While adding the documentation here for add_email_corpus I noticed
that the other email-adding functions in test-lib.sh were not yet
documented here, so add all of that documentation.
Carl Worth [Mon, 20 Sep 2010 23:13:15 +0000 (16:13 -0700)]
test: Remove useless NOTMUCH variable (in favor of simply "notmuch")
When the NOTMUCH variable was originally invented it was used as an
explicit path to the notmuch binary being tested. Today, the test
suite sets the PATH variable instead, so the NOTMUCH variable always
has a value of simply "notmuch".
We simplifying that by using the constant value rather than the
continual variable reference.
Carl Worth [Mon, 20 Sep 2010 22:01:52 +0000 (22:01 +0000)]
lib: Fix use-after-free bug.
Thanks to the new git-based test suite, it's easy to run the whole
test suite in valgrind, (simply "make test OPTIONS="--valgrind"), and
doing so showed this obvious use-after-free bug, (triggered by the
thread-order tests).
Carl Worth [Mon, 20 Sep 2010 21:39:40 +0000 (14:39 -0700)]
test: Fix to actually report errors (!).
A bug in the results-aggregation code was causing the test suite to report
"all tests passed" even when there were failures, (as long as there were
also no "broken" tests). Fix this.
Carl Worth [Mon, 20 Sep 2010 21:36:31 +0000 (14:36 -0700)]
test: Print section names, and rename all test sections
Now that we can usefully pass section names via the NOTMUCH_SKIP_TESTS
environment variable, it's useful to actually print those names out
for the user. Then, since we're now printing these names, let's use
nicer names, (not excessively long but also not using abbreviations
like "msg").
Carl Worth [Mon, 20 Sep 2010 21:28:13 +0000 (14:28 -0700)]
test: Make the --valgrind option useful, (and drop --verbose).
In order for --valgrind to be useful, we drop noisy additional output of
all of the commands being executed in verbose mode. This makes --verbose
alone quite useless, so we don't document it any more.
Also, add a zlib valgrind suppression that was showing up frequently in the
test suite.
Carl Worth [Mon, 20 Sep 2010 21:06:38 +0000 (14:06 -0700)]
test/README: Update to become notmuch-specific rather than git-specific
This file was obviously describing the git test suite previously, and
would have been very hard to understand in the context of the notmuch
test suite. HOpefully it's easier to follow now.
Carl Worth [Mon, 20 Sep 2010 20:54:57 +0000 (13:54 -0700)]
test: Rename GIT_SKIP_TESTS to NOTMUCH_SKIP_TESTS
By scanning test-lib.sh for occurrences of "git" or "GIT", I found
that most of those are internal things, (like the GIT_TEST_TEE_STARTED
variable). But GIT_SKIP_TESTS is part of the user-interface to the
test suite, so we rename it to reference notmuch rather than git.
Also, the GIT_TRACE warning is git-specific, so we drop that as well.
Carl Worth [Mon, 20 Sep 2010 19:41:10 +0000 (12:41 -0700)]
test: Add test to ensure that all available test scripts are run
Since we are now using an explicit list of tests to run in
notmuch-test we need to be careful that we don't add a new file of
tests and then forget to add it to the list.
Carl Worth [Fri, 17 Sep 2010 22:30:17 +0000 (15:30 -0700)]
test: Remove basic testing of broken, fixed, and skipped tests.
These were interfering with the aggregate statistics reported at the
end of the test-suite run. (Always reporting 1 broken, 1 fixed, and 1
skipped). The correct way to test the test-suite itself would be to
run the test suite externally for these cases, capture the expected
result, and then report that as a PASS test.
But, really, there's almost no value in these tests anyway. It's
almost to the level of testing that 'if false; exit 1; fi' returns
1. That is, there are so many ways that the test suite could be broken
internally, that these minor tests don't really help.
Carl Worth [Fri, 17 Sep 2010 22:25:39 +0000 (15:25 -0700)]
test: Rework test-suite input to avoid ulti-command strings
The original git test suite works by concatenating many commands into
a very long string (each separated by &&). This is painful to work
with since it prevents the editor from helping by parsing the shell
script, indenting, colorizing, etc.
Instead, we switch this back to something like the original notmuch
test suite, and add two new functions to test-lib.sh
(test_begin_subtest and test_expect_equal) to support these.
This also fixes the test suite to once again display the diff when a
test fails to generate the expected input.
Carl Worth [Fri, 17 Sep 2010 20:53:47 +0000 (13:53 -0700)]
test: Cleanup the test output
This makes the new, git-derived test suite report results in a manner
similar to the original notmuch test suite.
Notable changes include:
* No more initial '*' on every line
* Only colorize a single word
* Don't print useless test numbers
* Use "PASS" in place of "ok"
* Begin sentences with a capital letter
* Print test descriptions for each block
* Separate each block of tests with a blank line
* Don't summarize counts between each block
Carl Worth [Fri, 17 Sep 2010 19:10:54 +0000 (12:10 -0700)]
test: Fix test suite to integrate with our non-recursive Makefile system.
This avoids "make test" emitting messages from three (3!) recursive
invocations of make. We change the invocations of the tests themselves
to occur directly from the shell script rather than having the shell
script invoke make again and using wildcards in the Makefile.
Michal Sojka [Thu, 10 Jun 2010 06:48:03 +0000 (08:48 +0200)]
Convert the actual tests to the new framework
The changes are:
- The notmuch-test was split into several files (t000?-*.sh).
- Removed helper functions which were moved to test-lib.sh
- Replaced every printf with test_expect_success.
- Test commands chained with && (test-lib.sh doesn't use "set -e" in
order to complete the test suite even if something fails)
- Many variables such as ${MAIL_DIR} were properly quoted as they
contain spaces.
- Changed quoting patterns in add_message and generate_message (single
quotes are already used by the test framework).
- ${TEST_DIR} replaced by ${PWD}
QUICK HOWTO:
To run the whole test suite
make
To run only a single test
./t0001-new.sh
To stop on the first error
./t0001-new.sh -i
then mail store and database can be inspected in
"trash directory.t0001-new"
To see the output of tests
./t0001-new.sh -v
To not remove trash directory at the end:
./t0001-new.sh -d
To run all tests verbosely:
make GIT_TEST_OPTS="-v"
Michal Sojka [Thu, 10 Jun 2010 06:48:02 +0000 (08:48 +0200)]
test: Update helper functions
Modify the helper functions to work with git-based test suite i.e.
1) Quote arguments where it is necessary.
2) Do not use $NOTMUCH. It is equal to "notmuch" since $PATH is set to
the build tree.
3) Modify pass_if_equal to fit into the git-based test suite.
Michal Sojka [Thu, 10 Jun 2010 06:48:01 +0000 (08:48 +0200)]
Update test framework for use with notmuch
This removes Git specific things from the test-lib.sh and adds helper
functions for notmuch taken from Carl's notmuch-test script. README is
also slightly modified to reflect the current state.
Michal Sojka [Thu, 10 Jun 2010 06:48:00 +0000 (08:48 +0200)]
Copy test framework from Git
Git uses a simple and yet powerful test framework, written in shell.
The framework is easy to use for both users and developers so I think
it would help if it is used in notmuch as well.
Carl Worth [Thu, 16 Sep 2010 22:52:12 +0000 (15:52 -0700)]
emacs: Allow '|' to operate on multiple messages (by means of prefix argument).
We extend the '|' command so that passing a prefix argument, (for
example, "C-u |"), causes it to pipe all open messages in the current
thread rather than just the single, current message.
Carl Worth [Thu, 19 Aug 2010 22:09:55 +0000 (15:09 -0700)]
emacs: Fix line-wrapping for help message of notmuch-show mode.
This was previously wrapped for unsubtituted command names. It looks
much better in the notmuch-help (available with '?') if wrapped
according to the length of the substituted command names.
The purpose of that commit was to ensure that an attempt to manipulate
a non-standard database would not inadvertently manipulate the default
database only due to a typo in the NOTMUCH_CONFIG environment
variable. That is, a command like:
NOTMUCH_CONFIG=mistyped-config-filename notmuch tag -new tag:new
shouldn't modify the database at ${HOME}/mail, but should instead
simply report that the mistype configuration filename does not exist.
We fix both cases simultaneously by reporting the error message
whenever the function calling notmuch_config_open is not explicitly
prepared for a default configuration file.
Carl Worth [Wed, 9 Jun 2010 02:45:54 +0000 (19:45 -0700)]
notmuch show: Add a --format=mbox option
We don't love the mbox format, but it's still sometimes the most
practical way to share a collection of messages as a single file.
Here we implement the "mboxrd" variant of the mbox file format. This
variant applies reversible escaping by prefixing a '>' character to
all lines in the email messages matching the regular expression:
"^>*From "
This allows the escaping to be reliably removed. A reader should remove
a '>' from any line matching the regular expression:
"^>>*From "
More details on the mboxrd formats (and others as well) can be found
here:
Carl Worth [Sat, 5 Jun 2010 15:40:26 +0000 (08:40 -0700)]
Avoid giving GMime a NULL MIME-stream filter.
Micah Anderson reported an issue where a message failed to display in
the emacs interface, (it instead gave an error, "json-read-string: Bad
string format").
Micah tracked this down to the json output from "notmuch show" being
interrupted by a GMime error message:
I tracked this down further to notmuch passing a NULL value to
g_mime_stream_filter_add. And this was due to calling
g_mime_filter_charset_new with a value of "unknown-8bit".
So we add a test message withe a Conten-Type of "text/plain;
charset=unknown-8bit" from Micah's message. Then we fix "notmuch show"
to test for NULL before calling g_mime_stream_filter_add. Bug fixed.
Carl Worth [Fri, 4 Jun 2010 23:52:56 +0000 (16:52 -0700)]
make install: Run ldconfig or install a DT_RUNPATH in binary as appropriate.
Various users were confused as to why they couldn't run notmuch
immediately after "make install", (with linker errors saying that
libnotmuch.so could not be found). The errors came from two different
causes:
1. The user had installed to a system library directory, but had not
yet run ldconfig.
2. The user had installed to some non-system directory, and had not
set the LD_LIBRARY_PATH variable.
With this change we fix both problems (on Linux) without the user
having to do anything additional. We first use ldconfig to find the
system library directories. If the user is installing to one of these,
then we run ldconfig as part of "make install".
For case (2) we use the -rpath and --enable-new-dtags linker options
to install a DT_RUNPATH entry in the binary. This entry tells the
dynamic linker where to find libnotmuch. Without the
--enable-new-dtags option only a DT_RPATH option would be installed,
(which has the drawback of not allowing any override with the
LD_LIBRARY_PATH variable).
Distributions (such as Debian and Fedora) don't want to see binaries
packaged with a DT_RPATH or DT_RUNPATH entry. This should be avoided
automatically as long as the packages install to standard locations,
(such as /usr/lib).
Carl Worth [Fri, 4 Jun 2010 23:51:32 +0000 (16:51 -0700)]
configure: Remove space from IFS (using tab as necessary)
The idea here is to more easily support filenames with spaces in them
in various loops. We're about to add a loop over the paths configured
by the dynamic linker. Hopefully, they wouldn't contain spaces, but
one never knows so we might as well be prepared.
Carl Worth [Fri, 4 Jun 2010 19:39:36 +0000 (12:39 -0700)]
Add support (and tests) for messages with really long message IDs.
Scott Henson reported an internal error that occurred when he tried to
add a message that referenced another message with a message ID well
over 300 characters in length. The bug here was running into a Xapian
limit for the length of metadata key names, (which is even more
restrictive than the Xapian limit for the length of terms).
We fix this by noticing long message ID values and instead using a
message ID of the form "notmuch-sha1-<sha1_sum_of_message_id>". That
is, we use SHA1 to generate a compressed, (but still unique), version
of the message ID.
We add support to the test suite to exercise this fix. The tests add a
message referencing the long message ID, then add the message with the
long message ID, then finally add another message referencing the long
ID. Each of these tests exercise different code paths where the
special handling is implemented.
A final test ensures that all three messages are stitched together
into a single thread---guaranteeing that the three code paths all act
consistently.
Carl Worth [Fri, 4 Jun 2010 19:38:11 +0000 (12:38 -0700)]
test suite: Generate message filenames from count, not Message-Id.
We're about to add a test with an excessively long message-id, (512
characters or so). This exceeds filename length limits, so just always
the simple counter to generate the filenames, (which we were doing for
messages with non-custom IDs anyway).
Carl Worth [Fri, 4 Jun 2010 17:16:53 +0000 (10:16 -0700)]
Avoid database corruption by not adding partially-constructed mail documents.
Previously we were using Xapian's add_document to allocate document ID
values for notmuch_message_t objects. This had the drawback of adding
a partially constructed mail document to the database. If notmuch was
subsequently interrupted before fully populating this document, then
later runs would be quite confused when seeing the partial documents.
There are reports from the wild of people hitting internal errors of
the form "Message ... has no thread ID" for example, (which is
currently an unrecoverable error).
We fix this by manually allocating document IDs without adding
documents. With this change, we never call Xapian's add_document
method, but only replace_document with either the current document ID
of a message or a new one that we have allocated.
David Edmondson [Wed, 19 May 2010 07:03:37 +0000 (08:03 +0100)]
emacs: In search mode, truncate authors using invisible text.
Rather than discarding authors when truncated to fit the defined
column width, mark the text beyond the end of the column as invisible
and allow `isearch' to be used over the text so hidden.
This allows us to retain the compact display whilst enabling a user to
find the elided text.
David Edmondson [Wed, 19 May 2010 06:44:18 +0000 (07:44 +0100)]
emacs: Allow the display of absolute dates in the header line.
Add `notmuch-show-relative-dates' to control whether the summary line
in `notmuch-show' mode displays relative dates (e.g. '26 mins. ago') or
the full date string from the message. Default to `t' for
compatibility with the previous behaviour.
Nelson Elhage [Tue, 11 May 2010 15:38:44 +0000 (11:38 -0400)]
emacs: Bind <backtab> (shift-TAB) to notmuch-show-previous-button
Shift-TAB is standard "opposite" of TAB -- in GUI interfaces they
typically cycle through input elements in opposite orders -- so it
makes sense to behave the same way.
Signed-off-by: Nelson Elhage <nelhage@ksplice.com>
David Edmondson [Tue, 4 May 2010 13:44:39 +0000 (14:44 +0100)]
emacs: Pretty print the numbers of matching messages.
Insert a separator every three digits when outputting numbers. Allow
the user to choose the separator by customizing
`notmuch-decimal-separator'. Widen the space allocated for message
counts accordingly.
Nelson Elhage [Fri, 14 May 2010 17:15:38 +0000 (13:15 -0400)]
emacs: notmuch-hello: Make widget-keymap a parent of notmuch-hello-keymap
This lets us pick up later changes to widget-keymap if the user
customizes it in some way. This is the recommended way to use
`widget-keymap', according to its help.
This enables the nifty '?' key binding to work in notmuch-hello
(although for some strange reasons I don't see any descriptions for
specific key bindings yet. Not sure how that is supposed to work
though.
But this starts, runs and behaves identical to the existing code.
Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Sebastian Spaeth [Thu, 29 Apr 2010 09:10:01 +0000 (11:10 +0200)]
emacs: Reuse rather than reinvent message header filtering
In notmuch-mua-reply we were filtering out the Subject and To headers
manually in a loop, but message mode offers a nice function for
exactly that. Simplify the code by using it. Also, as notmuch-mua-mail
already sorts and hides headers that we want sorted and hidden, we can
safely remove those 2 functions from here as well. Also remove the
(require 'cl), the only reason for its existence was the now removed
"loop" function.
Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
David Edmondson [Thu, 29 Apr 2010 08:28:23 +0000 (09:28 +0100)]
emacs: Allow tuning of the tag/saved search layout.
Add `notmuch-column-control', which has three potential sets of
values:
- t: automatically calculate the number of columns per line based on
the tags to be shown and the window width,
- an integer: a lower bound on the number of characters that will be
used to display each column,
- a float: a fraction of the window width that is the lower bound on
the number of characters that should be used for each column.
So:
- if you would like two columns of tags, set this to 0.5.
- if you would like a single column of tags, set this to 1.0.
- if you would like tags to be 30 characters wide, set this to
30.
- if you don't want to worry about all of this nonsense, leave
this set to `t'.
David Edmondson [Thu, 29 Apr 2010 07:32:31 +0000 (08:32 +0100)]
emacs: Allow control over faces for search mode columns.
Add face declarations for the date, count, matching author and subject
columns in search mode and apply those faces when building the search
mode display.
David Edmondson [Thu, 29 Apr 2010 06:52:22 +0000 (07:52 +0100)]
emacs: Display non-matching authors with a different face.
In search mode some messages don't match the search criteria. Show
their authors names with a different face - generally darker than
those that do match.
Sebastian Spaeth [Wed, 28 Apr 2010 14:31:00 +0000 (16:31 +0200)]
NEWS: describe easier FCC configuration
I know I should be writing something witty here to make cworth happy,
but I can't think for any verbose justification of this patch beyond
that submitting a NEWS blurb will make cworth happy too. So let's make
him happy.
Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Sebastian Spaeth [Wed, 28 Apr 2010 14:30:59 +0000 (16:30 +0200)]
Easier way to define a fcc directory
In the common case that a user only has one FCC (save outgoing mail in
the Mail directory, it is now possible to simply configure a string
such as "Sent" in the notmuch-fcc-dirs variable. More complex options,
depending on a users email address, are possible and described in the
variable customization help text.
The whole function notmuch-fcc-header-setup has been cleaned up a
little while working on that.
Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
David Edmondson [Wed, 28 Apr 2010 11:18:32 +0000 (12:18 +0100)]
emacs: Usability improvements for `notmuch-hello'.
- If no saved searches exist or are displayed, don't signal an error,
- If no saved searches exist or are displayed, leave the cursor in the
search bar,
- Minor layout improvements.
Carl Worth [Thu, 3 Jun 2010 23:43:21 +0000 (16:43 -0700)]
Add a test case for the previous commit.
The commit said it fixed a problem with headers >200 characters
long. But examination of the code suggests that it was a header of
exactly 200 characters long that caused the problem. So we add a test
case for that here.
Before the fix in the previous commit, valgrind would detect many
errors when replying to the message created with this test case. After
that commit, those errors are gone.
Carl Worth [Tue, 1 Jun 2010 17:46:52 +0000 (10:46 -0700)]
Makefile: Improve the "what to do now" message from "make install"
This was already telling the user how to run notmuch within emacs, but
not how to just run the notmuch command-line interface, (which, as it
turns out, is a prerequisite for running the emacs interface anyway).
Carl Worth [Tue, 1 Jun 2010 17:44:32 +0000 (10:44 -0700)]
INSTALL/README: Clean up the description of how to run the emacs interface.
The INSTALL file still had old information about the "make
install-emacs" command which no longer exists. README was also giving
pointers on how to develop a real interface, (which is not the right
thing since README should be addressed to users, not coders).
So remove the stale and misplaced information, and instead add a new
"Running notmuch" section to the README describing how to run the
notmuch command-line interface and how to run the emacs interface.
Carl Worth [Tue, 1 Jun 2010 17:42:43 +0000 (10:42 -0700)]
INSTALL: Note the advantages of Xapian 1.0.18+ and 1.1.4+
These versions provide greatly desired performance advantages for
notmuch.
Previously, theses details existed in an old NEWS entry, but most
users are unlikely to find those details there. Put them here where we
mention the Xapian dependency.
Carl Worth [Tue, 1 Jun 2010 17:40:38 +0000 (10:40 -0700)]
TODO: Add idea for fixing "notmuch count" to always be exact.
I had previously thought Xapian only offered an estimate for the
number of results that might match a search. But Olly let me know
that we can easily ask for Xapian to provide the exact count.
Sebastian Spaeth [Sun, 16 May 2010 22:56:27 +0000 (15:56 -0700)]
python: have docs reflect current return value behavior
Database.find_message() used to be able to reliably indicate whether a
message exists or not (in which case it returns None). However, the
recent API change of the notmuch library means we will return None
even for all Xapian exceptions, which happens e.g. when the current
Database has been modified by another project. Therefore the return
value of None cannot be reliably be used to indicate whether a message
exists or not. Make the docs state that explicitely.
Signed-off-by: Sebastian Spaeth <Sebastian@SSpaeth.de>
Carl Worth [Wed, 28 Apr 2010 00:04:52 +0000 (17:04 -0700)]
debian: Instruct git-buildpackage that it's OK to build from this branch
I wish I had something with better support for a native Debian package
here. I shouldn't ever have to configure any branch---I just want it
to build a package from the current branch. Instead it makes me tell
it (twice!) what the current branch actually is.
Carl Worth [Tue, 27 Apr 2010 23:56:24 +0000 (16:56 -0700)]
NEWS: Reword the latest bug description slightly.
"Still needs to be handled correctly" could be misread to suggest that
the bug has not actually been fixed yet. So clarify what is actually
meant here, (that the bug is unlikely but we're still motivated to fix
it).
Fix SEGV in _thread_cleanup_author if author ends with ', '
Admittedly, an author name ending in ',' guarantees this is spam, and
indeed this was triggered by a spam email, but that doesn't mean we
shouldn't handle this case correctly.
We now check that there is actually a component of the name (presumably
the first name) after the comma in the author name.
Carl Worth [Tue, 27 Apr 2010 18:40:26 +0000 (11:40 -0700)]
Merge branch '0.3.x'
This doesn't pull in any code, (everything in 0.3.x was originally
cherry-picked from master anyway). But the merge does give us a
correct NEWS file showing which fixes are included in 0.3.1 and which
features have been commited "since" then, (topologically, not
chronologically).
Tomas Carnecky [Tue, 27 Apr 2010 17:27:17 +0000 (19:27 +0200)]
Wrap the compat header in extern "C" { } when compiling C++ sources
This fixes a build error on OpenSolaris where the final liking of
notmuch fails because the linker can't find strcasestr() referenced
from thread.cc.
(cherry picked from commit aab54b4ce752b16725399d1543602ef3f7736c91)
emacs: Remove conditional from notmuch-fcc-initialization.
The fcc code would only initialize if notmuch-fcc-dirs was set. This was
a problem if you reset the variable, or added the variable later during
initialization. Now we always add the fcc hook, but it doesn't do
anything unless notmuch-fcc-dirs are set.
(cherry picked from commit 80a90787163690d2d87571327ba504a470798c60)
David Edmondson [Tue, 27 Apr 2010 10:04:36 +0000 (11:04 +0100)]
emacs: If 'all tags' is not shown, don't use it when calculating widths.
If the 'all tags' section of the hello buffer will not be shown, don't
consider those tags when determining the number of saved searches that
can be displayed on a single line.
(cherry picked from commit 18d41192d2cf0c71045c2f6420eb2311553f58b4)
Carl Worth [Tue, 27 Apr 2010 17:13:04 +0000 (10:13 -0700)]
test: Exercise magic-from guessing with a single configured address
Immediately after releasing 0.3 we learned that the magic-from-guessing
code could hang in an infinite loop in some cases. The bug occurred
only when the user had configured only a primary email addresss and no
other email addresses.
Carl Worth [Tue, 27 Apr 2010 17:13:04 +0000 (10:13 -0700)]
test: Exercise magic-from guessing with a single configured address
Immediately after releasing 0.3 we learned that the magic-from-guessing
code could hang in an infinite loop in some cases. The bug occurred
only when the user had configured only a primary email addresss and no
other email addresses.
The test suite wasn't previously covering this case, so address this
shortcoming.
emacs: Remove conditional from notmuch-fcc-initialization.
The fcc code would only initialize if notmuch-fcc-dirs was set. This was
a problem if you reset the variable, or added the variable later during
initialization. Now we always add the fcc hook, but it doesn't do
anything unless notmuch-fcc-dirs are set.
Add a (require 'notmuch-message) to notmuch.el. This is for functions that
specifically target message mode (and, in the future, notmuch-message
mode).
Add `notmuch-message-mark-replied', a function for automatically tagging
replied messages with user-defined tags. The tags (which can be either
added or removed) can be customized with the customization variable
`notmuch-message-replied-tags'. This is a simple list of strings. Any
string prefaced with a "-" will be removed; any string prefaced with a "+"
(or neither "+" nor "-") will be added.
This adds a new file notmuch-message.el, for functions which target
message mode (and in the future, notmuch-message mode). Based on some
conversation, notmuch-message.el will probably end up subsuming
notmuch-mua.el, but until we figure out exactly how we want to do that,
they will remain separate files.
Edited-by: Carl Worth <cworth@cworth.org>: Remove trailing whitespace
and add newline at end of file.
Detect inline patches and convert them to fake attachments, in order
that `diff-mode' highlighting can be applied to the patch. This can be
enabled by customising `notmuch-show-insert-text/plain-hook'.