02:00
Mirek Kratochvil 95de70acbc sha_hash: use crypto++ lib instead of bundled one 2013-09-15 15:25:17 +02:00
Mirek Kratochvil 93dee89a4a algos_enc: do not collide with library min() macro 2013-09-15 15:09:01 +02:00
Mirek Kratochvil 54c9129ab1 actions: handle possible keyring saving failures 2013-09-12 13:16:57 +02:00
Mirek Kratochvil 09136ef046 actions: better keyring opening behavior 
This delays opening (and locking) the keyring until all input has been
processed, and all possible errors that would make a chance that the
keyring doesn't have to be opened have been checked for.

Whole purpose was to enable codecrypt to be chained with pipes in UNIX-y way,
like this primitive certificate creation:

(echo "At `date` I certify this is The Key:" ; ccr -pa -F "the key") | ccr -s
 2013-09-12 12:41:54 +02:00
Mirek Kratochvil d1421926f9 actions: neatified message for failed verification 2013-09-12 12:30:38 +02:00
Mirek Kratochvil 329953d575 keyring: better open/close behavior 2013-09-12 12:18:17 +02:00
Mirek Kratochvil 54e45bd3d4 style modifications because of newer astyle 2013-09-12 12:15:21 +02:00
Mirek Kratochvil 93cd8f377f keyring: squash opening and loading into one func 2013-09-12 12:14:58 +02:00
Mirek Kratochvil 60749aa4cf sencode: add integer size limits 2013-07-26 09:54:55 +02:00
Mirek Kratochvil 66d7d84b13 fix CXXFLAGS="-Wall" warnings 2013-07-23 15:58:02 +02:00
Mirek Kratochvil 36b68d90c1 std namespace cleaning 2013-06-21 20:35:59 +02:00
Mirek Kratochvil 1a2d25e555 main: correct licensing information 2013-05-27 18:23:20 +02:00
Mirek Kratochvil 9c9e49ae7b fmtseq: forgotten include 2013-05-25 20:03:42 +02:00
Mirek Kratochvil 9a268d231b forgotten styleup 2013-05-25 19:27:49 +02:00
Mirek Kratochvil b5ae7ca4dd algos_sig: added H=20 fmtseq variant 
...that provide around 1 million signatures per key. That seems more
than sufficient for human usage.

Note that generating the key takes 16 times longer than for already
present algorithms (that have H=16). On my computer, it is around 4
minutes for fmtseq128N20 and 16 minutes for fmtseq256N20.
 2013-05-25 19:19:22 +02:00
Mirek Kratochvil 0980ee827a sha2: fix the ugly warning 
memcpy() is better than dereferencing a type-punned-pointer for
simulating memcpy()
 2013-05-25 18:56:07 +02:00
Mirek Kratochvil 18449bdc8a arcfour: greatly increase discard parameters 
Okay, this thing got public so it's time to make the RC4 rugged. Not
that I'd know about something that would break current implementation,
but it's nice to at least do the recommended discard correctly.

We'll probably be adding better symmetric ciphers anyway.

Note that this is an incompatible change (again). FMTSeq private keys
will need to be replaced. Existing signature validity doesn't change.
Encrypted messages will not be possible to decrypt.
 2013-05-25 17:13:33 +02:00
Mirek Kratochvil 8c185c51ad mce_qd: compute check matrix on the fly 
codecrypt is usually invoked only with one prepare() and decrypt(), so
this basically saves several megs of memory and cuts needed computation
time in half.
 2013-05-18 10:52:57 +02:00