David Bremner [Tue, 24 Aug 2021 15:17:22 +0000 (08:17 -0700)]
lib/parse-sexp: support phrase queries.
Anything that is quoted or not purely word characters is considered a
phrase. Phrases are not stemmed, because the stems do not have
positional information in the database. It is less efficient to scan
the term twice, but it avoids a second pass to add prefixes, so maybe
it balances out. In any case, it seems unlikely query parsing is very
often a bottleneck.
David Bremner [Tue, 24 Aug 2021 15:17:19 +0000 (08:17 -0700)]
lib/parse-sexp: support and, not, and or.
All operations and (Xapian) fields will eventually have an entry in
the prefixes table. The flags field is just a placeholder for now, but
will eventually distinguish between various kinds of prefixes.
David Bremner [Tue, 24 Aug 2021 15:17:16 +0000 (08:17 -0700)]
lib/parse-sexp: parse single terms and the empty list.
There is not much of a parser here yet, but it already does some
useful error reporting. Most functionality sketched in the
documentation is not implemented yet; detailed documentation will
follow with the implementation.
David Bremner [Tue, 24 Aug 2021 15:17:13 +0000 (08:17 -0700)]
lib: define notmuch_query_create_with_syntax
Set the parsing syntax when the (notmuch) query object is
created. Initially the library always returns a trivial query that
matches all messages when using s-expression syntax.
It seems better to select the syntax at query creation time because
the lazy parsing is an implementation detail.
David Bremner [Tue, 24 Aug 2021 15:17:11 +0000 (08:17 -0700)]
configure: optional library sfsexp
The configure part is essentially the same as the other checks using
pkg-config. Since the optional inclusion of this feature changes what
options are available to the user, include it in the "built_with"
pseudo-configuration keys.
David Bremner [Tue, 24 Aug 2021 15:17:10 +0000 (08:17 -0700)]
CLI: make variable n_requested_db_uuid file scope.
It turns out that now that we pass an open database into the
subcommands, it is easy to check any requested uuid against the
database at the same time as we process the other shared
arguments. This results in overall less boilerplate code, as well as
making a CLI scope function and variable file scope in notmuch.c.
David Bremner [Sat, 4 Sep 2021 18:26:48 +0000 (11:26 -0700)]
debian: temporarily disable two subtests
These are failing on (surprisingly) the Debian amd64
autobuilder. There were also previous reports of failures on Ubuntu
s390x. Fixing this may require changing the way the default is
calculated.
David Bremner [Sun, 4 Jul 2021 13:16:38 +0000 (10:16 -0300)]
emacs: don't inline message/rfc822 parts without content
This avoids some ugly error messages and exceptions, and hopes that
some gnus method will display message/rfc822 parts that have only a
file, no :content part.
David Bremner [Sat, 3 Jul 2021 20:11:31 +0000 (17:11 -0300)]
test: known broken test for emacs display of message/rfc822 parts
There are at least 3 bugs present.
1) notmuch-show-insert-part-message/rfc822 assumes that message/rfc822
parts will have a ":content" property, but that turns out not to be
the case.
2) something deep in gnus wants gnus-newsgroup-charset, but that is
defined in gnus-sum, which is not loaded by default.
3) If gnus-sum is loaded, then the display of the message/rfc822 part
succeeds, but the buffer gets put into gnus-article-mode, which means
that, inter alia, notmuch text properties and keybindings get wiped.
lib: consider all instances of Delivered-To header
When using notmuch-reply and guessing the From: address from
Delivered-To headers, I had the wrong address chosen today. This was
because the messages from the notmuch list contain these headers in this
order:
Delivered-To: hannu.hartikainen@gmail.com
... Delivered-To: hannu@hrtk.in
In my .notmuch-config I have the following configuration:
Before this change, notmuch-reply would guess From: @gmail.com because
that is the first Delivered-To header present. After the change, the
primary address is chosen as I would expect.
Add a known broken subtest for guessing From: correctly when there are
multiple Delivered-To: headers. The address configured as primary_email
should get picked.
David Bremner [Sun, 22 Aug 2021 03:51:07 +0000 (20:51 -0700)]
emacs/rstdoc: escape '*'
This is just a regular character in docstrings (as it is fairly often
used in lisp identifiers and buffer names) but is the start of
emphasis in rst. This change is needed to quell a noisy warning when
including notmuch-tree.rsti
David Bremner [Sun, 22 Aug 2021 00:00:11 +0000 (17:00 -0700)]
CLI: define and use format version 5
This is a bit of a cheat, since the format does not actually
change. On the other hand it is fairly common to do something like
this to shared libary SONAMEs when the ABI changes in some subtle way.
It does rely on the format-version argument being early enough on the
command line to generate a sensible error message.
jao [Sun, 22 Aug 2021 00:50:49 +0000 (01:50 +0100)]
emacs: new command notmuch-tree-filter-by-tag
This new command for notmuch-tree-mode is analogous to
notmuch-search-filter-by-tag, bound to "t" in notmuch-search-mode; it
gets therefore the same "t" keybinding in notmuch-tree-mode (replacing
the current assignment to notmuch-search-by-tag).
David Bremner [Wed, 4 Aug 2021 10:42:33 +0000 (07:42 -0300)]
test/path-config: use test_expect_equal_file_nonempty
This is more robust against crashes when the expected output is also
generated by notmuch. In the case where the expected output is
explicit, it seems like overkill.
David Bremner [Wed, 4 Aug 2021 10:42:32 +0000 (07:42 -0300)]
test: add test_expect_equal_file_nonempty
A common bug in tests is that the code used to generate the EXPECTED
file fails, generating no output. When the code generating the OUTPUT
file fails in the same way, the test passes, even though there is a
failure being hidden. Add a new test function that guards against
this.
David Bremner [Sun, 18 Jul 2021 17:58:22 +0000 (14:58 -0300)]
CLI/config: restore "notmuch config get built_with.*"
We need to special case the config section "built_with" because it is
not (currently) handled by the library. This seems consist with the
other sub-sub-commands 'list' and 'set'.
David Bremner [Wed, 4 Aug 2021 10:42:34 +0000 (07:42 -0300)]
test: add known broken tests for finding database via MAILDIR
This highlights a bug reported by several users, including
Mohsin Kaleem [1].
The inconsistent use of test_begin_subtest_known_broken is because
some of these tests pass even though the database cannot be
located. This problem is left for a future commit.
Austin Ray [Fri, 13 Aug 2021 16:50:28 +0000 (12:50 -0400)]
lib: bump libnotmuch minor version
Notmuch 0.32 corresponds to libnotmuch 5.4 as indicated by docstrings;
however, the minor number wasn't bumped. Any libnotmuch downstream
consumer using the LIBNOTMUCH_CHECK_VERSION macro to support multiple
versions won't be able to access the new 5.4 functions.
Austin Ray [Sat, 14 Aug 2021 22:39:17 +0000 (18:39 -0400)]
lib: correct deprecated db open functions' docs
Both notmuch_database_open() and notmuch_database_open_verbose()'s
documentation state they call notmuch_database_open_with_config() with
config_path=NULL; however, their implementations pass an empty string.
The empty string is the correct value to maintain their original
behavior of not loading the user's configuration so their documentation
is incorrect.
The intent of the 'notmuch-jump-key' face is to allow users/themes to
differentiate the text of the minibuffer prompt from the keys that are
associated with jump actions. Commit 5cc106b0 correctly introduced the
'notmuch-jump-key' face for keys, but mistakenly applied it to the
prompt as well.
jao [Wed, 4 Aug 2021 00:02:41 +0000 (01:02 +0100)]
emacs: honour notmuch-show-text/html-blocked-images in w3m messages
When mm-text-html-renderer is set to 'w3m, the variable playing the
role of a regular expression for blocked images is
w3m-ignored-image-url-regexp. We bind it when the renderer is not
'shr.
It was assumed the destructor of notmuch_rb_database_type did return a
notmuch_status_t because that's what notmuch_database_close returns, and
that value was checked by notmuch_rb_database_close in order to decide
if to raise an exception.
It turns out notmuch_database_destroy was called instead, so nothing was
returned (void).
All the destroy functions are void, and that's what we want.
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Jonas Bernoulli [Tue, 16 Feb 2021 00:01:38 +0000 (01:01 +0100)]
emacs: notmuch-tree: mark the initial message at point as read
When moving between message in a tree or show buffer, the message at
point is marked as read. Likewise when creating such a buffer, then
the message that is initially at point is supposed to be marked as
read as well.
The latter worked for `notmuch-show' but not for `notmuch-tree'.
Press "RET" or "M-RET" in a search buffer to observe these behaviors.
In both cases the marking is supposed to be done by the function
`notmuch-show-command-hook'. In the case of `notmuch-show' that
function is added directly to `post-command-hook'.
`notmuch-tree' instead adds the function `notmuch-tree-command-hook'
to `post-command-hook' and that calls `notmuch-show-command-hook',
in the respective show buffer, but of course only if that exists.
Because the tree buffer is created asynchronously, the show buffer
doesn't exist yet by the time the `post-command-hook' is run, so
we have to explicitly run `notmuch-tree-command-hook' once the
show buffer exists.
The show buffer is created when `notmuch-tree-goto-and-insert-msg'
calls `notmuch-tree-show-message-in'. `notmuch-tree-process-filter'
is what finally brings us here.
Felipe Contreras [Mon, 17 May 2021 19:39:15 +0000 (14:39 -0500)]
ruby: enable garbage collection using talloc
We basically steal all the objects from their notmuch parents, therefore
they are completely under Ruby's gc control.
The order at which these objects are freed does not matter any more,
because destroying the database does not destroy all the children
objects, since they belong to Ruby now.
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
Felipe Contreras [Mon, 17 May 2021 19:39:14 +0000 (14:39 -0500)]
ruby: create an actual wrapper struct
Currently Ruby data points directly to a notmuch object (e.g.
notmuch_database_t), since we don't need any extra data that is fine.
However, in the next commit we will need extra data, therefore we create
a new struct notmuch_rb_object_t wrapper which contains nothing but a
pointer to the current pointer (e.g. notmuch_database_t).
This struct is tied to the Ruby object, and is freed when the Ruby
object is freed by the garbage collector.
We do nothing with this wrapper, so no functionality should be changed.
Signed-off-by: Felipe Contreras <felipe.contreras@gmail.com>
David Bremner [Wed, 7 Jul 2021 01:35:30 +0000 (22:35 -0300)]
emacs: update declaration of notmuch-tree
This fixes an a warning from the byte-compiler. The commit 74ab62a34018b38a40db4d363fff3faae964b47a changed the function
signature but did not update the declaration.
doc: new notmuch show --sort and related emacs commands
New --sort CLI option documented in notmuch-show's man page, and
notmuch-search-toggle-order mentioned in doc/notmuch-emacs.rst and
devel/emacs-keybindings.org (in the latter, there's also some
whitespace changes in a table introduced by org-mode).
Add the command-line option --sort to the show command of the CLI
notmuch interface, with the same possible values as the same option in
notmuch search.
David Bremner [Sat, 26 Jun 2021 12:04:28 +0000 (09:04 -0300)]
emacs/tree use notmuch-show-single-message
This is more efficient than notmuch-show-only-matching-messages, since
we do not parse the potentially large thread structure to find a
single message.
This is only a partial fix for notmuch-tree view, because displaying
the thread structure in the tree-mode window still crashes on long
threads. It is however enough to make unthreaded view handle long
threads.
This dynamically bound variable can be set when the caller of
notmuch-show guarantees that exactly one message will match the
query. It avoids transporting and parsing the complete thread
structure.
edef [Sun, 13 Jun 2021 08:23:58 +0000 (08:23 +0000)]
emacs: remap send-message and send-message-and-exit
All three of C-c C-c, <menu-bar> <Message> <Send Message>,
and <tool-bar> <Send Message> are bound to message-send-and-exit by
message.el, but notmuch-mua.el only had an explicit override for the
keyboard binding. This mostly manifests as confusing Fcc behaviour for
GUI users.
Patching the bindings for specific keys is rather brittle, since it has
to be aware of every relevant binding. This change switches to instead
using a remap binding, which turns any binding for message-send or
message-send-and-exit into a binding for the corresponding notmuch-mua
command.
David Bremner [Tue, 18 May 2021 02:06:01 +0000 (23:06 -0300)]
lib: autocommit after some number of completed transactions
This change addresses two known issues with large sets of changes to
the database. The first is that as reported by Steven Allen [1],
notmuch commits are not "flushed" when they complete, which means that
if there is an open transaction when the database closes (or e.g. the
program crashes) then all changes since the last commit will be
discarded (nothing is irrecoverably lost for "notmuch new", as the
indexing process just restarts next time it is run). This does not
really "fix" the issue reported in [1]; that seems rather difficult
given how transactions work in Xapian. On the other hand, with the
default settings, this should mean one only loses less than a minutes
worth of work. The second issue is the occasionally reported "storm"
of disk writes when notmuch finishes. I don't yet have a test for
this, but I think committing as we go should reduce the amount of work
when finalizing the database.
David Bremner [Tue, 18 May 2021 11:03:22 +0000 (08:03 -0300)]
database/close: remove misleading code / comment
Unfortunately, it doesn't make a difference if we call
cancel_transaction or not, all uncommited changes are discarded if
there is an open (unflushed) transaction.
The minibuffer-prompt face that was used before made it impossible to
differentiate between two distinct UI elements: (i) the prompt's text
which itself cannot be acted upon, (ii) the actionable keys used to
jump to searches/tags.
The use of a named face, notmuch-jump-key, makes it possible for users
or theme developers to apply properties that are specific to each of
those two cases.
In the interest of backward compatibility, the new face inherits from
minibuffer-prompt.