Commit Graph

1132 Commits

Author SHA1 Message Date
Alexandre Pujol aaa4637ed0
Track access/modification time of sensitive files
Collects the stats of tomb keys and tomb files then restore them when
Tomb exits. Can be extended to any file opened by Tomb. See #266
2018-09-23 22:21:05 +01:00
Jaromil cdd3c5804a updated documentation on KDF whitespace bug
fix #307
2018-09-23 12:05:02 +02:00
Jaromil e8919af867
Merge pull request #308 from AitorATuin/kdf_input
Fixes a problem reading passwords when using the kdf wrapper
2018-09-23 11:04:10 +01:00
Jaromil 27fd026cd5 documentation updates
also fix #319
2018-09-23 11:41:08 +02:00
Jaromil a27f15aada
Merge pull request #318 from roddhjav/master
Add test coverage for GPG subkeys support.
2018-08-20 13:03:50 +02:00
Alexandre Pujol c293aa7261
Add regression test for tomb 2.4 2018-08-19 23:08:48 +01:00
Alexandre Pujol 24fee7a076
Add test coverage for GPG subkeys support. See #317 2018-08-19 22:53:21 +01:00
Jaromil 7c8067fef3
Merge pull request #317 from bjonnh/patch-1
Add support for multiple GPG subkeys
2018-08-19 23:25:03 +02:00
bjonnh 18b7541a98
subkey_id for display must be out of _fingerprint
I did a typo, the ($gpg_id) should have been out of the _fingerprint call.
2018-08-18 17:41:37 -05:00
bjonnh b49a36a07b
Add support for multiple keys especially subkeys
Correcting issue #316
2018-08-17 14:03:33 -05:00
Jaromil 8d698ad46a
Merge pull request #309 from Ganondolf/gtk-tray-path
Fix mounted volume default path
2018-02-18 11:10:22 +01:00
Ganondolf f290904f71
Fix mounted volume default path 2018-02-17 16:23:26 +01:00
Jaromil 68a9589925 set ownership after dig and forge 2018-02-11 21:41:58 +01:00
ATuinDev 3cb8ebefd3
Don't impose any limitation in password size
Remove the hardcoded buffer with dynamic memory to get password input
(uses now calloc / realloc)
2018-02-04 20:53:35 +01:00
ATuinDev 35cf8572bf
Rename variable `j` to something more descriptive
Remove `+ 1` in conditions, easier to understand using the comparators
2018-02-03 23:24:20 +01:00
ATuinDev a66224d549
Run extras/kdf-keys tests in travis 2018-02-03 21:23:07 +01:00
ATuinDev 510c8f6430
Add check for password lens 2018-02-03 21:05:56 +01:00
ATuinDev 5c419b3117
Add explicit comment before reading the password 2018-02-03 19:20:01 +01:00
ATuinDev 257e5ee99a
whitespaces are now part of the password
NULL character can also be part of the password. There was a failing
test that was checking exactly that (now all the tests pass).
2018-02-03 19:16:20 +01:00
ATuinDev bce58cae3e
Add test testing that whitespaces are used by kdf wrapper
Adds a new test that assures that spaces are part of the passwords
2018-02-03 19:12:05 +01:00
Jaromil 3440a32839
Merge pull request #305 from dyne/assuan_deduplication
put assuan commands to pinentry into a single function
2018-02-02 16:47:28 +01:00
Jaromil f5ceddc0b7 put assuan commands to pinentry into a single function 2018-01-29 09:47:42 +01:00
Jaromil f4913e0744
Merge pull request #303 from sargo-devel/master
Added pinentry-qt5 support
2018-01-29 09:38:47 +01:00
SargoDevel 66ade86441 Added pinentry-qt5 support 2018-01-28 21:49:21 +01:00
Jaromil d0805084a2 added doc to avoid logging of invokations in syslog
see issue #302 and thanks to @stevesbrain
2018-01-23 09:38:52 +01:00
Jaromil f42a4c2a0b
Merge pull request #301 from Narrat/small_things
Small things
2018-01-09 10:42:40 +01:00
Narrat 08ca0a8eef list_gnupg_ciphers: Remove check on gpg
This function is called after _ensure_dependencies(), which would bail out if gpg is not found
2018-01-07 21:56:19 +01:00
Narrat 5e3b0dec84 list_gnupg_ciphers: be language agnostic
The old awk implementation always worked on lines beginning with 'Ciphers:' until it found 'Hash:'.
This fails for locales where a respective gnupg2.mo entry exists (Example: Ciphers in german is translated as Verschlü.:).
This is replaced by pointing awk on a specific line, which is for gpg1 and gpg2 the same. Work is done until awk stumbles up on a line which marks a new section (marked by keyword and :)

This closes #299
2018-01-07 21:40:32 +01:00
Narrat e15c58dfd7 list_gnupg_ciphers: Pipe everything into /dev/null
Firstly the printed binary path is in the wrong place. Reading the text, one assumes Ciphers coming next.
Secondly it doesn't make sense to check there for a missing gnupg installation. Before calling list_gnupg_ciphers(), there is a direct call for gpg --version. If that fails the whole text is scrambled and no error reported

Dropping the output from which allows to remove the space from printing the ciphers. The text is correctly aligned now
2018-01-07 21:03:21 +01:00
Narrat 61fdab85be Show only version of pinentry
pinentry --version invocation includes License information.
As the same applies for gpg, and the information is not displayed there, we should the same with pinentry.
And tomb doesn't deal with the gpg sourcecode in any way.

This closes #300
2018-01-07 20:03:04 +01:00
Jaromil 74689ea484 tagging the release 2018-01-03 20:34:18 +01:00
Jaromil 7951645db5 switch code indentation to hard tabs
also update code guidelines.
2018-01-03 20:27:14 +01:00
Jaromil 70abf31bab documentation updates for a new release 2018-01-03 19:53:35 +01:00
Jaromil ae78659efb
Merge pull request #297 from dyne/restore-chown
restored change of ownership on tomb's contents
2018-01-03 19:42:52 +01:00
Jaromil ab3044c6f5 restored change of ownership on tomb's contents
this reverts commit 843b7fdfc4
and refers to various issues, among them #268

on the long term its easy to realise how this is a usability feature for most
users, so we just provide a new '-p' flag to preserve ownership on open.
2018-01-03 18:34:50 +01:00
Jaromil 214ec8ecbd
Merge pull request #298 from roddhjav/fix-version-check
Fixes: version check & tests
2018-01-03 16:36:16 +01:00
Alexandre Pujol cec0c01b3e
Test: remove useless cleanup 2018-01-03 14:50:00 +00:00
Alexandre Pujol c7dc379c0d
Tests: support non-standard zsh location. See #283 2018-01-03 14:49:08 +00:00
Alexandre Pujol b20daeea6f
Fix: use is-at-least function to check program version 2018-01-03 14:46:23 +00:00
Jaromil bc963cd1ae fix for correct execution of exec-hooks
this was broken in the latest release and fixes #271
2018-01-03 14:37:15 +01:00
Jaromil 3b1759b2b3 fix for hidden files without an extension as tombs
fix #147 introducing an extra check on TOMBNAME that, if returned
empty by the first transormation that removes the last .extension,
then is filled with the full TOMBFILE name without any transformation
2018-01-03 13:43:05 +01:00
Jaromil c8616787f7 added a comment in manpage about resize fail and restore
documentation according to #269
2018-01-03 11:14:21 +01:00
Jaromil 31a78de23f removed known_bugs entry according to #276
This was confusing. Later found time to verify facts. We use 512 bytes
long keys (4096 bits) so AES-256 was always used in XTS mode.
2018-01-03 11:09:33 +01:00
Jaromil 60b72ad91f documentation and version updates
findmnt version shown
2018-01-03 10:44:27 +01:00
Jaromil cf93551efa
Merge pull request #296 from dyne/use-findmnt
Replace 'mount -l' with 'findmnt'
2018-01-03 10:38:48 +01:00
Jaromil 9b1d1891cc
Merge pull request #290 from parazyd/use-shred-instead-wipe-patch
use shred instead of wipe
2018-01-03 10:03:50 +01:00
Jaromil b72d67618b restore square parens detection in list_tomb_mounts() 2018-01-02 15:35:32 +01:00
Jaromil 3f107f9d31 small fix for usage of findmnt --raw flag 2018-01-02 14:04:53 +01:00
Jaromil 3593721967 drop zsh/regex module in favor of =~
As debated in issue #282 Zsh introduced a bug in v5.3.1 which briefly affected
our mechanism for closing tombs. The bug is fixed, but while investigating the
issue @aude realised there can be a better way to apply this regex for the
detection of mounted volumes on distro dependent /run/media/$USER paths.

This change drops usage of the regex optional module in Zsh to use the built-in
=~ comparison and improves the match using round parenthesis. It may fix the
close command on some distributions.
2018-01-02 14:04:53 +01:00
Victor Calvert 09ff889c1d Replace mount with findmnt
this fixes a mount-related functionality (finding the volume label) in new
versions of util-linux, that since v2.30 does not list anymore volume labels
with its mount -l command. Since findmnt needs sudo to list labels, this also
introduces the need for sudo in more commands: is_valid_tomb(), list, index and
search. The issue was examined in PR #283 and this is a rebase of it.
2018-01-02 13:39:35 +01:00