The object where pointer to `data` was received was deleted before
it was used in _notmuch_string_list_append().
Relevant Coverity messages follow:
3: extract
Assigning: data = std::__cxx11::string(message->doc.()).c_str(),
which extracts wrapped state from temporary of type std::__cxx11::string.
4: dtor_free
The internal representation of temporary of type std::__cxx11::string
is freed by its destructor.
5: use after free:
Wrapper object use after free (WRAPPER_ESCAPE)
Using internal representation of destroyed object local data.
(cherry picked from commit 06adc27668)
Apparently some systems (MacOS?) have a system library called libutil
and the name conflict causes problems. Since this library is quite
notmuch specific, rename it to something less generic.
The object where pointer to `data` was received was deleted before
it was used in _notmuch_string_list_append().
Relevant Coverity messages follow:
3: extract
Assigning: data = std::__cxx11::string(message->doc.()).c_str(),
which extracts wrapped state from temporary of type std::__cxx11::string.
4: dtor_free
The internal representation of temporary of type std::__cxx11::string
is freed by its destructor.
5: use after free:
Wrapper object use after free (WRAPPER_ESCAPE)
Using internal representation of destroyed object local data.
Internet_address_list_to_string() and
g_mime_message_get_date_as_string() return allocated string buffers
and not const, so from what I can tell from taking a look at the
sprinter-sexp.c’s sexp_string() function, the code leaks the
recipients_string as well as the date string.
Often Linux systems are shipped with wget(1) by default (and no curl).
Many BSDs, macOS, and e.g. some Linux minimal/container images
comes with curl(1) (and no wget).
Attempting to download with curl if wget is not available increases
the likelihood for this to succeed.
For reasons not completely understood at this time, gmime (as of
2.6.22) is returning a date before 1900 on bad date input. Since this
confuses some other software, we clamp such dates to 0,
i.e. 1970-01-01.
Adjusted notmuch-crypto gpg call-process function to respect
the GPG program set by the EasyPG epg-gpg-program variable.
This is to correct a problem observed on NixOS where only gpg2 is
installed by default. The Notmuch Emacs frontend fails when trying to
access someone's key to verify their signature when it cannot find the
GPG binary.
In addition to use ${srcdir} and deliver ${NOTMUCH_SRCDIR} where needed,
source from ruby bindings had to be copied to the out-of-tree target
directory -- if the source files in source directory were referenced
in build and there were also built object files there, those could have
been considered as target files (and then not found when attempting
to create bindings/ruby/notmuch.so).
The ${srcdir} -- usually relative path to notmuch source -- works fine
in current ./configure and all makefiles. To have simple access to
notmuch source in tests and out of tree builds holding absolute path to
the source directory is useful.
Some gdb python exceptions on some os environments (e.g. macOS
Sierra, non-codesigned gdb) do not make gdb exit (but to drop down
to nonexistent command line?).
Mitigate this chance by explict SystemExit on all exceptions.
The contents of output file 'gdb.out' is unchanged.
Now make goal combinations starting with *clean goals,
ending with *clean coals, and having non-*clean goals in between
should work. What does not expected to work are
non-*clean - *clean - *non-clean goal combinations.
Also, if first goals are *clean goals, re-creation of Makefile.config
is inhibited when Makefile.config exists and ./configure is newer.
The formatter structs are only needed for the formatter array
initialization. Move them closer to use. This also lets us drop some
forward declarations. No functional changes.
The --entire-thread option handling is split around, making the logic
harder to follow than necessary. Put it in one place. While at it,
make the true/false values match notmuch_bool_t values for
simplicity. No functional changes.
The mixed use of the format pointer and the format selection variables
is confusing. Add more clarity by using format_sel alone. No
functional changes.
The raw member has been unused since b1130bc71c ("show: Convert raw
format to the new self-recursive style, properly support interior
parts"). Good riddance. No functional changes.
The use of params.part has become rather convoluted in notmuch
show. Add another variable for selecting single message display to
make the code easier to read. No functional changes.
When pkg-config does not find configure, a compat version of the
zlib.pc is created. In creation of that configure attempted to
read values of $zlib_cflags and $zlib_ldflags. In the usual case
those were undefined, and with `set -a` now in the beginning of
configure, configure broke.
Even if $zlib_cflags and $zlib_ldflags had values which were used
to create zlib.pc, the values were overwritten (with static content)
a few lines later in next pkg-config --cflags and --libs run. These
values would not be different and probably useless -- the following
boild would probably fail.
But instead of using those, CPATH and LIBRARY_PATH environment
variables can be used successfully (both while configuring and
building).
Unify the subtests by requiring test_begin_subtest before
test_expect_code. (Similar change for test_expect_success has already
been done.)
This increases clarity in the test scripts by having a separate line
for the start of the subtest with the heading, and makes it possible
to simplify the test infrastructure by making all subtests similar.
Unify the subtests by requiring test_begin_subtest before
test_expect_success. (Similar change for test_expect_code will
follow.)
This increases clarity in the test scripts by having a separate line
for the start of the subtest with the heading, and makes it possible
to simplify the test infrastructure by making all subtests similar.
The only place where we use the implicit prereq check is T000-basic.sh
where we check that it works. It's an added complication that we don't
use. Remove it.
The test_have_prereq function can still be used for the same effect in
subtests that use test_begin_subtest. For now, this will make it
impossible to have prereqs in one-line subtests that don't require
test_begin_subtest. This will be fixed in follow-up work.
Known broken tests are, well, known broken. Do not print the result
diff for them unless V=1 is specified. Now that the test description
is printed also when known broken tests fail, the user can also skip
to running the individual failing tests.
