Follow-up of 6273966d, now that sfsexp 1.4.1 doesn't rely on globals
anymore by default (https://github.com/mjsottile/sfsexp/issues/21).
This simply defers the initial query generation to use the thread-safe
helper (xapian_query_match_all) instead of Xapian::Query::MatchAll.
According to the now deleted commentary, the hack of using run-at-time
was needed for Emacs 24. It seems to be no longer needed for Emacs
28.2, and removing it makes further changes to the code simpler.
Sometimes merging is not what we want with tags; in particular it
tends to keep tags in the local repo that have been removed elsewhere.
This commit provides a new reset command; the reset itself is trivial,
but the work is to provide a safety check that uses the existing
--force and git.safe_fraction machinery.
37c022ae ("Use `without-restriction` in `with-temporary-notmuch-message-buffer`", 2024-03-14)
introduced `delete-line` in a test, but this is Emacs 29 and above only.
Replace it with its (almost) definition.
37c022ae ("Use `without-restriction` in `with-temporary-notmuch-message-buffer`", 2024-03-14)
introduced a fix for draft saving in a way which is supported on Emacs
29 and above only. Replace this with a construct which we have used
before, so that we keep the same compatibility level.
This ensures that the temporary copy of the current message-mode
buffer is whole and not limited by a current restriction.
An example of such restriction is the default one established by
message-mode when composing a reply, that hides the References,
In-Reply-To and similar headers.
In [1], Jakub Wilk observes that the current behaviour is confusing
since it looks like there are two mailboxes in From, while in fact
there is only one. It seems to me that notmuch should at least quote
the display-name part of a mailbox if it has "funny" characters in it,
and perhaps always quote it. Either way will require changing the
indexing code, since the structure is lost when writing the headers to
the database.
[1]: https://bugs.debian.org/cgi-bin/bugreport.cgi?bug=1021614
In certain scenarios involving symlinks and setting
find-file-visit-truename, text/calendar parts were not displayed
properly.
Following a suggestion of Al Haji-Ali [1], replace the use of
get-file-buffer with find-buffer-visiting.
[1]: id:m2wmneguh8.fsf@gmail.com
The first test is just a general regression test, while the second
duplicates the problem discussed in the thread starting at [1].
[1]: id:m2leo2u0uo.fsf@gmail.com
The new notmuch-search-hide-excluded option allows users to configure whether
to show or hide excluded messages (as determined by search.exclude_tags
in the local notmuch config file). It defaults to true for now to maintain
backwards-compatibility with how notmuch-{search,tree} already worked.
New commands notmuch-search-toggle-hide-excluded and
notmuch-tree-toggle-exclude have also been added. They toggle the value
of notmuch-search-hide-excluded for the search in the current search or
tree buffer. It's bound to "i" in the respective keymaps for these
modes.
Lastly I've amended some calls to notmuch-tree and notmuch-unthreaded
which didn't pass through the buffer local value of
notmuch-search-oldest-first (and now notmuch-search-exclude).
Examples of where I've done this include:
+ notmuch-jump-search
+ notmuch-tree-from-search-current-query
+ notmuch-unthreaded-from-search-current-query
+ notmuch-tree-from-search-thread
A new test file for Emacs has been added which covers the usage of the
new `notmuch-search-hide-excluded' option and interactively hiding or
showing mail with excluded tags. These test cover the basic usage of
the `notmuch-search-toggle-hide-excluded' command in notmuch-search,
notmuch-tree and notmuch-unthreaded searches. These tests also cover
the persistence of the current value of the hide-excluded mail option
as a user switches from between these different search commands.
[1]: id:87ilxlxsng.fsf@kisara.moe
Amended-by: db, fix indentation in T461-emacs-search-exclude.sh
Today someone asked me the (reasonable) question of how much
performance impact there is from synching tags to maildir flags. It
turns out it is noticeable, about a 50% overhead compared to
non-synched tags (according to these tests). In practice I don't know
if it's a big problem for users, since I don't know what fraction of
tagging operations involve "special" tags.
Some people (e.g: me) prefer to read their email with threaded
representation by default.
Without this commit, I have to run "notmuch-search" (autoloaded)
first to get "notmuch-tree" loaded.
Previously we both found a database and returned a message saying that
the database could not be found (along with a success code). This
change should prevent spurious error output.
Duplicate bug reported in id:87wmtvcor5.fsf@alyssa.is
The error message is nonsense, because notmuch config list actually
includes the database in those two cases.
Currently the --to/--cc/--bcc options add "user@example.com, " to the
message headers, with the the unnecessary ", " separator after the
last address, regardless of how many addresses are being added.
This used to be fine, but with recent emacs mm, trying to send the
email with the trailing commas leads to prompt:
Email address looks invalid; send anyway? (y or n)
Fix this by only adding the commas between addresses, avoiding the
trailing commas.
qsort(3) does not promise stability, and recent versions of glibc have
been showing more unstable behaviour [2]. Michael Gruber observed [1] test
breakage due to changing output order for message properties.
We provide a sorting order of (key,value) pairs that _looks_ stable by
breaking ties based on value if keys are equal. Internally there may
be some instability in the case of duplicate (key,value) pairs, but it
should not be observable via the iterator API.
[1]: id:CAA19uiSHjVFmwH0pMC7WwDYCOSzu3yqNbuYhu3ZMeNNRh313eA@mail.gmail.com
[2]: id:87msv3i44u.fsf@oldenburg.str.redhat.com
