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

470 commits

This branch
This branch
All branches
Author SHA1 Message Date
Tom Li f7ae29ccc2
keyring.cpp: fix std::vector access violation 
Signed-off-by: Tom Li <tomli@tomli.me>
 2019-01-09 23:09:54 +08:00
Tom Li 7021f6c734
sc.h: correct access violation in load_key_vector(), close #2. 
In load_key_vector(), the program passes a std::vector<byte> to
a C-style function, load_key (const byte*begin, const byte*end)
by creating references

    load_key (& (K[0]), & (K[K.size()]));

However, accessing the one-past-the-last element in a std::vector
via [] is not allowed in C++, it triggers an assertion failure.

    Assertion '__builtin_expect(__n < this->size(), true)' failed.

In this commit, we use K.data() and K.data() + K.size() to expose
the underlying pointers and pass them to the C function.

Signed-off-by: Tom Li <tomli@tomli.me>
 2019-01-02 11:42:04 +08:00
Mirek Kratochvil 4441f8e514 document existing python bindings 2018-10-18 14:37:48 +02:00
Mirek Kratochvil c3f926b759 add windows port info 2018-10-03 10:46:10 +02:00
Mirek Kratochvil 59af6b2ae8 produce a conf-time warning if using getpass(3) 2018-05-28 14:11:57 +02:00
Mirek Kratochvil f0ecdcd9c9 autogen.sh: produce better Makefile.am 2018-05-28 14:11:02 +02:00
Mirek Kratochvil e615d48732 Merge 'amynka/cfg', better autotools autotooling 2018-05-28 13:46:17 +02:00
amynka@gentoo.org 1b1b07d9e6 Buildsystem improvements 
- Adding Makefile.am
- Fixing configure.ac hacks
- Cleaning configure.ac
 2018-05-21 15:51:46 +02:00
Mirek Kratochvil 58c8beefd2 Merge branch 'cmake' 2018-02-01 10:21:51 +01:00
Mirek Kratochvil fb616a748e CMakeLists commentary 2018-02-01 10:21:31 +01:00
Mirek Kratochvil d7362a0e5b keep .gitignore sorted and complete 2018-02-01 10:11:07 +01:00
tearsofphoenix 4a00aec0b9 add support for CMake 
Note that the CMakeLists.txt file is not included in autotools dist package to
avoid any packaging confusion. `configure.ac` is still the primary
build-configuration information source. (Btw isn't there a way to build
CMakeLists.txt from autotools?)
 2018-02-01 10:10:57 +01:00
Mirek Kratochvil 91b12452d6 update docs 2017-12-15 16:26:26 +01:00
Mirek Kratochvil 25cb1f081d contribution guide 2017-11-22 09:32:40 +01:00
Mirek Kratochvil 0e541c4fbf version bump to 1.8 
bemp.
 2017-10-31 21:42:06 +01:00
Mirek Kratochvil 66d5488133 privfile: add forgotten sources 2017-10-31 21:34:38 +01:00
Mirek Kratochvil 8a55068d89 configure.ac: check for pkg-config when using it 2017-10-27 10:41:40 +02:00
Mirek Kratochvil df3ed069d7 README: someone added a comma here 2017-10-23 22:23:19 +02:00
Mirek Kratochvil 65c2bcc546 man: document the -u defaulting 2017-10-23 16:33:04 +02:00
Mirek Kratochvil 798de9f70a update docs 2017-10-23 14:56:02 +02:00
Mirek Kratochvil 76fb3b059d main: default -u from CCR_USER env 2017-10-23 14:33:47 +02:00
Mirek Kratochvil c27285e750 actions: only show 40 bits of keyID by default 
It must be enough for everyone! It also copypastes better and is short enough
not to provide a false feeling of security.
 2017-10-23 14:32:10 +02:00
Mirek Kratochvil 5dceb7c07d styleup 2017-10-23 14:17:35 +02:00
Mirek Kratochvil 4f2680134e privfile: common sk/pk file creation/saving code 2017-10-23 14:14:26 +02:00
Mirek Kratochvil 7ec0823834 keyring: fix forgotten check 2017-10-23 14:13:49 +02:00
Mirek Kratochvil 104ee12951 implement private key locking 
Included:
- gazillion changes in actions
- keyring decodes privkey structures lazily, if not needed they will
  pass by as strings
 2017-10-22 23:10:23 +02:00
Mirek Kratochvil 7b0bc06d45 add forgotten seclock files 
whew boy.
 2017-10-22 21:12:57 +02:00
Mirek Kratochvil 578691f45e secret-locking on symmetric keys 2017-10-22 21:00:58 +02:00
Mirek Kratochvil c0770926e9 configure.ac: it was a bit too verbose 2017-10-22 21:00:58 +02:00
Mirek Kratochvil 3f378e14d7 readme: typo 2017-08-04 08:52:50 +02:00
Mirek Kratochvil 0d772da18c Merge tag 'v1.7.6' 2017-07-12 16:24:59 +02:00
Mirek Kratochvil 4f4e4f4df9 minor typos and styles 
Actually, debian lintian told me that:

I: codecrypt: spelling-error-in-manpage
   usr/share/man/man1/ccr.1.gz overriden overridden

Isn't that cute?
 2017-07-12 16:10:59 +02:00
Mirek Kratochvil 33a9c14661 Merge branch 'master' of github.com:exaexa/codecrypt 2017-07-12 15:08:09 +02:00
Mirek Kratochvil 8dd84892ef README: update debian information 2017-07-12 15:06:33 +02:00
Mirek Kratochvil e5b8b4bc82 version bump to 1.7.6 2017-07-10 12:40:37 +02:00
Mirek Kratochvil cc9b6b5858 base64: simplify debase64ing a bit 2017-07-10 11:17:34 +02:00
Mirek Kratochvil be8c3faa55 arcfour: register keyword is deprecated now 2017-07-10 11:13:10 +02:00
Mirek Kratochvil ad40fbe5dc keyring: write backup before actually touching keys 2017-06-26 14:30:07 +02:00
Mirek Kratochvil dc3a874cd5 generator: allow own PRNG seed source 2017-06-26 14:28:54 +02:00
Mirek Kratochvil d53586d582 remove obsolete qd_utils header 2017-04-25 12:15:06 +02:00
Mirek Kratochvil d7dc743016 Merge branch 'master' of github.com:exaexa/codecrypt 2016-10-18 21:02:18 +02:00
Mirek Kratochvil ac2047a915 Merge pull request #19 from mapmeld/mac 
run glibtoolize on OSX
 2016-10-18 21:01:13 +02:00
Nick Doiron cadaadac73 run glibtoolize on OSX 2016-10-18 13:12:33 -04:00
Mirek Kratochvil 87337db922 README: add explicit key distribution notice 2016-06-13 08:50:29 +02:00
Mirek Kratochvil 0bd9ab6d54 hashfile: avoid being fine with informative-grade hashes only 2016-04-28 21:58:52 +02:00
Mirek Kratochvil 91c5e9ba35 gitignore: remove obsolete debian/ entries 2016-04-28 21:57:56 +02:00
Mirek Kratochvil fdddafda81 version bump to 1.7.5 2016-04-28 13:16:13 +02:00
Mirek Kratochvil 8e608b4d8d mce_qd: remove obsolete MCE-QD encryption 2016-04-28 13:14:24 +02:00
Mirek Kratochvil 77c626aab2 man: update documentation to reflect on cubehash change 2016-04-28 13:01:34 +02:00
Mirek Kratochvil 213331903e cubehash_impl: fix cubehash implementation mistake 
Well, there's a reason for that test vectors are published on wikipedia.

Although this looks scary (like writing past array bounds), cubehash B
parameter is in all cases smaller than 63 (which is the first B value where
this would write behind the array), so no harm is done. For similar reason, the
"misimplemented" cubehash was cryptographically correct (i.e. without
cryptographic weakness), only implemented differently and producing different
results than those prescribed by the standard.

Practical implications of changing the hash functions are:
- everyone gets a new KeyID
- FMTSeq keys that used cubehash are invalid now, users are forced to generate
  new ones
 2016-04-28 12:49:13 +02:00