David Bremner [Mon, 26 Oct 2015 23:22:47 +0000 (00:22 +0100)]
emacs: replace use of notmuch-address-message-insinuate
This allows e.g. Gnus users to load this file without changing
message-mode behaviour.
This will disable completion for those that did not customize the
variable but relied on the existence of a file named "notmuch-addresses"
in their path. In the next commit the default behaviour will change to
use a "workalike" internal completion mechanism.
David Bremner [Mon, 26 Oct 2015 01:09:56 +0000 (22:09 -0300)]
debian: release 0.21~rc3-3
Build-conflict with gdb-minimal, since there is no gdb-python to depend
on
David Bremner [Sun, 25 Oct 2015 16:44:06 +0000 (13:44 -0300)]
debian: bug fix upload for #802952
David Bremner [Fri, 23 Oct 2015 10:54:29 +0000 (07:54 -0300)]
debian: start changelog entry for 0.21
David Bremner [Thu, 22 Oct 2015 11:01:19 +0000 (08:01 -0300)]
NEWS: trim discussion of keybindings
Try to reduce the total size of NEWS
David Bremner [Thu, 22 Oct 2015 10:47:43 +0000 (07:47 -0300)]
version: bump to 0.21~rc3
Mark Walters [Wed, 21 Oct 2015 18:40:41 +0000 (19:40 +0100)]
NEWS for minor emacs updates
This adds NEWS items for
the new limit/filter function in notmuch-show,
the saved-search option for tree view,
the binding S to run the current tree-view search in notmuch-search,
the increase in max text part size
the bugfix for replying to encrypted messages from tree view
Mark Walters [Wed, 21 Oct 2015 18:51:37 +0000 (19:51 +0100)]
emacs: hello: bugfix: update docs for saved-searches
Update the documentation string for notmuch-saved-searches to include
the new :search-type option.
Jani Nikula [Sat, 17 Oct 2015 21:03:35 +0000 (00:03 +0300)]
completion: complete lastmod: search prefix
Jani Nikula [Sat, 17 Oct 2015 16:30:03 +0000 (19:30 +0300)]
completion: clean up top level completion
Add note about missing features.
Jani Nikula [Sat, 17 Oct 2015 16:30:02 +0000 (19:30 +0300)]
completion: complete shared options for each command
Add support for completing --help, --uuid=, and --version after the
subcommand. Do not support shared options at the top level yet due to
difficulties in handling options with arguments.
--uuid necessitates adding 'compopt -o nospace' also to notmuch new
completion, but that just brings it in line with the rest.
Jani Nikula [Sat, 17 Oct 2015 16:30:01 +0000 (19:30 +0300)]
completion: complete notmuch count --lastmod
Jani Nikula [Sat, 26 Sep 2015 11:12:50 +0000 (14:12 +0300)]
completion: handle notmuch address --deduplicate= option
Complete notmuch address --deduplicate=(no|mailbox|address).
Mark Walters [Wed, 14 Oct 2015 08:38:23 +0000 (09:38 +0100)]
emacs: tree: bind S to run current query in search mode
Mark Walters [Wed, 14 Oct 2015 08:38:22 +0000 (09:38 +0100)]
emacs: allow saved searches to select tree-view
This patch allows the user to customize a saved search to choose tree
view rather than the default search view. It also updates notmuch-jump
so that it respects this choice.
Jani Nikula [Sat, 17 Oct 2015 20:58:48 +0000 (23:58 +0300)]
man: clarify the parameters for lastmod: range query
<since> and <until> for the lastmod: prefix right below the date:
prefix description give the impression one could use last modified
dates to lastmod: which is not at all the case. Use
<initial-revision>..<final-revision> instead.
Tomi Ollila [Mon, 19 Oct 2015 18:09:21 +0000 (21:09 +0300)]
NEWS: a few formatting updates for 0.21 news items (wiki compatibility)
Trailing dots were removed from 3 NEWS items so that those appear in
same level as surrounding "heading" lines in generated wiki page.
One trailing dot was added to nmbug-status item so it appears as normal
text in generated wiki page. `nmbug-status` was put in backticks so it
looks the same as in older nmbug-status news text.
David Bremner [Mon, 19 Oct 2015 10:26:04 +0000 (07:26 -0300)]
debian: changelog stanza for 0.21~rc2-1
David Bremner [Mon, 19 Oct 2015 10:12:18 +0000 (07:12 -0300)]
version: bump to 0.21~rc2
Jani Nikula [Sat, 17 Oct 2015 13:06:53 +0000 (16:06 +0300)]
NEWS: news for notmuch_directory_delete
Jani Nikula [Sat, 17 Oct 2015 13:06:52 +0000 (16:06 +0300)]
NEWS: news for notmuch address --deduplicate option
Jani Nikula [Sat, 17 Oct 2015 13:06:51 +0000 (16:06 +0300)]
NEWS: news for date:<expr>..!
Jani Nikula [Sat, 17 Oct 2015 13:06:50 +0000 (16:06 +0300)]
NEWS: nmbug-status supports sort order
David Bremner [Sat, 17 Oct 2015 13:06:01 +0000 (10:06 -0300)]
NEWS: note some Emacs UI changes for 0.21
David Bremner [Sat, 17 Oct 2015 13:01:01 +0000 (10:01 -0300)]
NEWS: new count api / deprecated count/search functions for 0.21
David Bremner [Sat, 17 Oct 2015 12:57:40 +0000 (09:57 -0300)]
NEWS: note some build system changes for 0.21
The ruby bindings changes in particular will be user visible. The
skipped tests status may also be interesting for packagers
David Bremner [Sat, 17 Oct 2015 12:39:19 +0000 (09:39 -0300)]
NEWS: news for initial revision tracking support (0.21)
Mark Walters [Thu, 15 Oct 2015 18:53:09 +0000 (19:53 +0100)]
emacs: show: increase default max-text-part-size
Currently notmuch-show-max-text-part-size is 10000 which means some
relatively normal messages have all parts hidden by default. Increase
this to 100000 by default.
The setting was introduced to alleviate problems with notmuch being
very slow on large threads. Users hitting these problems may wish to
customize this variable to something smaller (like 10000).
Mark Walters [Thu, 15 Oct 2015 18:46:34 +0000 (19:46 +0100)]
emacs: tree bugfix
Formerly replying to an encrypted message in tree-view did not work:
the message was not decrypted. This commit makes notmuch-tree respect
the setting of notmuch-crypto-process-mime. In particular, if
notmuch-crypto-process-mime is set to t, then replying to encrypted
messages in tree mode will now decrypt the reply (as it already did in
show mode).
Tomi Ollila [Thu, 15 Oct 2015 17:57:43 +0000 (20:57 +0300)]
doc: update copyright year to contain range 2009-2015
Now it is accurate and provides more information to the
reader of the documents.
David Bremner [Thu, 15 Oct 2015 13:09:24 +0000 (10:09 -0300)]
debian: update symbols file
we are only adding symbols, which is exactly the interesting case for a
symbols file.
David Bremner [Thu, 15 Oct 2015 11:16:26 +0000 (08:16 -0300)]
version: bump to 0.21~rc1
David Bremner [Thu, 15 Oct 2015 11:14:19 +0000 (08:14 -0300)]
debian: update changelogs for pre-release
a sensible changelog can wait for upstream NEWS to be updated
Jani Nikula [Fri, 25 Sep 2015 20:48:46 +0000 (23:48 +0300)]
cli: delete directory documents on directory removal
There was a problem with the directory documents being left behind when
the filesystem directory was removed. This was worked around in [1].
However, that ignored the fact that the directory documents are also
still listed by notmuch_directory_get_child_directories() leading to
confusing results when running notmuch new. The directory documents are
found and queued for removal over and over again.
Fix the problem for real by removing the directory documents. This fixes
the tests flagged as broken in [2].
The (non-deterministic) hack test from [3] also still passes with this
change.
[1] commit
acd66cdec075312944e527febd46382e54d99367
[2] commit
ed9ceda623d3f22fb66365b33db63c5c982067d3
[3] id:
1441445731-4362-1-git-send-email-jani@nikula.org
Jani Nikula [Fri, 25 Sep 2015 20:48:45 +0000 (23:48 +0300)]
lib: add interface to delete directory documents
As mentioned in
acd66cdec075312944e527febd46382e54d99367 we don't have
an interface to delete directory documents, and they're left behind. Add
the interface.
Jani Nikula [Fri, 25 Sep 2015 20:48:44 +0000 (23:48 +0300)]
test: flag one more notmuch new test as broken
Drop the test update added in [1] and mark the test as broken, like the
tests flagged as broken in [2]. These all reflect the same underlying
breakage with (lack of) directory deletion.
[1] commit
e4e04bbc328f990e36b77f508aef904d156029b1
[2] commit
ed9ceda623d3f22fb66365b33db63c5c982067d3
David Bremner [Mon, 5 Oct 2015 23:16:59 +0000 (20:16 -0300)]
lib: update doxygen comments to add @since for the new _st API
We should probably to this for all new functions introduced from now on.
David Bremner [Sun, 27 Sep 2015 15:32:03 +0000 (12:32 -0300)]
ruby: use new query_search API
These changes should not be too surprising for users because the
routines were already potentially throwing exceptions.
David Bremner [Sun, 27 Sep 2015 15:32:02 +0000 (12:32 -0300)]
lib: migrate thread.cc to new query_search API
here we rely on thread_id_query being attached to the local talloc
context, so no new cleanup code is needed.
David Bremner [Sun, 27 Sep 2015 15:32:01 +0000 (12:32 -0300)]
lib: migrate notmuch_database_upgrade to new query_search API
Here we depend on the error path cleaning up query
David Bremner [Sun, 27 Sep 2015 15:32:00 +0000 (12:32 -0300)]
python: update bindings for new count API
Note that any mismatches are not detected until runtime (if at all)
with the python bindings, so tests are crucial
David Bremner [Sun, 27 Sep 2015 15:31:59 +0000 (12:31 -0300)]
ruby: use new count API
This change of replacing ignoring errors with exceptions is intended,
and indeed one of the main motivations for the libnotmuch API changes.
David Bremner [Sun, 27 Sep 2015 15:31:58 +0000 (12:31 -0300)]
cli: update to use new count API
Essentially replace each call to notmuch_count_* with the corresponding
_st call, followed by print_status_query.
David Bremner [Sun, 27 Sep 2015 15:31:57 +0000 (12:31 -0300)]
lib: add versions of n_q_count_{message,threads} with status return
Although I think it's a pretty bad idea to continue using the old API,
this allows both a more gentle transition for clients of the library,
and allows us to break one monolithic change into a series
David Bremner [Sun, 27 Sep 2015 15:31:56 +0000 (12:31 -0300)]
cli/count: simplify and document return value of print_count
Essentially a cosmetic change.
David Bremner [Sun, 27 Sep 2015 15:31:55 +0000 (12:31 -0300)]
lib: move query variable to function scope
This is a prelude to deallocating it (if necessary) on the error path.
Jani Nikula [Sat, 3 Oct 2015 07:57:56 +0000 (10:57 +0300)]
nmbug-status: print config errors to stderr
Particularly scripted usage with stdout redirection can be confusing
if errors are printed to stdout instead of stderr.
Tomi Ollila [Wed, 30 Sep 2015 06:36:52 +0000 (09:36 +0300)]
notmuch-emacs-mua: make --auto-daemon imply --create-frame
Before this change with --auto-daemon but without --create-frame
emacs server was started but no clients stay connected to it
(in both graphical and terminal displays).
Note that this changes how --client --auto-daemon works on
graphical display; New emacs frame is now created for the
message (and message-exit-actions hook appended).
Jani Nikula [Sat, 15 Aug 2015 09:05:15 +0000 (12:05 +0300)]
notmuch-emacs-mua: do not create a frame by default with --client
Make the default behaviour for --client the same as emacsclient
default: do not create a new frame. Add a new option --create-frame,
passing the same option to emacsclient to create a frame.
Jani Nikula [Sat, 15 Aug 2015 09:05:14 +0000 (12:05 +0300)]
notmuch-emacs-mua: add --auto-daemon option to start Emacs daemon as needed
Automatically starting Emacs in daemon mode if the Emacs server is not
running is a matter of preference better not hard coded in
notmuch-emacs-mua. Add an option to control the behaviour.
Jani Nikula [Sat, 15 Aug 2015 09:05:13 +0000 (12:05 +0300)]
notmuch-emacs-mua: move --client option handling around
Move --client handling to a more suitable location. No functional
changes.
Jani Nikula [Sat, 15 Aug 2015 09:05:12 +0000 (12:05 +0300)]
notmuch-emacs-mua: support --no-window-system also for non-client
Clearly --no-window-system should be applicable without --client as
well. Prepare for future changes with the naming of the variables.
Jani Nikula [Sat, 15 Aug 2015 09:05:11 +0000 (12:05 +0300)]
man: clean up notmuch-emacs-mua man page
Move --client description to a more suitable location, stylistic
changes.
Jani Nikula [Fri, 25 Sep 2015 18:47:35 +0000 (21:47 +0300)]
cli: use designated initializer to initialize add_files_state
The side effect is that all of add_files_state will be initialized to
zero, removing any lingering doubt that some of it might not be
initialized. It's not a small struct, and the initialization is
scattered around a bit, so this makes the code more readable.
Jani Nikula [Sat, 26 Sep 2015 11:37:51 +0000 (14:37 +0300)]
nmbug-status: add support for specifying sort order for each view
Let each view have a "sort" key, typically used with values
"oldest-first" or "newest-first" (although all values in Query.SORT
are accepted), and sort the results accordingly. Oldest first remains
the default.
The dynamic approach of mapping sort values is as suggested by
W. Trevor King <wking@tremily.us>.
David Bremner [Sat, 26 Sep 2015 13:31:50 +0000 (10:31 -0300)]
Merge branch 'release'
bring debian specific changes into master
David Bremner [Sat, 26 Sep 2015 12:27:34 +0000 (09:27 -0300)]
debian: changelog stanza for 0.20.2-2
David Bremner [Sat, 26 Sep 2015 12:22:11 +0000 (09:22 -0300)]
debian: fix linking in emacsen-install script
When upgrading emacs, the install file can (apparently) be called with
the lisp files already linked into the flavour specific
directories. In this case we should not fail.
Jani Nikula [Thu, 3 Sep 2015 19:40:05 +0000 (22:40 +0300)]
cli: do not sort addresses on --output=count or --deduplicate=address
The order of the results with --output=count and --deduplicate=address
are unspecified as they're based on a hash table traversal. This being
the case, optimize the query by explicitly requesting unsorted
results. Clarify the documentation accordingly.
Jani Nikula [Thu, 3 Sep 2015 19:40:04 +0000 (22:40 +0300)]
man: document notmuch address --deduplicate=address option
Document the deduplication based on case insensitive address.
Jani Nikula [Fri, 25 Sep 2015 17:08:09 +0000 (20:08 +0300)]
test: add notmuch address --deduplicate=(no|mailbox|address) tests
First a simple smoke test first, next generate messages with multiple
email address variants and check the behaviour of deduplication
schemes with these.
Jani Nikula [Thu, 3 Sep 2015 19:40:03 +0000 (22:40 +0300)]
cli: add support for deduplicating based on case insensitive address
Consider all variants of an email address as one, and print the most
common variant.
Jani Nikula [Fri, 25 Sep 2015 16:48:20 +0000 (19:48 +0300)]
cli: change the data structure for notmuch address deduplication
Currently we key the address hash table with the case sensitive "name
<address>". Switch to case insensitive keying with just address, and
store the case sensitive name and address in linked lists. This will
be helpful in adding support for different deduplication schemes in
the future.
There will be a slight performance penalty for the current full case
sensitive name + address deduplication, but this is simpler as a whole
when other deduplication schemes are added, and I expect the schemes
to be added to become more popular than the current default.
Aparet from the possible performance penalty, the only user visible
change should be the change in the output ordering for
--output=count. The order is not guaranteed (and is based on hash
table traversal) currently anyway, so this should be of no
consequence.
Jani Nikula [Fri, 25 Sep 2015 16:48:19 +0000 (19:48 +0300)]
util: add strcmp_null, a strcmp that handles NULL parameters
Add strcmp_null, a strcmp that handles NULL strings; in strcmp terms a
NULL string is considered to be less than a non-NULL string.
Jani Nikula [Sat, 15 Aug 2015 11:25:47 +0000 (14:25 +0300)]
lib: add support for date:<expr>..! to mean date:<expr>..<expr>
It doesn't seem likely we can support simple date:<expr> expanding to
date:<expr>..<expr> any time soon. (This can be done with a future
version of Xapian, or with a custom query query parser.) In the mean
time, provide shorthand date:<expr>..! to mean the same. This is
useful, as the expansion takes place before interpetation, and we can
use, for example, date:yesterday..! to match from beginning of
yesterday to end of yesterday.
Idea from Mark Walters <markwalters1009@gmail.com>.
David Bremner [Thu, 24 Sep 2015 11:13:30 +0000 (08:13 -0300)]
test suite: don't consider skipped individual tests as failing
It isn't completely clear what we want to do here, but
1) We currently don't fail if we skip a whole test file (mainly because
we neglect to count those skipped tests properly). This change at least
makes the two kinds of skipping consistent.
2) Automated build environments may have good reasons for building with
a minimal set of prereqs, and we don't want to discourage running our
test suite by breaking builds.
Tomi Ollila [Fri, 11 Sep 2015 20:44:11 +0000 (23:44 +0300)]
make test: NOTMUCH_TEST_QUIET=1 is now the default
make test V=1 (or any other value than 0) and make test V=0
works similar way as build in general
Jani Nikula [Wed, 23 Sep 2015 19:31:13 +0000 (22:31 +0300)]
man: document notmuch address --deduplicate=(no|mailbox) option
Document the deduplication of results.
Jani Nikula [Fri, 4 Sep 2015 18:35:16 +0000 (21:35 +0300)]
test: notmuch address --deduplicate=no tests
Test not using address deduplication. Incorporate some trivial output
sorting tests here, as they seem to lack tests.
Jani Nikula [Thu, 3 Sep 2015 19:39:59 +0000 (22:39 +0300)]
cli: add support for not deduplicating notmuch address results
Make it possible to use notmuch address as part of a | sort | uniq -c
pipe instead of forcing --output=count. This is useful for combining
results from multiple notmuch address queries.
Jani Nikula [Wed, 23 Sep 2015 18:48:05 +0000 (21:48 +0300)]
test: check argc in symbol-test
Check argc mainly to fix unused parameter warning:
test/symbol-test.cc:7:14: warning: unused parameter ‘argc’ [-Wunused-parameter]
int main(int argc, char** argv) {
^
This makes more sense than telling the compiler it's unused on
purpose.
David Bremner [Wed, 23 Sep 2015 11:31:19 +0000 (08:31 -0300)]
lib: constify arguments to notmuch_query_get_*
These functions are all just accessors, and it's pretty clear they don't
modify the query struct. This also fixes one warning I created when I
introduced status.c.
David Bremner [Sun, 6 Sep 2015 13:15:47 +0000 (10:15 -0300)]
cli: convert remainder of CLI to n_q_search_{messages,threads}_st
I think it would be no real problem to cut and paste the gdb based
error message test from count to the other clients modified here, but
I'm not currently convinced it's worth the trouble since the code path
being tested is almost the the same, and the tests are relatively
heavyweight.
David Bremner [Sun, 6 Sep 2015 13:15:46 +0000 (10:15 -0300)]
cli/count: update to use notmuch_query_search_messages_st
This brings back status information that may have been hidden by the
great library logging conversion.
Note the change of the internal API / return-value for count_files. The
other count calls to the lib will also get error handling when that API
is updated in the lib.
David Bremner [Sun, 6 Sep 2015 13:15:45 +0000 (10:15 -0300)]
cli: add utility routine to print error status.
No attention to formatting here, initially just focus on getting the
relevant strings out of the library.
David Bremner [Mon, 7 Sep 2015 13:29:14 +0000 (10:29 -0300)]
cli/count: apply uncrustify changes
Leave the indentation alone because I don't know how to duplicate it
with Emacs.
David Bremner [Mon, 7 Sep 2015 13:08:25 +0000 (10:08 -0300)]
cli/count: remove unused argument handling for output=modifications
In the end this UI was rejected, but bits of it stayed in the code to
confuse future generations.
Jani Nikula [Thu, 3 Sep 2015 19:40:01 +0000 (22:40 +0300)]
util: move strcase_equal and strcase_hash to util
For future use in both cli and lib.
Jani Nikula [Thu, 3 Sep 2015 19:39:58 +0000 (22:39 +0300)]
cli: abstract new mailbox creation
We'll be needing more mailbox creation soon, so abstract it
away. While at it, check for allocation failures. No other functional
changes.
Jani Nikula [Thu, 3 Sep 2015 19:39:57 +0000 (22:39 +0300)]
cli: g_hash_table_lookup_extended is overkill
Switch to normal glib hash table lookup. The extended version is only
required if the values may contain NULL.
Jani Nikula [Fri, 4 Sep 2015 18:49:14 +0000 (21:49 +0300)]
test: fix whitespace/indentation in symbol-test
Not of much consequence, but makes it nicer to do further edits.
Jani Nikula [Sat, 5 Sep 2015 09:35:31 +0000 (12:35 +0300)]
cli: reset db directory mtime upon directory removal
The library does not have a function to remove a directory document
for a path. Usually this doesn't matter except for a slight waste of
space. However, if the same directory gets added to the filesystem
again, the old directory document is found with the old mtime. Reset
the directory mtime on removal to avoid problems.
The corner case that can hit this problem is renaming directories back
and forth. Renaming does not change the mtime of the directory in the
filesystem, and thus the old db directory document mtime may match the
fs mtime of the directory.
The long term fix might be to add a library function to remove a
directory document, however this is a much simpler and faster fix for
the time being.
Uli Scholler [Sun, 6 Sep 2015 21:29:05 +0000 (23:29 +0200)]
emacs: wrap current search in parens when filtering
When filtering the current search further with notmuch-search-filter,
wrap the current search in parens (if necessary).
This fixes unexpected behavior when the current search is
complex (like "(tag:this and date:one_week_ago..) or tag:that").
David Bremner [Sun, 7 Jun 2015 15:02:00 +0000 (17:02 +0200)]
lib: constify argument to notmuch_database_status_string
We don't modify the database struct, so no harm in committing to that.
David Bremner [Sun, 7 Jun 2015 15:01:59 +0000 (17:01 +0200)]
test: move backup_database and restore_database to library
The plan is to re-use them in testing the same errors from the CLI
David Bremner [Sun, 7 Jun 2015 15:01:58 +0000 (17:01 +0200)]
lib: note remaining uses of deprecated message search API
The two remaining cases in the lib seem to require more than a simple
replacement of the old call, with the new call plus a check of the
return value.
David Bremner [Sun, 7 Jun 2015 15:01:57 +0000 (17:01 +0200)]
lib: remove use of notmuch_query_search_messages from query.cc
There is not too much point in worrying about the bad error reporting
here, because the count api is due for the same deprecation.
David Bremner [Sun, 16 Aug 2015 15:33:21 +0000 (17:33 +0200)]
cli/lib: remove support for GMime 2.4
It's becoming a maintenance burden to do anything things with the
crypto glue code twice, once for 2.4 and once for 2.6. I don't have
any 2.4 version available to test on my development machine anymore,
so the 2.4 specific code paths are likely not very well tested.
Austin Clements [Mon, 13 Oct 2014 06:20:03 +0000 (02:20 -0400)]
lib: Add "lastmod:" queries for filtering by last modification
The implementation is essentially the same as the date range search
prior to Jani's fancy date parser.
David Bremner [Sun, 5 Apr 2015 22:39:55 +0000 (07:39 +0900)]
cli: add global option "--uuid"
The function notmuch_exit_if_unmatched_db_uuid is split from
notmuch_process_shared_options because it needs an open notmuch
database.
There are two exceptional cases in uuid handling.
1) notmuch config and notmuch setup don't currently open the database,
so it doesn't make sense to check the UUID.
2) notmuch compact opens the database inside the library, so we either
need to open the database just to check uuid, or change the API.
David Bremner [Fri, 5 Jun 2015 06:16:17 +0000 (08:16 +0200)]
cli/count: add --lastmod
In the short term we need a way to get lastmod information e.g. for
the test suite. In the long term we probably want to add lastmod
information to at least the structured output for several other
clients (e.g. show, search).
Austin Clements [Mon, 13 Oct 2014 06:20:02 +0000 (02:20 -0400)]
lib: API to retrieve database revision and UUID
This exposes the committed database revision to library users along
with a UUID that can be used to detect when revision numbers are no
longer comparable (e.g., because the database has been replaced).
Austin Clements [Mon, 13 Oct 2014 06:20:01 +0000 (02:20 -0400)]
lib: Add per-message last modification tracking
This adds a new document value that stores the revision of the last
modification to message metadata, where the revision number increases
monotonically with each database commit.
An alternative would be to store the wall-clock time of the last
modification of each message. In principle this is simpler and has
the advantage that any process can determine the current timestamp
without support from libnotmuch. However, even assuming a computer's
clock never goes backward and ignoring clock skew in networked
environments, this has a fatal flaw. Xapian uses (optimistic)
snapshot isolation, which means reads can be concurrent with writes.
Given this, consider the following time line with a write and two read
transactions:
write |-X-A--------------|
read 1 |---B---|
read 2 |---|
The write transaction modifies message X and records the wall-clock
time of the modification at A. The writer hangs around for a while
and later commits its change. Read 1 is concurrent with the write, so
it doesn't see the change to X. It does some query and records the
wall-clock time of its results at B. Transaction read 2 later starts
after the write commits and queries for changes since wall-clock time
B (say the reads are performing an incremental backup). Even though
read 1 could not see the change to X, read 2 is told (correctly) that
X has not changed since B, the time of the last read. In fact, X
changed before wall-clock time A, but the change was not visible until
*after* wall-clock time B, so read 2 misses the change to X.
This is tricky to solve in full-blown snapshot isolation, but because
Xapian serializes writes, we can use a simple, monotonically
increasing database revision number. Furthermore, maintaining this
revision number requires no more IO than a wall-clock time solution
because Xapian already maintains statistics on the upper (and lower)
bound of each value stream.
Jani Nikula [Mon, 10 Aug 2015 12:39:16 +0000 (15:39 +0300)]
configure: clean up messages around reading libnotmuch version
Clean up punctuation for consistency, and add newline for clean
configure output.
David Bremner [Sun, 9 Aug 2015 14:25:41 +0000 (16:25 +0200)]
build: extract library versions from notmuch.h
- Make lib/notmuch.h the canonical location for the library versioning
information.
- Since the release-check should never fail now, remove it to reduce
complexity.
- Make the version numbers in notmuch.h consistent with the (now
deleted) ones in lib/Makefile.local
Tomi Ollila [Thu, 6 Aug 2015 09:13:36 +0000 (12:13 +0300)]
test: make script exit (1) if it "fails" to source (.) a file
The files (test) scripts source (with builtin command `.`) provides
information which the scripts depend, and without the `source` to
succeed allowing script to continue may lead to dangerous situations
(e.g. rm -rf "${undefined_variable}"/*).
At the end of all source (.) lines construct ' || exit 1' was added;
In our case the script script will exit if it cannot find (or read) the
file to be sourced. Additionally script would also exits if the last
command of the sourced file exited nonzero.
Tomi Ollila [Wed, 5 Aug 2015 20:05:02 +0000 (23:05 +0300)]
NEWS: emacs: notmuch-{cli,emacs}-version usage
News for notmuch-emacs-version usage in User-Agent header and
rename of function notmuch-version to notmuch-cli-version.
While at it, made some minor (consistency) punctuation changes
to make news to wiki conversion work as expected.
David Bremner [Fri, 7 Aug 2015 19:33:57 +0000 (21:33 +0200)]
emacs/notmuch-mua.el: whitespace cleanup
M-x whitespace-cleanup in Emacs.
David Bremner [Sat, 1 Aug 2015 09:44:56 +0000 (11:44 +0200)]
emacs: make modifications to message Fcc vars buffer-local
Previously we globally modified these variables, which tended to cause
problems for people using message-mode, but not notmuch-mua-mail, to
send mail.
User visible changes:
- Calling notmuch-fcc-header-setup is no longer optional. OTOH, it
seems to do the right thing if notmuch-fcc-dirs is set to nil.
- The Fcc header is visible during message composition
- The name in the mode line is changed, and no longer matches exactly
the menu label.
- Previously notmuch-mua-send-and-exit was never called. Either we
misunderstood define-mail-user-agent, or it had a bug. So there was
no difference if the user called message-send-and-exit directly. Now
there will be.
- User bindings to C-c C-c and C-c C-s in message-mode-map are
overridden. The user can override them in notmuch-message-mode-map,
but then they're on their own for Fcc handling.
David Bremner [Sat, 1 Aug 2015 09:44:55 +0000 (11:44 +0200)]
emacs: define a notmuch-compose-mode based on message mode.
This is to provide a clean way of overriding e.g. keybindings when
sending mail from notmuch.
This is needed in particular to allow somewhere to dynamically bind
certain message-mode variables which are not respected when buffer-local. See e.g.
http://debbugs.gnu.org/cgi/bugreport.cgi?bug=21174