Commit graph

7470 commits

Author SHA1 Message Date
Ori
2c96956b3b emacs: Add compatibility for org-msg.el
Enable `notmuch-company` with `org-msg`, which does not derive from
`message-mode` in composing. This change was first proposed in the
discussion here:
https://github.com/jeremy-compostella/org-msg/issues/53
2021-06-24 17:26:43 -03:00
David Bremner
5be9e024d1 doc: tweak hook configuration documentation.
Add a historical note, and hint to look below for more about hooks.
Capitalized the heading for consistency, removed blank line.
2021-06-23 08:42:48 -03:00
David Bremner
7406abf5e0 test/libconfig: make indexing more consistent
The other tests have converted to two digit decimal, the commit does
the same to the remaining test that was printing out indices in hex.
2021-06-21 06:57:17 -03:00
David Bremner
8dbd5deb8d Merge branch 'release' 2021-06-20 17:15:20 -03:00
David Bremner
86818626b3 NEWS: fix typo
Oops. Apparently our release checks do not check the spelling of
UNRELEASED (although Tomi's wiki conversion script does).
2021-06-20 16:59:30 -03:00
David Bremner
ce0fa835d3 NEWS: start NEWS for 0.32.2 2021-06-20 16:41:03 -03:00
David Bremner
dda50ede05 emacs: require seq for seq-some
Although the function dates from 2015, the autoload is newer. In
particular [1] it is not found in Emacs 25.3.

[1]: id:874ke85tqx.fsf@cgc-instruments.com
2021-06-08 20:12:33 -03:00
David Bremner
a19d2b0d2b test: check openssl prequisite for add_gpgsm_home
This is a fix for the test failures reported by Dan Čermák [1].

It is more robust to check for the prerequisite inside the function
that uses it, rather than in every test file that calls the function.

[1]: id:87k0n4fqgm.fsf@tethera.net
2021-06-08 17:15:46 -03:00
Tomi Ollila
6f0f83660e test: aggregate-results updates
notmuch-test will now call aggregate-results.sh with file list
that it compiles based on the test ran, and aggregate-results
will report failure is any of the test files are missing.

With this notmuch-test no longer has to exit in non-parallel
run if some test fail to write its report file -- so it works
as parallel tests in this sense.

Changed test_done() in test-lib.sh write report file in one write(2),
so there is (even) less chance it being partially written. Also,
now it writes 'total' last and aggregate-results.sh expects this
line to exist in all report files for reporting to be successful.

Added 'set -eu' to notmuch-test and modified code to work with
these settings. That makes it harder to get mistakes slipped
into committed code.
2021-06-07 20:16:33 -03:00
David Bremner
651a1b085b lib/message: use passed database for error handling
'message' should always be initialized if we reach here, but in case it
is not, we still want to be able to log an error message.
2021-06-05 15:41:28 -03:00
David Bremner
b0a11dbc38 lib/{open,message}: make some internal functions static
They are not used outside their file, so being extern seems like an oversight
2021-06-05 15:40:00 -03:00
Jani Nikula
87234a8ef3 emacs: update default mailing list archives for stashing links
Gmane web interface is long gone, remove it. Make MARC the new
default. Update LKML to Lore, where it already redirects anyway. Also
add Notmuch web archive.
2021-06-05 15:39:53 -03:00
David Bremner
2c85cc29bd test: add known broken test for indexing RFC822 group names
Austin Clements diagnosed this indexing problem in [1].

[1]: id:20130711215207.GR2214@mit.edu
2021-06-05 15:31:03 -03:00
Tomi Ollila
03366a3c5a emacs: fix declare-function definitions with related line breaks
- declare-function notmuch-unthreaded lacked file name
- declare-function notmuch-search had differently named last arg
  - note: check-declare-directory did not complain about that
- declare-function notmuch-search-show-thread without nil
- some functions declared to be in different file than those
  existed ("notmuch" -> "notmuch-lib")

- some related function/declare lines were (/are now) wider than
  80-columns; added line breaks (and proper indentation) there
2021-06-05 08:24:51 -03:00
David Bremner
d7ddfa0d0e emacs: drop setting mail-user-agent, and document how to set it
After some discussion [1], I decided it is better to make notmuch users
who rely on this behaviour customize mail-user-agent. This is
consistent with the behaviour of other emacs mail packages.

[1]: id:87k0nuhfrk.fsf@toryanderson.com
2021-06-04 20:05:31 -03:00
Tomi Ollila
572af27950 test: source $NOTMUCH_SRCDIR/test/test-lib-emacs.sh
Sourcing test-lib.sh will cd to TMP_DIRECTORY, so
relative path in $0 will not work in previous version
 . $(dirname "$0")/test-lib-emacs.sh

Now individual test scripts -- e.g. ./test/T310-emacs.sh
will work.
2021-06-03 09:29:27 -03:00
David Bremner
100106a45d Merge remote-tracking branch 'origin/release' 2021-05-31 20:26:31 -03:00
Daniel Kahn Gillmor
0478c0e077 NEWS/emacs: document changes in 0.32 that affect keybindings.
See list discussion in thread starting with
id:87h7ip2baq.fsf@fifthhorseman.net for more details.

Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
Amended-by: David Bremner (s/.emacs/Emacs configuration/)
2021-05-31 20:25:23 -03:00
Felipe Contreras
85ae2bcf56 ruby: use notmuch_exclude_t enum
It exists since 2013, let's allow it to be used in Ruby.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-23 09:05:33 -03:00
David Bremner
f44c83c083 test: add regression test for large number of deletions
Prior to 9ad19e4454 there was an unhandled Xapian exception when
reindexing after a large number of deletes.  This test was used for
bisection, and will subsequently serve as a regression test.
2021-05-23 08:31:36 -03:00
David Bremner
8cb30b08d8 perf-test: add test for reindexing after removal/renaming msgs
At this point it is a bit tricky to measure the performance increase
from the new message deletion code, since the same commit (9ad19e4)
that improved the performance also seems to have fixed a bug with an
uncaught Xapian exception triggered by this test.
2021-05-23 08:22:31 -03:00
Tomi Ollila
f2533b9e73 test: say_color() in one write(2)
say_color() used to call (builtin) printf (and tput(1) to stdout)
several times, which caused attempts to write messages with color
to have partial content (e.g. escape sequences) often intermixed
with other tests when parallel tests were run.

Now, with all output collected, then written out using one
printf, all strings with color print out correctly
((at least short) write(2)'s appear to write out "atomically").

While at it, used only one tput(1) execution to determine whether
color output works, and made bold/colors/sgr0 to tput(1) their
values once per test.
2021-05-23 08:05:15 -03:00
David Bremner
69c2c930ec test: quiet some extra debugging output
This output does not cause test failures, but may make it harder to
interpret the output.
2021-05-23 08:04:07 -03:00
David Bremner
119a41b9ed test: ignore debugging messages
Previously building with "-DDEBUG" broke the test suite in several places.
2021-05-23 08:03:02 -03:00
David Bremner
748352693c lib/thread: add common prefix to debug messages, join lines
This will simplify filtering these message, e.g. in the test suite.
2021-05-23 08:01:38 -03:00
Jani Nikula
1222cba7e4 doc: example command-line option reference
Example reference to a command-line option using the option role
reference. This creates a hyperlink in html, and the usual boldface
style in man page. This could be used throughout the man pages.
2021-05-22 16:47:32 -03:00
Jani Nikula
f2e2f2aa96 doc: use program and option directives to document options
Use the program and option directives to document the subcommand
options. This unifies a lot of option documentation throughout.

This also makes it possible to reference options with :option:`--foo`
(within .. program::) or :option:`subcommand --foo` (globally). This
is left for later work.

See https://www.sphinx-doc.org/en/master/usage/restructuredtext/domains.html#directive-program

Note: There is a lot of indentation change, but intentionally there is
no reflow. Using 'git diff -w' or 'git show -w' to ignore white space
changes makes this a very easy change to review.
2021-05-22 16:43:24 -03:00
Jani Nikula
574b2436ee doc: use envvar directive and role for environment variables
Make man1/notmuch.rst the single point of truth for describing notmuch
environment variables. Use the envvar directive for that, and
reference them with the envvar role.

Drive-by cleanup configuration file and hook directory search order
documentation.
2021-05-22 16:41:20 -03:00
Jani Nikula
ff4e81ac57 doc: cross-reference notmuch man pages with actual links
Add internal hyperlink targets for man pages and cross-reference them
using the any role reference. There are a number of alternatives to
accomplish this, but this seems like the combination that retains the
man page section number and the same boldface style in the man pages.

As a bonus, we get sanity checking on the links; for example
notmuch-search-terms.rst had a reference to notmuch-properties(1)
i.e. the wrong section.

The obvious semantic follow-up change would be to only have meaningful
"see also" references instead of having them all everywhere.
2021-05-22 16:38:56 -03:00
Tomi Ollila
8675be1d30 test: replace notmuch_passwd_sanitize() with _libconfig_sanitize()
notmuch_passwd_sanitize() in test-lib.sh is too generic, it cannot
work in many cases...

The more specific version _libconfig_sanitize() replaces it in
T590-libconfig.sh and the code that uses it is modified to output
the keys (ascending numbers printed in hex) so the sanitizer knows
what to sanitize in which lines...

"@" + fqdn -> "@FQDN" replacement is used as fqdn could
-- in theory -- be substring of 'USERNAME'.

'user -> 'USER_FULL_NAME replacement to work in cases where user
is empty -- as only first ' is replaced that works as expected.

In addition to ".(none)" now also ".localdomain" is filtered from
USERNAME@FQDN.

/dev/fd/{n} is not defined in posix, but it is portable enough
(if it weren't it is easy to fix -- now code looks clearer).
2021-05-22 11:51:29 -03:00
Jani Nikula
3baa61e0e5 doc: use manpage role references to external man pages
Using manpage role references generates helpful links in html
documentation, while retaining the same boldface style in the man
pages.

The external man page site is configurable. The Debian manpage site
seems like a good fit for Notmuch.
2021-05-22 09:56:52 -03:00
David Bremner
702635d5f6 Merge branch 'release' 2021-05-22 09:34:55 -03:00
Felipe Contreras
3dae253c4f ruby: improve compilation with CFLAGS
The ruby MakeMakefile generates a makefile that is suboptimal, which has
CFLAGS like this:

  CFLAGS   = $(CCDLFLAGS) -march=x86-64 -mtune=generic \
    -O2 -pipe -fno-plt -fPIC $(ARCH_FLAG)

This works as long as the user doesn't modify the Makefile.

Certain flags (namely -fPIC) need to be present regardless of what
CFLAGS are specified.

The Makefile should have done this instead:

  CFLAGS = -march=x86-64 -mtune=generic -O2
  override CFLAGS += $(CCDLFLAGS) -pipe -fno-plt -fPIC $(ARCH_FLAG)

Unfortunately they didn't, so we need to workaround their lack of
foresight.

We can simply add the necessary flags in the parent Makefile so everyone
is happy.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-22 09:17:21 -03:00
David Bremner
3f4de98e7c lib/n_d_index_file: re-use thread-id of existing message
This prevents the message document getting multiple thread-id terms
when there are multiple files with the same message-id.

This change shifts some thread ids, requiring adjustments to other tests.
2021-05-22 09:08:02 -03:00
David Bremner
349fc29803 test: add known broken test for duplicate thread-id terms
According to my bijection, this bug has been present since commit
411675a6ce in 2017. It is not completely clear what harm it causes in
regulary use, but it (at least) makes notmuch crash when compiled with
-DDEBUG_DATABASE_SANITY.
2021-05-22 09:02:58 -03:00
David Bremner
ff07183a02 CLI/new: check status of notmuch_message_maildir_flags_to_tags
This improves error reporting since it prints the specifics of the exception.
2021-05-19 09:00:08 -03:00
David Bremner
564a83077a perf-test/T00-new: fix backup creation.
"xargs tar cf backup.tar < $manifest" recreates the tar file with each
"batch" execed by xargs. In general this results in only a fraction of
the desired files being backed up.
2021-05-19 08:54:44 -03:00
Luis Henriques
0ab28068c1 doc: fix variable name in documentation
Variable 'notmuch-saved-searches-sort-function' does not exist;
'notmuch-saved-search-sort-function' is the correct name.

Signed-off-by: Luis Henriques <henrix@camandro.org>
2021-05-19 08:53:53 -03:00
Jani Nikula
59c953656d doc: fix man page build for Sphinx 4.x
Sphinx 4.0 changed the default value of man_make_section_directory
from False to True. We create the section directories and move the
files manually, so fix the immediate man build failure by disabling
the feature.

The Sphinx documentation on this [1] is confusing, and has the change
backwards. Git history says the default changed from False to True.

[1] https://www.sphinx-doc.org/en/master/usage/configuration.html#confval-man_make_section_directory
2021-05-19 08:52:48 -03:00
David Bremner
892f4e65ae perf-test/T00-new: fix backup creation.
"xargs tar cf backup.tar < $manifest" recreates the tar file with each
"batch" execed by xargs. In general this results in only a fraction of
the desired files being backed up.
2021-05-19 08:39:58 -03:00
Luis Henriques
e715ec9371 doc: fix variable name in documentation
Variable 'notmuch-saved-searches-sort-function' does not exist;
'notmuch-saved-search-sort-function' is the correct name.

Signed-off-by: Luis Henriques <henrix@camandro.org>
2021-05-19 08:35:25 -03:00
Jani Nikula
5197d3e11f doc: fix man page build for Sphinx 4.x
Sphinx 4.0 changed the default value of man_make_section_directory
from False to True. We create the section directories and move the
files manually, so fix the immediate man build failure by disabling
the feature.

The Sphinx documentation on this [1] is confusing, and has the change
backwards. Git history says the default changed from False to True.

[1] https://www.sphinx-doc.org/en/master/usage/configuration.html#confval-man_make_section_directory
2021-05-19 08:26:22 -03:00
Felipe Contreras
92454bc093 test: split emacs functionality to its own file
This way it's easier to identify the tests that do require emacs stuff.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:29:04 -03:00
Felipe Contreras
bba4788aec test: emacs: fix a couple of shellcheck complaints
In test-lib-emacs.sh line 20:
    test_require_external_prereq ${TEST_EMACS} || ret=1
                                 ^-----------^ SC2086: Double quote to prevent globbing and word splitting.

Did you mean:
    test_require_external_prereq "${TEST_EMACS}" || ret=1

In test-lib-emacs.sh line 21:
    test_require_external_prereq ${TEST_EMACSCLIENT} || ret=1
                                 ^-----------------^ SC2086: Double quote to prevent globbing and word splitting.

Did you mean:
    test_require_external_prereq "${TEST_EMACSCLIENT}" || ret=1

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:28:44 -03:00
Felipe Contreras
29ba523275 test: emacs: check for configured emacs
Commit d59d9c81 (test: Make the emacsclient binary user-configurable,
2012-11-27) modified the prereq check for the configured emacsclient,
but we probably want to do the same for emacs itself.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:28:33 -03:00
Felipe Contreras
cd9ec20a49 test: emacs: simplify missing dependencies check
No functional changes.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:28:23 -03:00
Felipe Contreras
4561f694e9 test: more style fixes
In order to fit the git coding style.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:28:01 -03:00
Felipe Contreras
645bea13cb test: trivial style cleanups
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:27:52 -03:00
Felipe Contreras
5f49e3421f ruby: new notmuch_rb_object_destroy() helper
The struct used to store the types (rb_data_type_t) contains a "data"
field where we can store whatever we want. I use that field to store a
pointer to the corresponding destroy function. For example
notmuch_rb_database_type contains a pointer to notmuch_database_destroy.

I cast that pointer as a notmuch_status_t (func*)(void *) and call
that function passing the internal object (e.g. notmuch_database_t).

Using the rb_data_type_t data we can call the correct notmuch destroy
function.

Therefore this:

  ret = ((notmuch_status_t (*)(void *)) type->data) (nm_object);

Is effectively the same as this:

  ret = notmuch_database_destroy (database);

The advantage of doing it this way is that much less code is necesary
since each rb_data_type_t has the corresponding destroy function stored
in it.

Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:25:14 -03:00
Felipe Contreras
9574fb6099 ruby: add all data types
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
2021-05-17 07:25:14 -03:00