Tomi Ollila [Sun, 2 Aug 2015 14:48:14 +0000 (17:48 +0300)]
emacs: prefer notmuch-emacs-version in User-Agent: header
Now that we have `notmuch-emacs-version' defined in notmuch emacs MUA
use that as a part of User-Agent: header to provide more accurate
version information when sending emails.
In case some incomplete installation of notmuch emacs MUA is used and
`notmuch-emacs-version' is defined as "unknown" then fall back to ask
version info from cli (as it used to be before this commit).
Requiring notmuch-version[.elc] and if that is missing setting
"fallback" notmuch-emacs-version (to "unknown") was moved from
notmuch.el to notmuch-lib.el as notmuch-mua.el (which provides
User-Agent: information) require's the latter.
David Bremner [Sun, 2 Aug 2015 06:40:55 +0000 (08:40 +0200)]
test: use the python interpreter in sh.config
The configure script chooses "python" if both python and python{2,3}
exist exists, so this could change the version of python used to run
the test suite.
The checking for ${NOTMUCH_PYTHON} in the test suite is arguably
over-engineering, since the configure step will fail if it can't find
it.
Mark Walters [Sat, 13 Jun 2015 08:08:09 +0000 (09:08 +0100)]
emacs: add a filter option to show
Show the current thread with a different filter (i.e., open messages
in the thread matching the new query).
Bound to 'l' for "limit".
Note that it is not the same as filter in search mode as it replaces
the existing query rather than ANDing with it (but it does keep the
thread-id part of the query).
Mark Walters [Sat, 13 Jun 2015 08:08:08 +0000 (09:08 +0100)]
emacs: split notmuch-show-apply-state
Separate out a notmuch-show-goto-msg-id sub-function from
notmuch-show-apply-state. There should be no functional change but the
next patch will call the new function.
The CLI (and bindings) code should really be updated to use the new
status-code-returning versions. Here are some warnings to prod us (and
other clients) to do so.
Austin Clements [Fri, 5 Jun 2015 17:28:33 +0000 (19:28 +0200)]
lib: Only sync modified message documents
Previously, we updated the database copy of a message on every call to
_notmuch_message_sync, even if nothing had changed. In particular,
this always happens on a thaw, so a freeze/thaw pair with no
modifications between still caused a database update.
We only modify message documents in a handful of places, so keep track
of whether the document has been modified and only sync it when
necessary. This will be particularly important when we add message
revision tracking.
David Bremner [Mon, 27 Jul 2015 18:42:32 +0000 (20:42 +0200)]
build: add "set -eu" to version script generation
It turns out that on certain systems like FreeBSD, c++filt is not
installed by default. It's basically OK if we fail the build in that
case, but what's really not OK is for the build to continue and
generate bad binaries.
Previously documentation was build automatically if sphinx/doxygen
executable were found. The switch is used to unconditionally disable
sphinx/doxygen detection and therefor, building of documentation
(including man pages).
Daniel Schoepe [Tue, 16 Jun 2015 21:47:20 +0000 (23:47 +0200)]
Fix documentation for notmuch--tag-hook functions
The second argument to notmuch-tag is now called tag-changes, but the
documentation for notmuch-before-tag-hook and notmuch-after-tag-hook
still used the old argument name `tags'. This resulted in broken hooks
when following the documentation.
Mark Walters [Tue, 9 Jun 2015 09:51:07 +0000 (10:51 +0100)]
emacs: tree: mark read change
The mark read code for tree mode did not get updated in the recent
changes. This updates it to match. Since the user can customize the
mark read logic we just call the show logic in the message pane.
David Bremner [Mon, 1 Jun 2015 07:09:01 +0000 (09:09 +0200)]
build/ruby: use notmuch configure script values for shared lib
This is supposed to help build on systems like MacOS with different
conventions for naming shared libraries. We have already computed the
relevant names, so doing it again in ruby seems like a bad idea.
David Bremner [Mon, 1 Jun 2015 07:08:59 +0000 (09:08 +0200)]
build: integrate building ruby bindings into notmuch build process
Because ruby generates a Makefile, we have to use recursive make.
Because mkmf.rb hardcodes the name Makefile, put our Makefile{.local}
in the parent directory.
David Bremner [Mon, 8 Jun 2015 06:02:22 +0000 (08:02 +0200)]
lib: reject relative paths in n_d_{create,open}_verbose
There are many places in the notmuch code where the path is assumed to be absolute. If someone (TM) wants a project, one could remove these assumptions. In the mean time, prevent users from shooting themselves in the foot.
Update test suite mark tests for this error as no longer broken, and
also convert some tests that used relative paths for nonexistent
directories.
Tomi Ollila [Tue, 2 Jun 2015 14:43:45 +0000 (17:43 +0300)]
release-checks: check that git working directory is clean
Before release check that there are no uncommitted changes and
that there are no files in working directory that possibly should
have been added to the repository.
Amended by db: remove --ignored, since that seems like too much
trouble.
David Bremner [Sat, 23 May 2015 20:28:47 +0000 (22:28 +0200)]
test: redirect man output to /dev/null
In the case the these tests fail, they generate a bunch of output;
this output is not very interesting because it is just the successful
output of a man page. It does however make it hard to see what tests are actually failing, even with NOTMUCH_TEST_QUIET
David Bremner [Sun, 5 Apr 2015 20:34:55 +0000 (05:34 +0900)]
cli: add standard option processing to config, help and setup
In particular this fixes a recently encountered bug where the
"--config" argument to "notmuch setup" is silently ignored, which the
unpleasant consequence of overwriting the users config file.
David Bremner [Sun, 5 Apr 2015 13:24:42 +0000 (22:24 +0900)]
cli: ignore config argument of notmuch_help_command
We call it with NULL at one point anyway, so it needs to work with
NULL. Since the only place we use talloc is right before exec, there
is no harm in always using NULL.
Charles Celerier [Sat, 27 Dec 2014 09:18:27 +0000 (10:18 +0100)]
emacs: Added "is:<tag>" style completion to notmuch-read-query.
The notmuch-search-terms man page states that "tag:<tag>" is equivalent
to "is:<tag>". Completion for "is:<tag>" style searches is now supported
in the Emacs interface.
Amended by David Bremner: combine lexical-let and let into
lexical-let*
David Bremner [Tue, 19 May 2015 20:48:24 +0000 (22:48 +0200)]
configure: Add sanity checking for environment variables
Passing in environment variables incompatible with the compiler may
cause other parts of the configure script to fail in hard to
understand ways, so we abort early.
David Bremner [Tue, 7 Apr 2015 12:10:15 +0000 (21:10 +0900)]
test: move nonexistent directory to something under notmuch's control.
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.
Mark Walters [Sat, 28 Mar 2015 11:08:59 +0000 (11:08 +0000)]
emacs: show: hide large text attachments by default
notmuch-show can be slow displaying large attachments so hide them by
default. The default maximum size is 10000 bytes/characters but it is
customizable.
Note that notmuch-show-insert-bodypart is also called from the reply
code so we need to be a little careful.
Tomi Ollila [Sun, 29 Mar 2015 16:37:34 +0000 (19:37 +0300)]
notmuch-emacs-mua: non-forking escape () usage with backslash '\' escape
Use the printf -v convention to give output variable as argument
to escape () function so no subshell needs to be executed for
escaping input. The '-v' option to escape () is just syntactic
sugar for better understanding.
Also, backslash is now escaped with another backslash for emacs. This
ie especially important at the end of string.
`echo` is no longer used to write escaped output -- it might interpret
the escapes itself.
W. Trevor King [Sun, 22 Mar 2015 22:51:41 +0000 (15:51 -0700)]
nmbug-status: Use 'show-ref --heads' for loading configs
When loading configs from Git, the bare branch name (without a
refs/heads/ prefix or similar) matches all branches of that name
(including remote-tracking branches):
David Bremner [Fri, 26 Dec 2014 17:34:49 +0000 (18:34 +0100)]
lib: eliminate fprintf from _notmuch_message_file_open
You may wonder why _notmuch_message_file_open_ctx has two parameters.
This is because we need sometime to use a ctx which is a
notmuch_message_t. While we could get the database from this, there is
no easy way in C to tell type we are getting.
David Bremner [Fri, 26 Dec 2014 16:25:35 +0000 (17:25 +0100)]
lib: replace almost all fprintfs in library with _n_d_log
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.
David Bremner [Sat, 27 Dec 2014 18:12:49 +0000 (19:12 +0100)]
lib: add "verbose" versions of notmuch_database_{open,create}
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.
David Bremner [Sat, 14 Mar 2015 07:32:22 +0000 (08:32 +0100)]
test: Add two tests for error output from notmuch_database_open
This is arguably testing the same thing twice, but in the brave new
future where we don't use printf anymore, each subcommand will be
responsible for handling the output on it's own.
W. Trevor King [Sat, 10 May 2014 19:16:38 +0000 (12:16 -0700)]
nmbug-status: Clarify errors for illegible configs
Carl Worth pointed out that errors like:
$ ./nmbug-status
fatal: Not a git repository: '/home/cworth/.nmbug'
fatal: Not a git repository: '/home/cworth/.nmbug'
Traceback (most recent call last):
File "./nmbug-status", line 254, in <module>
config = read_config(path=args.config)
File "./nmbug-status", line 73, in read_config
return json.load(fp)
File "/usr/lib/python2.7/json/__init__.py", line 290, in load
**kw)
File "/usr/lib/python2.7/json/__init__.py", line 338, in loads
return _default_decoder.decode(s)
File "/usr/lib/python2.7/json/decoder.py", line 366, in decode
obj, end = self.raw_decode(s, idx=_w(s, 0).end())
File "/usr/lib/python2.7/json/decoder.py", line 384, in raw_decode
raise ValueError("No JSON object could be decoded")
ValueError: No JSON object could be decoded
are not particularly clear. With this commit, we'll get output like:
$ ./nmbug-status
fatal: Not a git repository: '/home/wking/.nmbug'
No local branch 'config' in /home/wking/.nmbug. Checkout a local
config branch or explicitly set --config.
which is much more accessible. I've also added user-friendly messages
for a number of other config-parsing errors.
Jani Nikula [Tue, 14 Oct 2014 16:32:54 +0000 (19:32 +0300)]
lib: make notmuch_query_count_messages explicitely exact
The default is actually exact if no checkatleast parameter is
specified. This change makes that explicit, mainly for documentation,
but also to be safe in the unlikely event of a change of default.
[ commit message rewritten by db based on id:87lho0nlkk.fsf@nikula.org
]
Jani Nikula [Sat, 7 Mar 2015 12:44:42 +0000 (14:44 +0200)]
completion: complete addresses in from:/to: search terms
Use the new notmuch address command to do completion for addresses in
from: and to:. Use --output=sender for both for efficiency, even
though --output=recipients would be more accurate for to: prefix
completion.
Jani Nikula [Sun, 8 Mar 2015 16:18:55 +0000 (18:18 +0200)]
cli: add support for notmuch command --help
Recognize 'notmuch command --help' at the top level as a special case,
and show help for the command. Note that for simplicity, --help is
only recognized as the first option for the subcommand.
David Bremner [Tue, 9 Dec 2014 20:01:11 +0000 (21:01 +0100)]
CLI: make gpg binary used by libgmime configurable.
Previously we set up a way for the top level notmuch command to choose
which gpg binary was invoked by libgmime. In this commit we add the
(mostly boilerplate) code to allow the notmuch-config command to read
and write this path, and use it in the appropriate struct.
David Bremner [Tue, 9 Dec 2014 20:01:10 +0000 (21:01 +0100)]
CLI: set up infrastructure to make path to gpg configurable.
GMIME takes a path to gpg, but we hardcode that path. In this commit
we set up argument passing and option storage to allow this path to
specified in the top level notmuch command.
David Bremner [Sat, 7 Mar 2015 07:31:14 +0000 (08:31 +0100)]
python: replace hardcoding of SONAME version
Failing to update this string in globals.py causes failures when the
SONAME changes. In order to hopefully reduce the number of such
errors, automate the process of setting the SONAME in the python
bindings.
Jani Nikula [Sat, 7 Mar 2015 14:30:19 +0000 (16:30 +0200)]
notmuch-emacs-mua: remove -C to keep short options compatible with mutt
Notmuch uses long options exclusively all around. The short options in
notmuch-emacs-mua are intentionally just a compatible subset of
mutt(1). Keep it this way, if only to make documenting the fact easy!
The Notmuch style --client long option remains, of course.
Jani Nikula [Mon, 2 Mar 2015 17:47:04 +0000 (19:47 +0200)]
parse-time-string: fix setting and rounding of seconds
If seconds are not specified in the string to be parsed, they're not
set according to the reference time (in the no rounding case) nor
rounded properly (in the rounding up cases). Fix this.
The bug caused searches such as date:10:30..10:30 to match messages
with date exactly 10:30:00 only, and not in range 10:30:00..10:30:59
(inclusive) as documented.
Note that date searches referring "noon" or "5pm" will still be
interpreted as exact to the second.
Jani Nikula [Mon, 2 Mar 2015 17:47:03 +0000 (19:47 +0200)]
test: extract and flag second rounding tests as broken
Extract and add some more tests relating to rounding seconds into a
separate subtest, and flag it as broken. If seconds are not specified,
the time is not set or rounded according to the interface defined in
parse-time-string/parse-time-string.h. Instead, seconds are always set
to 00 in reality, which is broken.
Jani Nikula [Mon, 2 Mar 2015 17:47:01 +0000 (19:47 +0200)]
test: improve the time parsing tests
Set the reference time to 12:13:14 instead of 11:11:00 to ensure hours
and minutes are not mixed up, and seconds are really set instead of
defaulted to 00.
We may as well update the list of commands while we decide if it's
worth automating the process. Note that there is a bit more noise than
expected because it alphabetizes all of the commands with their own
man pages.