thread-based email index, search, and tagging https://git.notmuchmail.org/git/notmuch/atom?h=0.20.2 2015-05-31T16:56:10Z 2015-05-31T16:56:10Z David Bremner david@tethera.net 2015-05-31T16:56:10Z urn:sha1:5df1b1bb81b1983578e636ac62000e280d241139 Quoting Debian bug 787341 It failed to build on arm64: the last ten tests in T070-insert failed. What's happening here is that GDB is segfaulting in response to the "file" command. GDB on arm64 can be a bit buggy. However, the "file" command is redundant here as GDB has already got the file from the --args on the command line. 2015-04-08T14:43:55Z David Bremner david@tethera.net 2015-04-07T12:10:15Z urn:sha1:294bb6de20de7d9c83a45dca22888d557ece588a Apparently some systems actually have a directory called /nonexist[ae]nt. It's hard to fathom a good reason for that, but oh well. As long as we don't create such a directory inside the notmuch source tree, the new version should be more robust. 2015-04-08T14:43:55Z David Bremner david@tethera.net 2015-04-08T14:31:43Z urn:sha1:2e9ecb611a7f698808add756b2ddcbdfcd8b7c8b Apparently most of the misspellings are my fault. 2015-04-03T00:27:57Z Tomi Ollila tomi.ollila@iki.fi 2015-03-29T15:30:36Z urn:sha1:3d1483df12050dec8a2df95249576162a0e79302 When creating $THREADS data it may end of not having 'None' at all or the numbers in line output yields a loop. To avoid loop the value in current array index is set to 'None' so that if the same item is reached again the loop will end. Also empty string as next array index will end the loop. 2015-03-29T07:50:01Z David Bremner david@tethera.net 2015-03-29T07:39:46Z urn:sha1:703c8f9511c6f96023fb180b0e9d831e762befc9 Since notmuch_database_status_string can return NULL, passing it directly to fputs is not a good idea. 2015-03-28T23:34:15Z David Bremner david@tethera.net 2015-03-28T23:24:38Z urn:sha1:6a99505625785022d490207bdf6e3e29ed156c6a This second half of the error reporting tests actually uses the function notmuch_database_status_string to retrieve the last logged error 2015-03-28T23:34:15Z David Bremner david@tethera.net 2014-12-26T16:25:35Z urn:sha1:736ac26407914425a9c94e86616225292cf716dd This is not supposed to change any functionality from an end user point of view. Note that it will eliminate some output to stderr. The query debugging output is left as is; it doesn't really fit with the current primitive logging model. The remaining "bad" fprintf will need an internal API change. 2015-03-28T23:34:15Z David Bremner david@tethera.net 2014-12-27T18:12:49Z urn:sha1:84d3b15d251623cbb66e5eca7ddb8d61aa596d33 The compatibility wrapper ensures that clients calling notmuch_database_open will receive consistent output for now. The changes to notmuch-{new,search} and test/symbol-test are just to make the test suite pass. The use of IGNORE_RESULT is justified by two things. 1) I don't know what else to do. 2) asprintf guarantees the output string is NULL if an error occurs, so at least we are not passing garbage back. 2015-03-28T23:34:15Z David Bremner david@tethera.net 2015-03-14T08:50:55Z urn:sha1:83298fa0f8e7717a260c1f88edd7bf545709562d This first half of the tests is all that don't need to retrieve the the error string explicitly from the notmuch database structure. 2015-03-28T23:33:56Z David Bremner david@tethera.net 2015-03-14T08:02:03Z urn:sha1:1e69afe8af3512c45061cfde99c9f98e1f94e20c This is to limit the copy-pasta involved in running C tests. I decided to keep things simple and not try to provide an actual C skeleton. The setting of LD_LIBRARY_PATH is to force using the built libnotmuch rather than any potential system one.