Activity for Matthias Andree
-
Matthias Andree modified ticket #85
bogofilter recursively decodes RFC-2047 enc'words
-
bogofilter tokenizes parts from a MIME body's base64 stream
-
about to upload bogofilter 1.3.0.beta2.
-
Complete build options for Pthread API
-
Check return codes everywhere
-
bogofilter recursively decodes RFC-2047 enc'words
-
Migrated -> https://gitlab.com/bogofilter/bogofilter/-/issues/3
-
Assorted bug, and dependency fixes (C89, autotools, time_t, OpenSSL)
-
Support for PCRE2?
-
1.12.0 released.
-
confirmed & work in progress. Gitlab has a preliminary leafnode 1.12.0 state that contains PCRE2-capable code and further updates and cleanups. Before release, I would also like to add systemd configuration examples since (x)inetd is going away on the Linux distros, too. Fedora 34 has neither.
-
Hans, sorry for being blunt here. I am pretty much unimpressed by useless insistence on by-proxy arguments. I see nothing wrong with David's assessment, and there are more reasons to keep existing practice rather than change it. For one, bogofilter's personality has not changed, and if your Desktop client does not provide the features you expect of it, talk to its maintainer, not to the bogofilter maintainer. You have not presented any valid and corroborated reasons why things are different 12 years...
-
Hans, sorry for being blunt here. I am pretty much unimpressed by useless insistence on by-proxy arguments. I see nothing wrong with David's assessment, and there are more reasons to keep existing practice rather than change it. For one, bogofilter's personality has not changed, and if your Desktop client does not provide the features you expect of it, talk to its maintainer, not to the bogofilter maintainer. You have not presented any valid and corroborated reasons why things are different 12 years...
-
https://www.gnu.org/software/autoconf-archive/ and http://git.savannah.gnu.org/gitweb/?p=autoconf-archive.git;a=history;f=m4/ax_pthread.m4 would appear to be newer versions. Still not clear if that's needed in bogofilter which isn't threaded in itself, but possibly libraries require mutexes or threads.
-
bogofilter tokenizes parts from a MIME body's base64 stream
-
Turns out that the MIME boundaries are in violation of RFC2046 which does not comprise @ as valid character, so the lexer would not detect the MIME boundaries at all. Extending src/lexer_v3.l's definitions. Fixed by commit a88322e8.
-
I have installed flex-2.5.39 from source on Fedora 31 so I could reproduce this issue, and have concluded that the bug was fixed in this commit: commit cd33fc00802a75fe7b3b8a967bf879f7bc33c320 Author: Matthias Andree <matthias.andree@gmx.de> Date: Sat Feb 28 20:25:42 2015 +0000 Fix realloc() aborts on t.passthrough-truncation. Reported for Fedora 21 by Matt Garretson. bogofilter/NEWS | 8 ++++++++ bogofilter/src/lexer.c | 11 +++++++---- 2 files changed, 15 insertions(+), 4 deletions(-) Doran's remaining...
-
Complete build options for Pthread API
-
setting to pending
-
POP3 Proxy
-
I think without TLS, we should not add it yet.
-
A pop3 wrapper for kmail vr 0.3
-
This is code that should be maintained and published separately.
-
Replace BDB with LMDB
-
LMDB support was released last year in bogofilter-1.2.5.
-
Fix memory-leak in mime stack cleanup
-
part of bogofilter-1.2.5.
-
Mail crashes bogofilter
-
I have installed flex-2.5.39 from source on Fedora 31 so I could reproduce this issue, and have concluded that the bug was fixed in this commit: commit cd33fc00802a75fe7b3b8a967bf879f7bc33c320 Author: Matthias Andree <matthias.andree@gmx.de> Date: Sat Feb 28 20:25:42 2015 +0000 Fix realloc() aborts on t.passthrough-truncation. Reported for Fedora 21 by Matt Garretson. bogofilter/NEWS | 8 ++++++++ bogofilter/src/lexer.c | 11 +++++++---- 2 files changed, 15 insertions(+), 4 deletions(-) Doran's remaining...
-
bogofilter: yyinput result may exceed max_size
-
I think this is similar to #116 - unfortunately, the URL to the flex bug is now defunct, but I believe this bug was fixed with commit cd33fc00802a75fe7b3b8a967bf879f7bc33c320 (refs/bisect/bad) Author: Matthias Andree matthias.andree@gmx.de Date: Sat Feb 28 20:25:42 2015 +0000 Fix realloc() aborts on t.passthrough-truncation. Reported for Fedora 21 by Matt Garretson.
-
Bogofilter 1.2.5 release candidate
-
WRT Oracle Berkeley DB licensing, Oracle still provides 5.3 for download, which was not using the AGPLv3 license.
-
Repository converted to Git
-
Replace BDB with LMDB
-
Repository converted to Git
-
Reproduced and committed as fa012f83cb6cf1e6d4c104e4b0d554713fab0abd.
-
Fix memory-leak in mime stack cleanup
-
Fix out-of-bounds read in spanword
-
Fix out-of-bounds read in spanword
-
Reproduced and committed as cc8758f702f8b43b193047ff6e2b84968424b281.
-
Fix memory-leak in db_open()
-
Reproduced and committed as b2ac0f2916740ee0ee8d0847bc340b51e374f2b4. Fedora 30 has 4 more tests that fail with leaks under AddressSanitizer.
-
This is what I've committed.
-
Fix heap-buffer-overlow after decoding long tokens
-
I have collected all tests into a testcase, and then used the patch from your final comment. Git commit 0b400bdb5bedba8d8e237e23500418ff2193e6af.
-
Fix heap-buffer-overflow in yy_get_next_buffer
-
while I could not reproduce this on Fedora 30, I have committed your patch as e179b2f665d9ddeb9bf1ce212ba97e9698b60377
-
Committed to Git as 88c1eb51bddab37956c652ef4ee6bb2136a32e5b
-
Committed to Git, with memset() change, as 5f8a9050f404ebfc4380a1f4a605c78afb0e9586.
-
Fix heap-buffer-overflow/conversion buffer mixup
-
Committed to Git as 60a26fd84e6845a8f0b7c62dfa681bef1aabb678
-
Committed to Git, with memset() change, as 5f5339722e8da6c5f2377862338969d724daa129.
-
Fix gconv: Assertion `outbuf == outerr' failed
-
Committed to Git, with memset() change, as b983640f68188c9baf2a741964bb5409943d621a
-
Fix abort: Invalid buffer size, exiting.
-
committed to Git, c03bd8dd83daedf0215b94694827118a8a730b5f
-
Fix out-of-bound read when setting message id
-
Committed to Git, 4b7cfbf62938b47e54bcf2d44e7c292cf7af7ecd
-
Lock out SVN repo builds after Git conversion. Update README.
-
I wonder what the concept would look like, and if we went that way, if we could replace one layer of indirection with a new one, rather than adding one in between...
-
Replace BDB with LMDB
-
LMDB support has been contributed by Steffen Nurpmeso and merged into the Subversion repository, this ticket will be closed once a release has been made that contains it.
-
LMDB driver: stricter overflow checking.
-
any chances for windows version?
-
Bogofilter is a POSIX application that is supported on Linux and FreeBSD, and it appears that it has been ported for http://cygwin.com/.
-
Matthew, just a hint when you're dealing with automake-based projects, the headers need to be listed in Makefile.am, too. I figured that out when running make distcheck. Consequence would have been that oauth2.h would have been missing from the tarball. You don't see that when building from Git. I have now added oauth2.h to Makefile.am.
-
Merged, thank you.
-
[v2] Use correct output buffer size when calling to64frombits() from oauth2.c
-
[v4] add oauthbearer (gmail) support for imap and pop
-
[v3] add oauthbearer (gmail) support for imap and pop
-
Just minor style things that you were unable to know and that can be fixed later. Personally, I'll prefer to use enum over long #define lists for new constant sets such as the IPLUS_NONE/IPLUS_OAUTHBEARER, but that's no hindrance to merging your branch. I have merged the branch into master. I needed to make two adjustments to oauth2.c. One was a missing #include <stdlib.h> for free(), the other was because I had changed the to64frombits() interface in the meanwhile as an in-depth robustness fix....
-
Further datastore_lmdb update by Steffen Nurpmeso.
-
Merge lmdb-support branch developments.
-
Bugfix sent in by Steffen Nurpmeso.
-
Resynch with trunk.
-
Fix several warnings.
-
Make LMDB code valid C++, too.
-
Fix out-of-bounds memory read found with t.valgrind.
-
Fix t.bogoutil under BF_CHECKTOOL=valgrind.
-
Merge LMDB support from ^/branches/lmdb-support.
-
Resynch trunk.
-
Add news and build docs about LMDB.
-
Update sent by Steffen Nurpmeso, fixes self-test failures.
-
Remove one more obsolete workaround.
-
Sync with trunk, and drop 'do not write to read-only DB' shortcut.
-
Do not write special tokens to read-only DBs.
-
Forgo DB_NOSYNC when closing transactional DB,
-
Initial LMDB support, by Steffen Nurpmeso.
-
Branching for LMDB support.
-
Fix double word.
-
Silence GCC fallthrough warning.
-
confirmed
-
The underlying ticket against RT is at https://github.com/Beep6581/RawTherapee/issues/4161 - and note the ARW files expire after c. 58 days, and note that darktable does not need the lens manually chosen.
-
Update contrib/spamitarium.pl to 0.5.2.
-
Mention JIK's updates to contrib/spamitarium.pl.
-
Another update to spamitarium.pl by Jonathan Kamens.
-
Update spamitarium with Jonathan Kamens's changes.
-
Hi Matthew, First impressions after a very cursory skimming of a squashed merge of your commits: no major obstacles in the way of a merge, since the "web browser part" is kept external to the fetchmail core. the contrib script appears to hardcode google URLs somewhere, which might prevent it from working with Yahoo or IEEE or others. We might want to refactor URLs and other hardwired data to a central place so they can easily be adjusted. I wonder if the "each ctl its own copy" (of commit [1a4fd6])...
-
Hi Matthew, First impressions after a very cursory skimming of a squashed merge of your commits: no major obstacles in the way of a merge, since the "web browser part" is kept external to the fetchmail core. the contrib script appears to hardcode google URLs somewhere, which might prevent it from working with Yahoo or IEEE or others. We might want to refactor URLs and other hardwired data to a central place so they can easily be adjusted. I wonder if the "each ctl its own copy" (of commit [1a4fd6])...
1 >
