Age | Commit message (Collapse) | Author |
|
Submitted by: Curtis Hamilton
(cherry picked from commit 55ffac651c9996916680cdc05b32eb1b091a5ccf)
Change-Id: I160ff9369a563d58469682ed96e25c792184708c
|
|
Change-Id: I81877e46cf95dcc7de16b797fca33658036bf9e8
|
|
VS2013 does not like stack-allocated arrays with runtime size
Change-Id: I58bd3a9feb6e2bef1b4a4f775669e5e3e03de5f7
Reviewed-on: https://gerrit.libreoffice.org/32922
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Ia66415ff9ee67958e84cfff7eda082cdbfc9b9b9
|
|
since "trace" is such a generic term, this makes it easier to actually find the feature when you need it.
And add feature to limit stack depth of the reported backtrace.
Change-Id: Iab3e4ceb2e8480e7b5e2b920eb6c5d7631e21c43
Reviewed-on: https://gerrit.libreoffice.org/31752
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: Iefc33784f21e7a0b88c8d6308618926e38ab8554
|
|
Change-Id: Idb4bfc97e7d758c58118742affd9a9345c7a99f2
Reviewed-on: https://gerrit.libreoffice.org/32491
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
...at least macOS doesn't like that; change from OSL_TRACE (effectively doing
nothing unless running a debug build with SAL_LOG=+INFO) to SAL_INFO (calling
std::stringstream -> malloc in debug builds, regardless of SAL_LOG env var) in
a005fbeddc4e615cfff56a9bd84355f8d42c1c8c "OSL_TRACE->SAL in framework..sal" made
the mistake obvious.
Change-Id: I43d35524a0861998a1ef58c6f448085a2fb01cd5
|
|
The target file and the option are controlled by SAL_LOG_FILE. The
variable should contain the target file in a system dependent notation.
Change-Id: Ie1ce9c39740c8b7a989e5af222be21a3e3142084
Reviewed-on: https://gerrit.libreoffice.org/32176
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
|
|
Change-Id: I8bcea5ffc74d48148bea78da8c17744e288c069a
Reviewed-on: https://gerrit.libreoffice.org/32004
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
The only actual use of fcvt across the codebase has been removed in 2002 with
31cb31335f3bf7e6e57d70ca61bf5a2a536d1e6e "GetStringFromDouble: replaced that
fcvt() and temporary LocaleDataWrapper instance crap" even before this
implementation for FREEBSD was introduce in 2003 with
c61bf41c39daa97d30016eedcd7a0b687c2d8558 "INTEGRATION: CWS ooo11rc: join: from
ooo11beta2".
Change-Id: Ieea9499f001c6eac6383c344afb9050dfc70f43c
|
|
Change-Id: Ib1fc80c79355322489b7ebd37cbbbbb6614ec67a
|
|
Change-Id: I345343f84f66299e8b91b31af0290ed0c622bf8d
|
|
Change-Id: Id9077ea29e7028c12f40abfd4c57756c863e3cd9
|
|
Change-Id: I9a897af88aa9f6f7ca98ce521c69b5a4ee8462e9
Reviewed-on: https://gerrit.libreoffice.org/31903
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
As per sberg' suggestion
Limit it to == and !=, because some people like the flow of inequalities
like: "0 < a && a < 42"
The changes to sal/ were made using the rewriter.
The rewriter still has one bug, in pipe.cxx, it managed to pick up
some random piece of macro. No idea why.
Change-Id: I01305f9c5396a4b6c7421d6e92f1b4b529388e82
Reviewed-on: https://gerrit.libreoffice.org/30962
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
Change-Id: I78e85db2794b01286be49e1be70b53ac4f058cbe
|
|
on windows vista and higher only
Change-Id: I19cc63ad7e96b58fa8eb341801f5ad55260abf85
Reviewed-on: https://gerrit.libreoffice.org/31021
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
At least on my macOS 10.12.1, PTHREAD_STACK_MIN is 8192 (so stacksize was
substantially smaller than 12MB), and without this change there were randomly
looking failures in e.g. JunitTest_chart2_unoapi with a -fsanitize=address
build.
Change-Id: Icfe989a0e5097a9a0ae76c6e0f6ffcca18271245
|
|
Change-Id: I96634c18ba6600f8f7d5d04a162dcd7e1e312923
Reviewed-on: https://gerrit.libreoffice.org/30474
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
...since 97354578d7195bce927f0c00c4e2ae9cd7344776 "Remove the obsolete
functionality originally underlying osl/diagnose.h"
Change-Id: I7c920bca9cdddb37fcbbc15d8629d01c21eb0787
|
|
Change-Id: Idb133e442b881d263fcb83a4858c4a08b451c0f0
Reviewed-on: https://gerrit.libreoffice.org/30032
Reviewed-by: Noel Grandin <noel.grandin@collabora.co.uk>
Tested-by: Noel Grandin <noel.grandin@collabora.co.uk>
|
|
...ever since 010f71130b0d55079d01fbda38d45901de98b133 "INTEGRATION: CWS sal01:
#105958# Remove usage of wininet.dll autodial functions" wrapped its central
code in
#ifdef SOCKET_USE_AUTODIAL
instead of removing it properly.
Change-Id: I4c077c71ce8eda3a3f8b83286178d0d7907a6e02
Reviewed-on: https://gerrit.libreoffice.org/29742
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
...when included as <time.h> (our rather sick convention, for include files
other than standard headers) rather than "time.h"
Change-Id: I5aacfb81de95ae4254e39badffc9c83c7ac0f58f
|
|
Change-Id: If04d3f07dc8bffac8887773b5424c09848cd8f1a
Reviewed-on: https://gerrit.libreoffice.org/29404
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: Iad0a2a7ba68b3f43c22e0b5a3cee646f8baba30e
|
|
regression introduced with c89f22ffcd4191f7cb8ef0e68adb2a0f0ec6c74e
"INTEGRATION: CWS pj72"
Change-Id: Icea8bc43e346dbe4cf7cbfdf1176704d52b339f4
|
|
Change-Id: I8547760d4ea24205e21f215f25f0c58757bd9329
|
|
Change-Id: I2644079cafa46db8689463e1d0abc1cf3bdf49de
Reviewed-on: https://gerrit.libreoffice.org/28222
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Tor Lillqvist <tml@collabora.com>
|
|
Change-Id: I020c1def83d7cc37f9d5e14f2fefacb31a56bbbf
|
|
Often the choice whether some particular message is displayed with
SAL_INFO or SAL_WARN has not necessarily been that well thought
through. Many SAL_WARNs are not actually warnings but purely
informative. If you want to see the INFOs for an area it makes sense
to want to see also the WARNs for it. So make it so, unless a specific
WARN selection is specified.
Change-Id: I6286c5f856b29e68bedb7f20efadf56953f94b72
|
|
Outputs a timestamp in decimal seconds (with millisecond accuracy).
Simplified the handling of SAL_LOG if no "level" is specified. Now
just a totally unset (or empty) SAL_LOG causes the default of "+WARN"
to be used. Given how the code works, it would have become too
unwieldy to check for all combinations of TIMESTAMP and RELATIVETIMER
but no WARN or INFO.
Change-Id: I7bb5bb665d4e764e7eee447e93486f6467042e97
|
|
Change-Id: I586c11d4fe61b56624ee158c04e365cfb0cc4012
|
|
Change-Id: I9bdcd8b2d7b46a087d7f5461b6f7c3d9f02ec084
|
|
Change-Id: I4504939a3957606979c6ac36af6611e1fe072d01
Reviewed-on: https://gerrit.libreoffice.org/27902
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Noel Grandin <noelgrandin@gmail.com>
|
|
Change-Id: I87e9ce45cf945e75e8140a9d4608da8abcddada6
Reviewed-on: https://gerrit.libreoffice.org/27187
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Use parentheses to clarify the code.
Change-Id: I864dc6dacadb5b9ba9dca8e0abd9fa4e6db1eddc
Reviewed-on: https://gerrit.libreoffice.org/25677
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
instead of being randomly embedded somewhere deep inside the CXX code.
Found with:
git grep -nP '^#include' -- *.cxx | sort -g -k 2 -t :
Change-Id: I9ee432d3b665ecb6ec600bd51cc4b735a1b1127a
Reviewed-on: https://gerrit.libreoffice.org/26764
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Caolán McNamara <caolanm@redhat.com>
Tested-by: Caolán McNamara <caolanm@redhat.com>
|
|
Change-Id: I8b0fe9354232a7b60f3605fa6f90f6741f7bf683
Reviewed-on: https://gerrit.libreoffice.org/26537
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: jan iversen <jani@documentfoundation.org>
|
|
This seems to be a good idea based on several discussions
in the project. In the end catching SEH exceptions is just
going to cause strange platform dependent bahavior.
This patch is based on on
http://thread.gmane.org/gmane.comp.documentfoundation.libreoffice.scm/39102/focus=55516
and includes some additional cleanup of the sal signal code.
Change-Id: Iedc998e37e6495afec445eccb60fa1c2b1a7defd
Reviewed-on: https://gerrit.libreoffice.org/26497
Reviewed-by: Michael Meeks <michael.meeks@collabora.com>
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
Change-Id: I67eac95686678e6f5a2d60798535b2c65a9ba5d7
|
|
Change-Id: I4ead30fa1736a3a99fb3379e89367934c8df2809
Reviewed-on: https://gerrit.libreoffice.org/26444
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Jenkins <ci@libreoffice.org>
|
|
This reverts commit a82e532ce006c54b2740de74d1da5d11307da7c1. That hack should
no longer be necessary now that the root cause for tdf#38913 has probably been
found and fixed with b5876bfcb69a65c87d602bae687b3c0634c0a1e7 "passing a NULL
pointer to fileno is not allowed".
Conflicts:
sal/osl/w32/salinit.cxx
Change-Id: Id3a3d3163945acc9b5d5e74a8d357dfa2ce63f37
Reviewed-on: https://gerrit.libreoffice.org/26411
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Stephan Bergmann <sbergman@redhat.com>
|
|
avoid warning whether pid_t is long or int
Change-Id: I533175b53166cb233169902d277906f5ce979421
Signed-off-by: Richard PALO <richard@NetBSD.org>
Reviewed-on: https://gerrit.libreoffice.org/25593
Reviewed-by: Michael Stahl <mstahl@redhat.com>
Tested-by: Michael Stahl <mstahl@redhat.com>
|
|
Change-Id: If35da46c209746a86687e1c3f60f72f2f08aa681
|
|
Change-Id: I9070bae914d923171b6eaf8209de00922dc87be7
Reviewed-on: https://gerrit.libreoffice.org/25052
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Jochen Nitschke <j.nitschke+logerrit@ok.de>
|
|
The LibreOffice signal handler overwrites the breakpad signal handler.
Just store the old handler and call that handler before we do any work
in our own handler.
Our own handler implementation is just insane as we are even trying to
take the SolarMutex which can easily deadlock.
Change-Id: I2401bba18701115561d00c7fc11138ead2a48205
Reviewed-on: https://gerrit.libreoffice.org/25022
Tested-by: Jenkins <ci@libreoffice.org>
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
|
|
Change-Id: Icfc4b73d5664c205ad642c660af5227f403d20ff
Reviewed-on: https://gerrit.libreoffice.org/25021
Reviewed-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
Tested-by: Markus Mohrhard <markus.mohrhard@googlemail.com>
|
|
Fix "fallthrough annotation in unreachable code". Clang is clever
enough to know that code after assert(false) is unreachable in a
non-NDEBUG build.
Seriously, micro-optimisation of the case that should never be
reached?
Change-Id: I10d7ae60895437be52c27aa93539d17e0f091ac4
|
|
Change-Id: Id8262e735b9f95d8b476a4939c362648e2a4328a
|