Since most memory allocation is (ultimately) in the talloc context
defined by a notmuch_database_t pointer, this gives a more complete
view of memory still allocated at program shutdown.
We also change the talloc report in notmuch.c to mode "a" to avoid
clobbering the newly reported log.
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.
Although this default worked for "notmuch config get", it didn't work
most other places. Restore the previous functionality, with the
wrinkle that XDG locations will shadow $HOME/mail if they exist.
This fixes a bug reported by Jack Kamm in id:87eeefdc8b.fsf@gmail.com
In principle this could be done without depending on C++11 features,
but these features should be available since gcc 4.8.1, and this
localized usage is easy to replace if it turns out to be problematic
for portability.
Prior to 0.32, notmuch had the (undocumented) behaviour that it
expanded a relative value of database.path with respect to $HOME. In
0.32 this was special cased for database.path but broken for
database.mail_root, which causes problems for at least notmuch-new
when database.path is set to a relative path.
The change in T030-config.sh reflects a user visible, but hopefully
harmless behaviour change; the expanded form of the paths will now be
printed by notmuch config.
When compat canonicalize_file_name was introduced, it was limited to
C code only because it was used by C code only during that time.
>From 5ec6fd4d, (lib/open: check for split configuration when creating
database., 2021-02-16), lib/open.cc, which is C++, relies on the
existent of canonicalize_file_name.
However, we can't blindly enable canonicalize_file_name for C++ code,
because different implementation has different additional signature for
C++ and users can arbitrarily add -DHAVE_CANONICALIZE_FILE_NAME=0 to
{C,CXX}FLAGS.
Let's move our implementation into a util library.
Helped-by: Tomi Ollila <tomi.ollila@iki.fi>
Signed-off-by: Đoàn Trần Công Danh <congdanhqx@gmail.com>
Ignoring this return value seems like a bad idea in general, and in
particular it has been hiding one or more bugs related to handling
long directory names.
This is intended to fix the slow behaviour of "notmuch new" (and possibly
"notmuch reindex") when large numbers of files are deleted.
The underlying issue [1] seems to be the Xapian glass backend spending
a large amount of time in db.has_positions when running queries with
large-ish amounts of unflushed changes.
This commit removes two uses of Xapian queries [2], and replaces them with
an approximation of what Xapian would do after optimizing the
queries. This avoids the calls to has_positions (which are in any case
un-needed because we are only using boolean terms here).
[1] Thanks to "andres" on IRC for narrowing down the performance
bottleneck.
[2] Thanks to Olly Betts of Xapian fame for talking me a through a fix
that does not require people to update Xapian.
Since the library searches in several locations for a config file, the
caller does not know which of these is chosen in the usual case of
passing NULL as a config file. This changes provides an API for the
caller to retrieve the name of the config file chosen. It will be
tested in a following commit.
Eventually we want to do all opening of databases in the top
level (main function). This means that detection of missing databases
needs to move out of subcommands. It also requires updating the
library to use the new NO_DATABASE status code.
This matches functionality in the the CLI function
notmuch_config_get_database_path, which was previously used in the CLI
code for all calls to open a database.
The layer of shims here seems a bit wasteful compared to just calling
the corresponding string map functions directly, but it allows control
over the API (calling with notmuch_database_t *) and flexibility for
future changes.
_trial_open can't know if the PATH_ERROR return value will cause the
error message to be returned from the library, so it's up the caller
to clean up if not.
Like the hook directory, we primarily need a way to communicate this
directory between various components, but we may as well let the user
configure it.
Most of the diff is generalizing choose_hook_dir to work for both
backup and hook directories.
Choose sibling directory of xapian database, as .notmuch may not
exist.
libgen.h is already used in debugger.c, so it is not a new dependency
/ potential portability problem.
This changes some error reporting, either intentionally by reporting
the highest level missing directory, or by side effect from looking in
XDG locations when given null database location.
The main functionality will be tested when notmuch-new is converted to
support split configuration. Here only the somewhat odd case of split
mail root which is actually symlinked to the database path is tested.
Introduce a new configuration value for the mail root, and use it to
locate mail messages in preference to the database.path (which
previously implied the mail messages were also in this location.
Initially only a subset of the CLI is tested in a split
configuration. Further changes will be needed for the remainder of the
CLI to work in split configurations.
The idea is to allow reuse in n_d_create_with_config. This is
primarily code movement, with some changes in error messages to reduce
the number of input parameters.
This is slightly more tidy, but more importantly it allows for re-use
of this code in n_d_create_with_config. That re-use will be crucial
when we no longer call n_d_open_with_config from
n_d_create_with_config.
This removes duplication between the struct element and the
configuration string_map entry. Create a simple wrapper for setting
the database path that makes sure the trailing / is stripped.
In the future Xapian will apparently support this more conveniently
for the cases other than READ_ONLY => READ_ONLY
Conceptually this function seems to fit better in lib/open.cc;
database.cc is still large enough that moving the function makes
sense.
This will allow re-opening in a different mode (read/write
vs. read-only) with current Xapian API. It will also prove useful when
updating the compact functions to support more flexible database
location.
Include the (currently unused) mode argument which will specify which
mode to re-open the database in. Functionality and docs to be
finalized in a followup commit.
Based on a patch from Michael J Gruber [1]. As of glib 2.67 (more
specifically [2]), including "gmime-extra.h" inside an extern "C"
block causes build failures, because glib is using C++ features.
Observing that "gmime-extra.h" is no longer needed in
notmuch-private.h, which can simply delete that include, but
we have to correspondingly move the includes which might include
it (in particular crypto.h) out of the extern "C" block also.
This seems less fragile than only moving gmime-extra, and relying on
preprocessor sentinels to keep the deeper includes from happening.
Move to the include to the outside of the extern block.
[1]: id:aee618a3d41f7889a7449aa16893e992325a909a.1613055071.git.git@grubix.eu
[2]: https://gitlab.gnome.org/GNOME/glib/-/merge_requests/1715
The hook directory configuration needs to be kept in synch with the
other configuration information, so add scaffolding to support this at
database opening time.
This will allow client code to provide more meaningful diagnostics. In
particular it will enable "notmuch new" to continue suggsting the user
run "notmuch setup" to create a config after "notmuch new" is
transitioned to the new configuration framework.
By using an enum we can have better error detection than copy pasting
key strings around.
The question of what layer this belongs in is a bit
tricky. Historically most of the keys are defined by the CLI. On the
other hand features like excludes are supported in the
library/bindings, and it makes sense to configure them from the
library as well.
The somewhat long prefix for notmuch_config_t is to avoid collisions
with the existing usage in notmuch-client.h.
Fill in the remainder of the documented functionality for
n_d_open_with_config with respect to config file location. Similar
searching default locations of the database file still needs to be
added.
The main goal is to allow configuration information to be temporarily
overridden by a separate config file. That will require further
changes not in this commit.
The performance impact is unclear, and will depend on the balance
between number of queries and number of distinct metadata items read
on the first call to n_d_get_config.
database.cc is uncomfortably large, and some of the static data
structures do not need to be shared as much as they are.
This is a somewhat small piece to factor out, but it will turn out to
be helpful to further refactoring.
As diagnosed by Olivier Taïbi in
id:20201027100916.emry3k2wujod4xnl@galois.lan, if an exception is
thrown while the initialization is happening (e.g. if the function is
called on a closed database), then the destructor is (sometimes)
invoked on an uninitialized Xapian object.
Solve the problem by moving the setting of the destructor until after
the placement new successfully completes. It is conceivable this might
cause a memory leak, but that seems preferable to crashing, and in any
case, there seems to be nothing better to be done if the
initialization is failing things are in an undefined state by
definition.
Use `makefile-gmake-mode' instead of `makefile-mode' because the
former also highlights ifdef et al. while the latter does not.
"./Makefile.global" and one "Makefile.local" failed to specify any
major mode at all but doing so is necessary because Emacs does not
automatically figure out that these are Makefiles (of any flavor).
static_cast is a bit tricky to understand and error prone, so add a
second pointer to (potentially the same) Xapian database object that
we know has the right subclass.
I'm not sure what the point of modifying that right before destroying
the object is. In a future commit I want to remove that element of the
object, so simplify that task.
The API docs promise to handle relative filenames, but the code did
not do it.
Also check for files outside the mail root, as implied by the API
description.
This fixes the bug reported at
id:87sgdqo0rz.fsf@tethera.net
In order to mimic the "best effort" API of Xapian to provide
information from a closed database when possible, do not
destroy the Xapian database object too early.
Because the pointer to a Xapian database is no longer nulled on close,
introduce a flag to track whether the notmuch database is open or not.
The original generic handler had an extra '%s' in the format
string. Update tests that failed to catch this because the template to
print status strings checked 'stat', which was not set.
As a side effect, we revert the switch from notmuch_bool_t to bool
here. This is because those two types are not actually compatible when
passing by reference.
It's not very nice to return FALSE for an error, so provide
notmuch_message_get_flag_st as a migration path.
Bump LIBNOTMUCH_MINOR_VERSION because the API is extended.
Currently I don't know of a good way of testing this, but at least in
principle a Xapian exception in _notmuch_message_{add,remove}_term
would cause an abort in the library.
This should not change functionality, but does slightly reduce code
duplication. Perhaps more importantly it allows consistent changes to
all of the similar exception handling in message.cc.
This will be mandatory as of Xapian 1.5. The API is also more
consistent with the FieldProcessor API, which helps code re-use a bit.
Note that this switches to using the built-in Xapian support for
prefixes on ranges (i.e. deleted code at beginning of
ParseTimeRangeProcessor::operator(), added prefix to constructor).
Another side effect of the migration is that we are generating smaller
queries, using one OP_VALUE_RANGE instead of an AND of two OP_VALUE_*
queries.
As we prepare to handle S/MIME-encrypted PKCS#7 EnvelopedData (which
is not multipart), we don't want to be limited to passing only
GMimeMultipartEncrypted MIME parts to _notmuch_crypto_decrypt.
There is no functional change here, just a matter of adjusting how we
pass arguments internally.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
When we are indexing, we should treat SignedData parts the same way
that we treat a multipart object, indexing the wrapped part as a
distinct MIME object.
Unfortunately, this means doing some sort of cryptographic
verification whose results we throw away, because GMime doesn't offer
us any way to unwrap without doing signature verification.
I've opened https://github.com/jstedfast/gmime/issues/67 to request
the capability from GMime but for now, we'll just accept the
additional performance hit.
As we do this indexing, we also apply the "signed" tag, by analogy
with how we handle multipart/signed messages. These days, that kind
of change should probably be done with a property instead, but that's
a different set of changes. This one is just for consistency.
Note that we are currently *only* handling signedData parts, which are
basically clearsigned messages. PKCS#7 parts can also be
envelopedData and authEnvelopedData (which are effectively encryption
layers), and compressedData (which afaict isn't implemented anywhere,
i've never encountered it). We're laying the groundwork for indexing
these other S/MIME types here, but we're only dealing with signedData
for now.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
strncmp looks for a prefix that matches, which is very much not what
we want here. This fixes the bug reported by Franz Fellner in
id:1588595993-ner-8.651@TPL520
Xapian 1.4 is over 3 years old now (1.4.0 released 2016-06-24),
and 1.2 has been deprecated in Notmuch version 0.27 (2018-06-13).
Xapian 1.4 supports compaction, field processors and retry locking;
conditionals checking compaction and field processors were removed
but user may want to disable retry locking at configure time so it
is kept.
Apparently doxygen needs its comments formatted in a specific way to
notice that the group is closed.
Without this fix, with doxygen 1.8.16-2 we see:
```
doxygen ./doc/doxygen.cfg
…/notmuch/lib/notmuch.h:2322: warning: end of file while inside a group
```
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
The documentation for notmuch_config_list_key warns that that the
returned value will be destroyed by the next call to
notmuch_config_list_key, but it neglected to mention that calling
notmuch_config_list_value would also destroy it (by calling
notmuch_config_list_key). This is surprising, and caused a use after
free bug in _setup_user_query_fields (first noticed by an OpenBSD
porter, so kudos to the OpenBSD malloc implementation). This change
fixes that use-after-free bug.
When encountering a message that has been mangled in the "mixed up"
way by an intermediate MTA, notmuch should instead repair it and index
the repaired form.
When it does this, it also associates the index.repaired=mixedup
property with the message. If a problem is found with this repair
process, or an improved repair process is proposed later, this should
make it easy for people to reindex the relevant message. The property
will also hopefully make it easier to diagnose this particular problem
in the future.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
When we notice a legacy-display part during indexing, it makes more
sense to avoid indexing it as part of the message body.
Given that the protected subject will already be indexed, there is no
need to index this part at all, so we skip over it.
If this happens during indexing, we set a property on the message:
index.repaired=skip-protected-headers-legacy-display
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
Our _notmuch_message_crypto_potential_payload implementation could
only return a failure if bad arguments were passed to it. It is an
internal function, so if that happens it's an entirely internal bug
for notmuch.
It will be more useful for this function to return whether or not the
part is in fact a cryptographic payload, so we dispense with the
status return.
If some future change suggests adding a status return back, there are
only a handful of call sites, and no pressure to retain a stable API,
so it could be changed easily. But for now, go with the simpler
function.
We will use this return value in future patches, to make different
decisions based on whether a part is the cryptographic payload or not.
But for now, we just leave the places where it gets invoked marked
with (void) to show that the result is ignored.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
This adds no functionality directly, but is a useful starting point
for adding new repair functionality.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
When indexing the cleartext of an encrypted message, record any
protected subject in the database, which should make it findable and
visible in search.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
This should not change the indexing process yet as nothing calls
_notmuch_message_gen_terms with a user prefix name. On the other hand,
it should not break anything either.
_notmuch_database_prefix does a linear walk of the list of (built-in)
prefixes, followed by a logarithmic time search of the list of user
prefixes. The latter is probably not really noticable.
This will be used to avoid needing a database access to resolve a db
prefix from the corresponding UI prefix (e.g. when indexing). Arguably
the setup of the separate header map does not belong here, since it is
about indexing rather than querying, but we currently don't have any
other indexing setup to do.
Previously this functioned scanned every term attached to a given
Xapian document. It turns out we know how to read only the terms we
need to preserve (and we might have already done so). This commit
replaces many calls to Xapian::Document::remove_term with one call to
::clear_terms, and a (typically much smaller) number of calls to
::add_term. Roughly speaking this is based on the assumption that most
messages have more text than they have tags.
According to the performance test suite, this yields a roughly 40%
speedup on "notmuch reindex '*'"
Without this,
$ make time-test OPTIONS=--small
leads to fatal errors from too many open files.
Thanks to st-gourichon-fid for bringing this problem to my attention in IRC.
Rather than storing the lower level stdio FILE object, we store a
GMime stream. This allows both transparent decompression, and passing
the stream into GMime for parsing. As a side effect, we can let GMime
close the underlying OS stream (indeed, that stream isn't visible here
anymore).
This change is enough to get notmuch-{new,search} working, but there is still
some work required for notmuch-show, to be done in a following commit.
This is a functional change, not a straight translation, because we
are no longer directly invoking g_mime_parser_options_get_default(),
but the GMime source has indicated that the options parameter for
g_mime_parser_construct_message() is "nullable" since upstream commit
d0ebdd2ea3e6fa635a2a551c846e9bc8b6040353 (which itself precedes GMime
3.0).
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
Several GMime 2.6 functions sprouted a change in the argument order in
GMime 3.0. We had a compatibility layer here to be able to handle
compiling against both GMime 2.6 and 3.0. Now that we're using 3.0
only, rip out the compatibility layer for those functions with changed
argument lists, and explicitly use the 3.0 argument lists.
Signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
Several of these #defines were not actually used in the notmuch
codebase any longer. And as of GMime 3.0, g_mime_init takes no
arguments, so we can also drop the bogus RFC2047 argument that we were
passing and then #defining away.
signed-off-by: Daniel Kahn Gillmor <dkg@fifthhorseman.net>
