Re: github mirror no longer updating

[Norbert Thiebaud]

On Mon, Jun 11, 2018 at 11:50 AM, Luke Benes  wrote:
> For network performance reasons, I have been using the mirror at:
> https://github.com/LibreOffice/core/commits/master
>
>
> But for some reason, it has stopped updating.

Indirect Microsoft takeover effect.
I cancelled my github account...
but gerrit is still using my account to push-replicate :-(

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: "Microsoft Excel Is Going Beyond Text and Numbers"

[Norbert Thiebaud]

On Mon, Apr 16, 2018 at 10:07 AM, Wol's lists  wrote:
>
>
> Concrete example which hit a couple of friends of mine ...
>
> Expat Americans, didn't declare their income to the American authorities.

even without declaration issues:
scenario:
you friend buy an asset in England. 20 years later he sells it.
for added-value purpose you need to calculate the profit/loss in the
taxable currency.
(note at the time the purchase is made, one may not know that 20 years
later he would need to convert that in some other currency)
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: "Microsoft Excel Is Going Beyond Text and Numbers"

[Norbert Thiebaud]

On Thu, Apr 5, 2018 at 12:02 AM, Tor Lillqvist  wrote:
>> (obligatory rant: _current_ exchange or stock market rates in
>> spreadsheet functions have close to zero practical value, and only
>> encourage people to do shitty accounting.
>
>
> Indeed. I am not an accountant or economist, but I would say that even
> historical exchange rates are of, well, historical interest only.

They are of tax-purpose interest.

> when you actually performed a currency transaction

you may not have performed a currency transaction at the time.
but 20 years later, for tax purpose, you need to 'convert' the value
of something at the time from one currency to another.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: PDFium, beware

[Norbert Thiebaud]

On Fri, Feb 9, 2018 at 6:33 PM, Stephan Bergmann  wrote:
> Recent master --enable-pdfium has at
> workdir/UnpackedTarball/pdfium/third_party/base/allocator/partition_allocator/page_allocator.h:31
>
>> // All Blink-supported systems have 4096 sized system pages and can handle
>> // permissions and commit / decommit at this granularity.
>> static const size_t kSystemPageSize = 4096;

and sysconf(_SC_PAGE_SIZE) is not working ?
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Let failed Jenkins builds set Verified: -1 in Gerrit?

[Norbert Thiebaud]

On Tue, Dec 5, 2017 at 1:16 AM, David Ostrovsky  wrote:
>
> RTFM: [1]? Quoting:
>
> "
> *Remove Reviewer*
>
> This category permits users to remove other users from the list of
> reviewers on a change.
> [...]
> Project owners and site administrators can always remove any reviewer
> (even without having the Remove Reviewer access right assigned).
>
> Users without this access right can only remove themselves from the
> reviewer list on a change.
> "
>
> Also note, that the ancient/legacy Gerrit version @LibreOffice
> has a major design bug/missing feature: [2] there is now way to
> remove a vote. Only the whole reviewer, with all votes can
> be wiped out.

I'm sorry but it is completely un-obvious that removing a 'reviewer'
from the list would wipe out the history of that reviewer.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Let failed Jenkins builds set Verified: -1 in Gerrit?

[Norbert Thiebaud]

On Fri, Dec 1, 2017 at 1:32 AM, David Ostrovsky  wrote:
> On Thu Nov 30 08:25:40 UTC 2017, Stephan Bergmann wrote:
>
>>Does "(with a help of maintainers)" imply that there are change
>>owners who could not override a VRFY-1 from Jenkins themselves?
>
> Yes. I believe that "registered users" group members will not be
> able to remove VRFY-1 vote.

I do not know of anyway, anyone can remove it using the ui, short of
impersonating jenkins' user.

the only available way to remove the -1 is to rebase the patch, since
hte -1 is bound to the particular patcheset version...
then hope that jenkins will not put another -1, or at least quickly
+2/+1 the rebased patch and submit... neither are desirable.

hence the choice of +1/0 rather than +1/-1 for jenkins.

For all practical purpose: 0 means jenkins has not yet try or was not
successful.
to submit one would have to manually add a verify +1. to do that you
need to go the the change view, and jenkins activity or lack thereof
is pretty apparent there.

I suspect the intent is to quicky avoid changeset that failed
jenkins... but then whether the change set has failed or has not yet
been built, what is the difference wrt to avoiding going in the
details of the change ?
from a review perspective no verification or failed verification means
the same thing: not ready
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Let failed Jenkins builds set Verified: -1 in Gerrit?

[Norbert Thiebaud]

On Thu, Nov 30, 2017 at 2:00 AM, Miklos Vajna  wrote:
>
> I think this was a decision from Norbert, probably because he was worried
> about false negatives. A workaround is to watch your inbox (which gets a
> mail, whatever is the outcome), rather than the above URL.
>
> (I would have no problem if Jenkins would put a Verified-1, though.)

Indeed False positive is a problem, especially since verify is a -1/+1 range
and
"By default, a change is submittable when it gets at least one highest
vote in each voting category and has no lowest vote (aka veto vote) in
any category."

so if jenkins did put a -1 that would be blocking until _jenkins_
reverse its vote.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

mac os build hung

[Norbert Thiebaud]

a slave bot got hung in libtest_chart2 somewhere in some opencl code:


Sampling process 11291 for 3 seconds with 1 millisecond of run time
between samples
Sampling completed, processing symbols...
Analysis of sampling cppunittester (pid 11291) every 1 millisecond
Process: cppunittester [11291]
Path:
/Users/tdf/lode/jenkins/workspace/lo_tb_master_mac_dbg/workdir/LinkTarget/Executable/cppunittester
Load Address:0x10c3f2000
Identifier:  cppunittester
Version: 0
Code Type:   X86-64
Parent Process:  ??? [1]

Date/Time:   2017-09-08 05:47:08.774 -0500
Launch Time: 2017-09-07 16:02:40.215 -0500
OS Version:  Mac OS X 10.12.6 (16G29)
Report Version:  7
Analysis Tool:   /usr/bin/sample


Call graph:
2013 Thread_216247   DispatchQueue_1: com.apple.main-thread  (serial)
+ 2013 start  (in libdyld.dylib) + 1  [0x7fffb1449235]
+   2013 main  (in cppunittester) + 39  [0x10c3f37f7]  cppunittester.cxx:373
+ 2013 sal_main()  (in cppunittester) + 3769  [0x10c3f46c9]
cppunittester.cxx:466
+   2013 (anonymous namespace)::ProtectedFixtureFunctor::run()
const  (in cppunittester) + 2890  [0x10c3f5b2a]  cppunittester.cxx:316
+ 2013 CppUnit::TestRunner::run(CppUnit::TestResult&,
std::__1::basic_string const&)  (in libcppunit-1.14.0.dylib) +
104  [0x10c5c0e88]  TestRunner.cpp:96
+   2013 CppUnit::TestResult::runTest(CppUnit::Test*)  (in
libcppunit-1.14.0.dylib) + 82  [0x10c5b40a2]  TestResult.cpp:149
+ 2013
CppUnit::TestRunner::WrappingSuite::run(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 78  [0x10c5c0bae]  TestRunner.cpp:47
+   2013
CppUnit::TestComposite::run(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 90  [0x10c598c4a]  TestComposite.cpp:23
+ 2013
CppUnit::TestComposite::doRunChildTests(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 125  [0x10c598dcd]  TestComposite.cpp:64
+   2013
CppUnit::TestComposite::run(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 90  [0x10c598c4a]  TestComposite.cpp:23
+ 2013
CppUnit::TestComposite::doRunChildTests(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 125  [0x10c598dcd]  TestComposite.cpp:64
+   2013
CppUnit::TestCase::run(CppUnit::TestResult*)  (in
libcppunit-1.14.0.dylib) + 739  [0x10c598013]  TestCase.cpp:91
+ 2013
CppUnit::TestResult::protect(CppUnit::Functor const&, CppUnit::Test*,
std::__1::basic_string const&)  (in libcppunit-1.14.0.dylib) + 99
 [0x10c5b4543]  TestResult.cpp:182
+   2013
CppUnit::ProtectorChain::protect(CppUnit::Functor const&,
CppUnit::ProtectorContext const&)  (in libcppunit-1.14.0.dylib) + 1346
 [0x10c5829b2]  ProtectorChain.cpp:86
+ 2013
CppUnit::ProtectorChain::ProtectFunctor::operator()() const  (in
libcppunit-1.14.0.dylib) + 40  [0x10c58e718]  ProtectorChain.cpp:20
+   2013
CppUnit::DefaultProtector::protect(CppUnit::Functor const&,
CppUnit::ProtectorContext const&)  (in libcppunit-1.14.0.dylib) + 79
[0x10c5645ef]  DefaultProtector.cpp:15
+ 2013
CppUnit::ProtectorChain::ProtectFunctor::operator()() const  (in
libcppunit-1.14.0.dylib) + 40  [0x10c58e718]  ProtectorChain.cpp:20
+   2013 (anonymous
namespace)::Prot::protect(CppUnit::Functor const&,
CppUnit::ProtectorContext const&)  (in unoexceptionprotector.dylib) +
79  [0x10c79bf7f]  unoexceptionprotector.cxx:63
+ 2013
CppUnit::ProtectorChain::ProtectFunctor::operator()() const  (in
libcppunit-1.14.0.dylib) + 40  [0x10c58e718]  ProtectorChain.cpp:20
+   2013 (anonymous
namespace)::Prot::protect(CppUnit::Functor const&,
CppUnit::ProtectorContext const&)  (in unobootstrapprotector.dylib) +
33  [0x10ceda9d1]  unobootstrapprotector.cxx:89
+ 2013
CppUnit::ProtectorChain::ProtectFunctor::operator()() const  (in
libcppunit-1.14.0.dylib) + 40  [0x10c58e718]  ProtectorChain.cpp:20
+   2013 (anonymous
namespace)::Protector::protect(CppUnit::Functor const&,
CppUnit::ProtectorContext const&)  (in libvclbootstrapprotector.dylib)
+ 33  [0x10d83fd31]  vclbootstrapprotector.cxx:39
+ 2013
CppUnit::TestCaseMethodFunctor::operator()() const  (in
libcppunit-1.14.0.dylib) + 110  [0x10c59871e]  TestCase.cpp:32
+   2013
CppUnit::TestCaller::runTest()  (in
libtest_chart2_import.dylib) + 25  [0x11717d409]  TestCaller.h:175
+ 2013
std::__1::function::operator()() const  (in

Re: Build failure with current master on MacOS

[Norbert Thiebaud]

On Mon, Jul 24, 2017 at 3:33 PM, Alexander Thurgood
 wrote:
> Le 24/07/2017 à 15:16, Michael Stahl a écrit :
>
>> (here "fixed" means you'll get an error message from configure instead
>> that tells you to install gettext)
>>
>
> At least the message is explicit enough ;-)
>
> I don't build via LODE, just the standard OSX Terminal.app.

The 'standard'  , without any build dep does not have the right make,
the right doxygen, the rigth autoconf/autogen, does not have cmake or
ant or junit...

lode is merely a few line of bash that mostly take care of that.
and lo's configure.ac is lode aware to automatically pick the 'right'
stuff in that case, so that you do not _have_ to pollute your path and
other things.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: Branch 'libreoffice-5-4' - configure.ac

[Norbert Thiebaud]

 configure.ac |   46 +-
 1 file changed, 13 insertions(+), 33 deletions(-)

New commits:
commit d0548cc49d20910e222a4fe84ebc48b8786a127d
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Fri May 19 07:07:27 2017 -0500

bumb MacOS baseline to 10.9

we desire libxml 2.8+ and as a side effect that also provide
thread-safe static in c++

(cherry picked from commit 7dcee9790c18a594236dc32a534c20f1e7447ee6)

Change-Id: I2a0f06a81c6a5734a8b3a859fb291e5ae80d12e9
Reviewed-on: https://gerrit.libreoffice.org/38028
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Miklos Vajna <vmik...@collabora.co.uk>

diff --git a/configure.ac b/configure.ac
index 35497b23cb58..b38623879abd 100644
--- a/configure.ac
+++ b/configure.ac
@@ -2329,7 +2329,7 @@ AC_ARG_WITH(macosx-sdk,
   not terribly useful. It works fine to build with a 
new SDK and run the result
   on an older OS.
 
-  e. g.: --with-macosx-sdk=10.8
+  e. g.: --with-macosx-sdk=10.9
 
   there are 3 options to control the MacOSX build:
   --with-macosx-sdk (referred as 'sdk' below)
@@ -2344,10 +2344,10 @@ AC_ARG_WITH(macosx-sdk,
command line  || config result
   ==
   min  | max  | sdk  || min  | max  | sdk  |
-  ?| ?| ?|| 10.8 | 10.s | 10.s |
-  ?| ?| 10.x || 10.8 | 10.x | 10.x |
-  ?| 10.x | ?|| 10.8 | 10.s | 10.s |
-  ?| 10.x | 10.y || 10.8 | 10.x | 10.y |
+  ?| ?| ?|| 10.9 | 10.s | 10.s |
+  ?| ?| 10.x || 10.9 | 10.x | 10.x |
+  ?| 10.x | ?|| 10.9 | 10.s | 10.s |
+  ?| 10.x | 10.y || 10.9 | 10.x | 10.y |
   10.x | ?| ?|| 10.x | 10.s | 10.s |
   10.x | ?| 10.y || 10.x | 10.y | 10.y |
   10.x | 10.y | ?|| 10.x | 10.y | 10.y |
@@ -2365,7 +2365,7 @@ AC_ARG_WITH(macosx-version-min-required,
 AS_HELP_STRING([--with-macosx-version-min-required=],
 [set the minimum OS version needed to run the built LibreOffice])
 [
-  e. g.: --with-macos-version-min-required=10.8
+  e. g.: --with-macos-version-min-required=10.9
   see --with-macosx-sdk for more info
 ],
 ,)
@@ -2374,7 +2374,7 @@ AC_ARG_WITH(macosx-version-max-allowed,
 AS_HELP_STRING([--with-macosx-version-max-allowed=],
 [set the maximum allowed OS version the LibreOffice compilation can 
use APIs from])
 [
-  e. g.: --with-macos-version-max-allowed=10.8
+  e. g.: --with-macos-version-max-allowed=10.9
   see --with-macosx-sdk for more info
 ],
 ,)
@@ -2740,7 +2740,7 @@ if test $_os = Darwin; then
 
 AC_MSG_CHECKING([what Mac OS X SDK to use])
 
-for _macosx_sdk in $with_macosx_sdk 10.8 10.9 10.10 10.11 10.12; do
+for _macosx_sdk in $with_macosx_sdk 10.9 10.10 10.11 10.12; do
 MACOSX_SDK_PATH=`xcrun --sdk macosx${_macosx_sdk} --show-sdk-path 2> 
/dev/null`
 if test -d "$MACOSX_SDK_PATH"; then
 with_macosx_sdk="${_macosx_sdk}"
@@ -2759,9 +2759,6 @@ if test $_os = Darwin; then
 AC_MSG_RESULT([SDK $with_macosx_sdk at $MACOSX_SDK_PATH])
 
 case $with_macosx_sdk in
-10.8)
-MACOSX_SDK_VERSION=1080
-;;
 10.9)
 MACOSX_SDK_VERSION=1090
 ;;
@@ -2775,12 +2772,12 @@ if test $_os = Darwin; then
 MACOSX_SDK_VERSION=101200
 ;;
 *)
-AC_MSG_ERROR([with-macosx-sdk $with_macosx_sdk is not a supported 
value, supported values are 10.8--12])
+AC_MSG_ERROR([with-macosx-sdk $with_macosx_sdk is not a supported 
value, supported values are 10.9--12])
 ;;
 esac
 
 if test "$with_macosx_version_min_required" = "" ; then
-with_macosx_version_min_required="10.8";
+with_macosx_version_min_required="10.9";
 fi
 
 if test "$with_macosx_version_max_allowed" = "" ; then
@@ -2795,9 +2792,6 @@ if test $_os = Darwin; then
 MACOSX_DEPLOYMENT_TARGET="$with_macosx_version_min_required"
 
 case "$with_macosx_version_min_required" in
-10.8)
-MAC_OS_X_VERSION_MIN_REQUIRED="1080"
-;;
 10.9)
 MAC_OS_X_VERSION_MIN_REQUIRED="1090"
 ;;
@@ -2811,7 +2805,7 @@ if test $_os = Darwin; then
 MAC_OS_X_VERSION_MIN_REQUIRED="101200"
  

[Libreoffice-commits] core.git: solenv/bin

[Norbert Thiebaud]

 solenv/bin/mkdocs_portal.sh |1 +
 1 file changed, 1 insertion(+)

New commits:
commit b8a1797e1511bdee779112f09b1e64098591b9fe
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Tue May 23 16:31:27 2017 -0500

mkdocs_portal: reset local variable in loop

Change-Id: I0c66717c0fac6d4d1bb6a33b8a196066f0f880d1

diff --git a/solenv/bin/mkdocs_portal.sh b/solenv/bin/mkdocs_portal.sh
index 468e94d68f73..352d2a7bb570 100755
--- a/solenv/bin/mkdocs_portal.sh
+++ b/solenv/bin/mkdocs_portal.sh
@@ -95,6 +95,7 @@ header "LibreOffice Modules" " " "$BASE_OUTPUT/index.html"
 
 for module_name in *; do
 if [ -d $module_name ]; then
+cur_file=
 if [ -f $module_name/readme.txt ] ; then
 cur_file="$module_name/readme.txt"
 elif [ -f $module_name/README ] ; then
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

[Libreoffice-commits] core.git: configure.ac

[Norbert Thiebaud]

 configure.ac |   46 +-
 1 file changed, 13 insertions(+), 33 deletions(-)

New commits:
commit 7dcee9790c18a594236dc32a534c20f1e7447ee6
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Fri May 19 07:07:27 2017 -0500

bumb MacOS baseline to 10.9

we desire libxml 2.8+ and as a side effect that also provide
thread-safe static in c++

Change-Id: I2a0f06a81c6a5734a8b3a859fb291e5ae80d12e9
Reviewed-on: https://gerrit.libreoffice.org/37818
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Miklos Vajna <vmik...@collabora.co.uk>

diff --git a/configure.ac b/configure.ac
index 8eea72fa4938..585c47efdfb7 100644
--- a/configure.ac
+++ b/configure.ac
@@ -2333,7 +2333,7 @@ AC_ARG_WITH(macosx-sdk,
   not terribly useful. It works fine to build with a 
new SDK and run the result
   on an older OS.
 
-  e. g.: --with-macosx-sdk=10.8
+  e. g.: --with-macosx-sdk=10.9
 
   there are 3 options to control the MacOSX build:
   --with-macosx-sdk (referred as 'sdk' below)
@@ -2348,10 +2348,10 @@ AC_ARG_WITH(macosx-sdk,
command line  || config result
   ==
   min  | max  | sdk  || min  | max  | sdk  |
-  ?| ?| ?|| 10.8 | 10.s | 10.s |
-  ?| ?| 10.x || 10.8 | 10.x | 10.x |
-  ?| 10.x | ?|| 10.8 | 10.s | 10.s |
-  ?| 10.x | 10.y || 10.8 | 10.x | 10.y |
+  ?| ?| ?|| 10.9 | 10.s | 10.s |
+  ?| ?| 10.x || 10.9 | 10.x | 10.x |
+  ?| 10.x | ?|| 10.9 | 10.s | 10.s |
+  ?| 10.x | 10.y || 10.9 | 10.x | 10.y |
   10.x | ?| ?|| 10.x | 10.s | 10.s |
   10.x | ?| 10.y || 10.x | 10.y | 10.y |
   10.x | 10.y | ?|| 10.x | 10.y | 10.y |
@@ -2369,7 +2369,7 @@ AC_ARG_WITH(macosx-version-min-required,
 AS_HELP_STRING([--with-macosx-version-min-required=],
 [set the minimum OS version needed to run the built LibreOffice])
 [
-  e. g.: --with-macos-version-min-required=10.8
+  e. g.: --with-macos-version-min-required=10.9
   see --with-macosx-sdk for more info
 ],
 ,)
@@ -2378,7 +2378,7 @@ AC_ARG_WITH(macosx-version-max-allowed,
 AS_HELP_STRING([--with-macosx-version-max-allowed=],
 [set the maximum allowed OS version the LibreOffice compilation can 
use APIs from])
 [
-  e. g.: --with-macos-version-max-allowed=10.8
+  e. g.: --with-macos-version-max-allowed=10.9
   see --with-macosx-sdk for more info
 ],
 ,)
@@ -2744,7 +2744,7 @@ if test $_os = Darwin; then
 
 AC_MSG_CHECKING([what Mac OS X SDK to use])
 
-for _macosx_sdk in $with_macosx_sdk 10.8 10.9 10.10 10.11 10.12; do
+for _macosx_sdk in $with_macosx_sdk 10.9 10.10 10.11 10.12; do
 MACOSX_SDK_PATH=`xcrun --sdk macosx${_macosx_sdk} --show-sdk-path 2> 
/dev/null`
 if test -d "$MACOSX_SDK_PATH"; then
 with_macosx_sdk="${_macosx_sdk}"
@@ -2763,9 +2763,6 @@ if test $_os = Darwin; then
 AC_MSG_RESULT([SDK $with_macosx_sdk at $MACOSX_SDK_PATH])
 
 case $with_macosx_sdk in
-10.8)
-MACOSX_SDK_VERSION=1080
-;;
 10.9)
 MACOSX_SDK_VERSION=1090
 ;;
@@ -2779,12 +2776,12 @@ if test $_os = Darwin; then
 MACOSX_SDK_VERSION=101200
 ;;
 *)
-AC_MSG_ERROR([with-macosx-sdk $with_macosx_sdk is not a supported 
value, supported values are 10.8--12])
+AC_MSG_ERROR([with-macosx-sdk $with_macosx_sdk is not a supported 
value, supported values are 10.9--12])
 ;;
 esac
 
 if test "$with_macosx_version_min_required" = "" ; then
-with_macosx_version_min_required="10.8";
+with_macosx_version_min_required="10.9";
 fi
 
 if test "$with_macosx_version_max_allowed" = "" ; then
@@ -2799,9 +2796,6 @@ if test $_os = Darwin; then
 MACOSX_DEPLOYMENT_TARGET="$with_macosx_version_min_required"
 
 case "$with_macosx_version_min_required" in
-10.8)
-MAC_OS_X_VERSION_MIN_REQUIRED="1080"
-;;
 10.9)
 MAC_OS_X_VERSION_MIN_REQUIRED="1090"
 ;;
@@ -2815,7 +2809,7 @@ if test $_os = Darwin; then
 MAC_OS_X_VERSION_MIN_REQUIRED="101200"
 ;;
 *)
-AC_MSG_ERROR([with-macosx-version-min-require

Re: Some baseline thoughts

[Norbert Thiebaud]

On Fri, May 19, 2017 at 5:58 AM, Tor Lillqvist  wrote:
>
>> that box use 10.8sdk which come with libxml:
>> #define LIBXML_DOTTED_VERSION "2.7.8"
>>
>
> In a current SDK it is 2.9.4. I don't know how it has been updated in SDK
> versions. But seriously, couldn't we bump the baseline to 10.9 by now? That
> would also help the thread-safe statics thing, if I understand correctly.
> Macs that can't be upgraded to 10.9 or beyond are about ten years old now, I
> think.

I don't mind bumping to 10.9. the box that do the realease is already
at 10.9 anyway

according to google there are about 2 to 3% of mac users still on 10.8.
but hey.. if they are content with sticking to 10.8 they can be
content with sticking with lo 5.3

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Some baseline thoughts

[Norbert Thiebaud]

On Wed, May 17, 2017 at 2:50 AM, Miklos Vajna  wrote:
> Hi,
>
> On Thu, Mar 16, 2017 at 12:20:54PM +0100, Stephan Bergmann 
>  wrote:
>> *  macOS 10.8 
>> ()
>>
>> Bumping this to macOS 10.9 would get rid of the last known case of
>> !HAVE_THREADSAFE_STATICS (cf. 
>> 
>> "Found out that Mac OS X 10.9 has working threadsafe statics").  What is
>> unknown however is the status on Android and iOS.  Having
>> HAVE_THREADSAFE_STATICS everywhere would allow to clean up code, see e.g.
>>  "replace double-checked-locking
>> with C++11 static initializer".
>
> Do we have any tinderbox building against the 10.8 SDK? I just realized
> that the xmlsec we bundle on master now requires libxml 2.8, and as far
> as I see Jenkins builds against the 10.9 SDK. So there is a risk that
> the 10.8 SDK provides an older libxml and this went unnoticed.

it got noticed today. the beta1 build faird for that very reason

the code use input_id of parserCtxt which was introduced in 2.8

that box use 10.8sdk which come with libxml:
#define LIBXML_DOTTED_VERSION "2.7.8"

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: -Og in dbgutil builds

[Norbert Thiebaud]

On Thu, Apr 6, 2017 at 10:39 AM, Michael Stahl  wrote:

>
> but i'm mostly interested in Norbert's opinion, since the biggest
> benefit of -Og is faster tinderbox/jenkins builds.

I am neutral. Considering that we are doing a clang-plugin build in
the mix, I doubt -Og on the gcc build is going to impact that much
either way the overall bottom line.
If that happen to be a real impact, I can always tweak the jenkins
build using the dedicated distro-config / lode mechanisms.

iow: make the default what make sens for dev. if that has a negative
impact on jenkins, we will sort it out there.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: Branch 'libreoffice-5-3' - solenv/bin

[Norbert Thiebaud]

 solenv/bin/macosx-codesign-app-bundle |   47 ++
 1 file changed, 37 insertions(+), 10 deletions(-)

New commits:
commit cb76a9c17dcf38817ca45c4675f52dcab01a7749
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sat Mar 18 09:43:47 2017 -0500

codesigning script for macosx compained about double signing

Release build of 5.3.2.1 failed in codesign
apparently LibreOfficePython.framework was being signed more than
once, which cause codesign to fail and due to a recent
patch to harden the codesign wrapper, the build itself to fail

This does not address why some part are signed multiple time
but merely tell codesign to ignore the issue and just sign

This also fix a bash un-initialize variable warning and
capture output of codesign in case of error to be able to diagnose
things.

Change-Id: Ibd6752702feb2bdf5163ac30ed7a3fd9c86f961c
Reviewed-on: https://gerrit.libreoffice.org/35407
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
(cherry picked from commit a332bb9a6cc19f0c212892b3f304583338b0a094)
Reviewed-on: https://gerrit.libreoffice.org/35542
Reviewed-by: Miklos Vajna <vmik...@collabora.co.uk>

diff --git a/solenv/bin/macosx-codesign-app-bundle 
b/solenv/bin/macosx-codesign-app-bundle
index 1149e70ea6d1..a330205e9915 100755
--- a/solenv/bin/macosx-codesign-app-bundle
+++ b/solenv/bin/macosx-codesign-app-bundle
@@ -26,7 +26,7 @@ for V in \
 done
 
 APP_BUNDLE="$1"
-
+entitlements=
 if test -n "$ENABLE_MACOSX_SANDBOX"; then
 # In a sandboxed build executables need the entitlements
 entitlements="--entitlements $BUILDDIR/lo.xcent"
@@ -47,7 +47,11 @@ fi
 find -d "$APP_BUNDLE" \( -name '*.jnilib' \) ! -type l |
 while read file; do
 id=`echo ${file#${APP_BUNDLE}/Contents/} | sed -e 's,/,.,g'`
-codesign --verbose --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$file" || exit 1
+codesign --verbose --force --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id 
--sign "$MACOSX_CODESIGNING_IDENTITY" "$file" > "/tmp/codesign_$(basename 
"$file").log" 2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_$(basename "$file").log"
 done
 
 # Sign dylibs
@@ -62,7 +66,11 @@ find "$APP_BUNDLE" \( -name '*.dylib' -or -name '*.dylib.*' 
-or -name '*.so' \
 $other_files \) ! -type l |
 while read file; do
 id=`echo ${file#${APP_BUNDLE}/Contents/} | sed -e 's,/,.,g'`
-codesign --verbose --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$file" || exit 1
+codesign --verbose --force --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id 
--sign "$MACOSX_CODESIGNING_IDENTITY" "$file" > "/tmp/codesign_$(basename 
"$file").log" 2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_$(basename "$file").log"
 done
 
 # Sign included bundles. First .app ones (i.e. the Python.app inside
@@ -74,7 +82,11 @@ while read app; do
 fn=${fn%.*}
 # Assume the app has a XML (and not binary) Info.plist
 id=`grep -A 1 'CFBundleIdentifier' $app/Contents/Info.plist | 
tail -1 | sed -e 's,.*,,' -e 's,.*,,'`
-codesign --verbose --identifier=$id --sign "$MACOSX_CODESIGNING_IDENTITY" 
$entitlements "$app" || exit 1
+codesign --verbose --force --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" $entitlements "$app" > "/tmp/codesign_${fn}.log" 
2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_${fn}.log"
 done
 
 # Then .framework ones. Again, be generic just for kicks.
@@ -87,8 +99,12 @@ while read framework; do
 if test ! -L "$version" -a -d "$version"; then
# Assume the framework has a XML (and not binary) Info.plist
id=`grep -A 1 'CFBundleIdentifier' 
$version/Resources/Info.plist | tail -1 | sed -e 's,.*,,' -e 
's,.*,,'`
-codesign --verbose --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$version" || exit 1
-fi
+codesign --verbose --force --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$version" > "/tmp/codesign_${fn}.log" 2>&1
+   if [ "$?" != "0" ] ; then
+   exit 1
+   fi
+   rm "/tmp/codesign_${fn}.log"
+   fi
 done
 done
 
@@ -96,7 +112,11 @@ done
 
 find "$APP_BUNDLE" -name '*.mdimporter' -type d |
 while read bundle; do
-codesign --verbose --prefix=$MACOSX_BUNDL

[Libreoffice-commits] core.git: solenv/bin

[Norbert Thiebaud]

 solenv/bin/macosx-codesign-app-bundle |   47 ++
 1 file changed, 37 insertions(+), 10 deletions(-)

New commits:
commit a332bb9a6cc19f0c212892b3f304583338b0a094
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sat Mar 18 09:43:47 2017 -0500

codesigning script for macosx compained about double signing

Release build of 5.3.2.1 failed in codesign
apparently LibreOfficePython.framework was being signed more than
once, which cause codesign to fail and due to a recent
patch to harden the codesign wrapper, the build itself to fail

This does not address why some part are signed multiple time
but merely tell codesign to ignore the issue and just sign

This also fix a bash un-initialize variable warning and
capture output of codesign in case of error to be able to diagnose
things.

Change-Id: Ibd6752702feb2bdf5163ac30ed7a3fd9c86f961c
Reviewed-on: https://gerrit.libreoffice.org/35407
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/solenv/bin/macosx-codesign-app-bundle 
b/solenv/bin/macosx-codesign-app-bundle
index 39d87246a92f..f5ccff1475f1 100755
--- a/solenv/bin/macosx-codesign-app-bundle
+++ b/solenv/bin/macosx-codesign-app-bundle
@@ -26,7 +26,7 @@ for V in \
 done
 
 APP_BUNDLE="$1"
-
+entitlements=
 if test -n "$ENABLE_MACOSX_SANDBOX"; then
 # In a sandboxed build executables need the entitlements
 entitlements="--entitlements $BUILDDIR/lo.xcent"
@@ -48,7 +48,11 @@ fi
 find -d "$APP_BUNDLE" \( -name '*.jnilib' \) ! -type l |
 while read file; do
 id=`echo ${file#${APP_BUNDLE}/Contents/} | sed -e 's,/,.,g'`
-codesign --verbose --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$file" || exit 1
+codesign --verbose --force --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id 
--sign "$MACOSX_CODESIGNING_IDENTITY" "$file" > "/tmp/codesign_$(basename 
"$file").log" 2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_$(basename "$file").log"
 done
 
 # Sign dylibs
@@ -63,7 +67,11 @@ find "$APP_BUNDLE" \( -name '*.dylib' -or -name '*.dylib.*' 
-or -name '*.so' \
 $other_files \) ! -type l |
 while read file; do
 id=`echo ${file#${APP_BUNDLE}/Contents/} | sed -e 's,/,.,g'`
-codesign --verbose --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$file" || exit 1
+codesign --verbose --force --identifier=$MACOSX_BUNDLE_IDENTIFIER.$id 
--sign "$MACOSX_CODESIGNING_IDENTITY" "$file" > "/tmp/codesign_$(basename 
"$file").log" 2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_$(basename "$file").log"
 done
 
 # Sign included bundles. First .app ones (i.e. the Python.app inside
@@ -75,7 +83,11 @@ while read app; do
 fn=${fn%.*}
 # Assume the app has a XML (and not binary) Info.plist
 id=`grep -A 1 'CFBundleIdentifier' $app/Contents/Info.plist | 
tail -1 | sed -e 's,.*,,' -e 's,.*,,'`
-codesign --verbose --identifier=$id --sign "$MACOSX_CODESIGNING_IDENTITY" 
$entitlements "$app" || exit 1
+codesign --verbose --force --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" $entitlements "$app" > "/tmp/codesign_${fn}.log" 
2>&1
+if [ "$?" != "0" ] ; then
+   exit 1
+fi
+rm "/tmp/codesign_${fn}.log"
 done
 
 # Then .framework ones. Again, be generic just for kicks.
@@ -88,8 +100,12 @@ while read framework; do
 if test ! -L "$version" -a -d "$version"; then
# Assume the framework has a XML (and not binary) Info.plist
id=`grep -A 1 'CFBundleIdentifier' 
$version/Resources/Info.plist | tail -1 | sed -e 's,.*,,' -e 
's,.*,,'`
-codesign --verbose --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$version" || exit 1
-fi
+codesign --verbose --force --identifier=$id --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$version" > "/tmp/codesign_${fn}.log" 2>&1
+   if [ "$?" != "0" ] ; then
+   exit 1
+   fi
+   rm "/tmp/codesign_${fn}.log"
+   fi
 done
 done
 
@@ -97,7 +113,11 @@ done
 
 find "$APP_BUNDLE" -name '*.mdimporter' -type d |
 while read bundle; do
-codesign --verbose --prefix=$MACOSX_BUNDLE_IDENTIFIER. --sign 
"$MACOSX_CODESIGNING_IDENTITY" "$bundle" || exit 1
+codesign --verbose --force --prefix=$MACOSX_BUNDLE_IDENTIFIER. --sign 
&q

Re: sha1sum problem

[Norbert Thiebaud]

On Fri, Feb 24, 2017 at 9:45 AM, Bryan Quigley
 wrote:
> I'm going to look at adding the SHA256SUM for all the files in
> download.lst.  It looks like we can add them without breaking
> anything.  Then we can test out how we want to switch or if we need to
> support both for a bit (platform dependent?).

shasum is already present and working on the supported platform
(I checked earlier on mac and windows)
so yeah that should be doable pretty much transparently.

Supporting a mix of method would be more painful and really not that
useful I think.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: sha1sum problem

[Norbert Thiebaud]

On Fri, Feb 24, 2017 at 9:25 AM, Stephan Bergmann <sberg...@redhat.com> wrote:
> On 02/24/2017 03:47 PM, Norbert Thiebaud wrote:
>>
>> The situation with checksum of 'external' files is much worse that you
>> thought.
>> They are actually checked with md5.
>> That being said they are not truly external, since they are hosted on
>> the project infrastructure
>
>
> ...but downloaded via plain HTTP

guilhem is fixing that on all our web services.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: sha1sum problem

[Norbert Thiebaud]

On Fri, Feb 24, 2017 at 8:03 AM, Necdet Yücel <necdetyu...@gmail.com> wrote:
> 2017-02-24 16:42 GMT+03:00 Norbert Thiebaud <nthieb...@gmail.com>:
>>
>> On Fri, Feb 24, 2017 at 5:13 AM, Necdet Yücel <necdetyu...@gmail.com>
>> wrote:
>> > Hi,
>> >
>> > in configure script file integrity is checked by using sha1sums.
>>
>> You are sure about that ? or you just did a quick git grep sha1sum and
>> went from there ?
>
>
> i saw SHA1SUM="openssl sha1" in configure script and wrote here. If it's ok
> for LO, then it's ok for me too. Sorry for inconvenience

This instance of sha1sum usage is in a test intended to detect a bug
in gmake. the input being sha-ed
are generated during the test.. this is not suceptible attacks, and
even it it was the consequences would be to
not detect a buggy gmake and have the build fail possibly later.

The situation with checksum of 'external' files is much worse that you thought.
They are actually checked with md5.
That being said they are not truly external, since they are hosted on
the project infrastructure
and the original motivation was not so much malicious injection
detection but faulty transfer.
using sha1 there would actually be an 'improvement' :-)

I guess we could convert that to shasum -a 256

Note though that the binaries published by tdf for download come with
a variety of hashes to verify against
http://download.documentfoundation.org/libreoffice/stable/5.3.0/mac/x86_64/LibreOffice_5.3.0_MacOS_x86-64.dmg.mirrorlist
including a pgp signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: sha1sum problem

[Norbert Thiebaud]

On Fri, Feb 24, 2017 at 5:13 AM, Necdet Yücel  wrote:
> Hi,
>
> in configure script file integrity is checked by using sha1sums.

You are sure about that ? or you just did a quick git grep sha1sum and
went from there ?


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: sha1sum problem

[Norbert Thiebaud]

On Thu, Feb 23, 2017 at 10:54 AM, Necdet Yücel  wrote:
> Hi,
>
> Since SHA1 has been broken in practice [1] maybe we should consider to move
> to safer alteratives, such as SHA-256.

What are you talking about ? (yes I read the news this morning too,
but what concretely are you talking about in the context of this
project ? )

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Build Time Query

[Norbert Thiebaud]

On Thu, Feb 23, 2017 at 3:53 PM, Jason Marshall
 wrote:
> Dear Sir/ Madam
>
> I am attempting to build LibreOffice on the following platform:
>
> OS: Lubuntu
> Processor: Intel Atom Z530 1.6 GHz
> RAM: 1Gb
> Disk: 60Gb SSD
>
> Would greater than ten hours be expected,
> as I am aware that the hardware specification above is quite low.

The spec of your processor are at least 70 times slower than the cpu
we use to build the product in our ci system, and 25 to 30 times
slower than an average modern laptop.
1GB RAM is also not a lot, and you will prolly swap out during some of
the link phases.

a full cold build on our ci sytem takes ~30 minutes

>  In
> addition, if the failed build is simply re-started by running the ‘make'

yes

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Infra call on Tue, Feb 21 at 17:30 UTC

[Norbert Thiebaud]

On Mon, Feb 20, 2017 at 11:27 AM, Guilhem Moulin
 wrote:
> On Fri, 17 Feb 2017 at 10:51:19 +0100, Guilhem Moulin wrote:
>> The next infra call will take place at `date -d 'Tue Feb 21 17:30:00 UTC 
>> 2017'`
>> (18:30:00 Berlin time).
>
> Gentle reminder: that's in 24h!  (And 17:30:00 *UTC* does not
> necessarily mean 17:30:00 local time :-P  Enter the above command in
> your favorite terminal to make the conversion.)

btw

$>:~ n_th$ `date -d 'Tue Feb 21 17:30:00 UTC 2017'`
usage: date [-jnu] [-d dst] [-r seconds] [-t west] [-v[+|-]val[ymwdHMS]] ...
[-f fmt date | [[[mm]dd]HH]MM[[cc]yy][.ss]] [+format]
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: compiler plugin

[Norbert Thiebaud]

On Fri, Jan 20, 2017 at 3:16 AM, Stephan Bergmann <sberg...@redhat.com> wrote:
> On 07/12/2016 09:32 AM, Norbert Thiebaud wrote:
>>
>> I've enabled an additional build for gerrit doing clang + plugins on linux
>> we will see how that perform in average.
>> preliminary observation is that there is way to much churn in the
>> plugins for this to be viable at this time
>
>
> What appears to be missing is a "normal" (not Gerrit-triggered) bot running
> that setup on master, so that a broken master wrt that setup can be spotted
> (and acted upon) more quickly.
>
> Today, a broken master is only spotted when somebody tracks down a broken
> Gerrit build to be due to a broken master

so, if I understand it, a direct push to master that broke it, surprise!

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Please add an option to set bug reports to "Private"

[Norbert Thiebaud]

On Thu, Dec 22, 2016 at 4:12 AM, Michael Meeks
 wrote:
>
>> You cannot expect volunteers to go through hoops to give you free
>> support because you are not willing to do your part.
>
> I really don't believe that is a fair characterization of Christoph's
> suggestion =)

s/you/one/

the 'you' there was used in a generic sens not a particular.. I should
have used the undefined pronoun 'one'

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Please add an option to set bug reports to "Private"

[Norbert Thiebaud]

On Wed, Dec 21, 2016 at 12:08 AM, "Christoph Schäfer"
 wrote:
> Dear LibreOffice developers,
>
>
> I'd be grateful if you could add an option to your bugtracker to hide bug 
> reports and/or sample files from public view, i.e., add a "Private" option, 
> so that only developers and admins can see them.

The whole world is a potential 'developper' in an open source project.
That is the whole point.

If a customer want secrecy and NDAs and the like.. they need to use a
consulting company to solve their bug. This is the only way they can
maintain any sort of 'secrecy' of their test documents.

The only viable alternative, as mentioned  elsewhere in this thread,
is to sanitize the documents to remove 'sensitive' stuff.
You cannot expect volunteers to go through hoops to give you free
support because you are not willing to do your part.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

idlc cpu loop

[Norbert Thiebaud]

Stephan,

as requested:

Analysis of sampling idlc (pid 3535) every 1 millisecond
Process: idlc [3535]
Path:
/Users/tdf/lode/jenkins/workspace/lo_gerrit/Config/macosx_clang_dbgutil/instdir/LibreOffice5.4_SDK/bin/idlc
Load Address:0x108408000
Identifier:  idlc
Version: 0
Code Type:   X86-64
Parent Process:  idlc [3503]

Date/Time:   2016-12-19 19:05:16.839 -0600
OS Version:  Mac OS X 10.9.5 (13F1077)
Report Version:  7

Call graph:
2391 Thread_117354590: Main Thread   DispatchQueue_
  2391 thread_start  (in libsystem_pthread.dylib) + 13  [0x7fff93ae5fc9]
2391 _pthread_start  (in libsystem_pthread.dylib) + 137
[0x7fff93ae172a]
  2391 _pthread_body  (in libsystem_pthread.dylib) + 138
[0x7fff93ae1899]
2391 osl_thread_start_Impl(void*)  (in libuno_sal.dylib.3)
+ 335  [0x1085c352f]  thread.cxx:240
  2391 ChildStatusProc(void*)  (in libuno_sal.dylib.3) +
8863  [0x1085ab35f]  process.cxx:208
2391 std::__1::basic_ostream&
std::__1::operator<<(std::__1::basic_ostream&, char
const*)  (in libuno_sal.dylib.3) + 68  [0x10852e674]  ostream:882
  2391 std::__1::basic_ostream&
std::__1::__put_character_sequence(std::__1::basic_ostream&, char
const*, unsigned long)  (in libuno_sal.dylib.3) + 743  [0x1085321b7]
ostream:743
2391 std::__1::ostreambuf_iterator std::__1::__pad_and_output
>(std::__1::ostreambuf_iterator,
char const*, char const*, char const*, std::__1::ios_base&, char)  (in
libuno_sal.dylib.3) + 927  [0x1085326bf]  streambuf:350
  2391 std::__1::basic_stringbuf::overflow(int)  (in libuno_sal.dylib.3) + 371  [0x1085343d3]
sstream:528
2391 std::__1::basic_string::push_back(char)  (in libc++.1.dylib) + 86  [0x7fff95591b62]
  2391 std::__1::basic_string::__grow_by(unsigned long, unsigned long, unsigned long, unsigned
long, unsigned long, unsigned long)  (in libc++.1.dylib) + 144
[0x7fff95591e7a]
2391 operator new(unsigned long)  (in
libc++.1.dylib) + 30  [0x7fff9558e28e]
  2391 malloc  (in libsystem_malloc.dylib)
+ 42  [0x7fff9576627c]
2391 malloc_zone_malloc  (in
libsystem_malloc.dylib) + 71  [0x7fff95765868]
  2388 szone_malloc_should_clear  (in
libsystem_malloc.dylib) + 1052  [0x7fff9576369f]
  + 2388 syscall_thread_switch  (in
libsystem_kernel.dylib) + 10  [0x7fff95be0b16]
  3 szone_malloc_should_clear  (in
libsystem_malloc.dylib) + 1055,1052,...
[0x7fff957636a2,0x7fff9576369f,...]

Total number in stack (recursive counted multiple, when >=5):

Sort by top of stack, same collapsed (when >= 5):
syscall_thread_switch  (in libsystem_kernel.dylib)2388
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Budgeting & investing in features ...

[Norbert Thiebaud]

On Thu, Dec 8, 2016 at 6:34 AM, Noel Grandin  wrote:
> Hi
>
> More CI hardware.
>
> With more people and people being keener to use the CI infrastructure, I'm
> frequently seeing 3h+ turn-around times.

what would help a lot that, without new hardware, it to improve the
testing so that hung test are a thing of the past
3 hours is the current global timeout for gerrit build...

I've just kill a linux tb build that was hung for 5 hours
and this
http://ci.libreoffice.org/job/lo_gerrit/Config=windows_msc_dbgutil_32/3921/console
is already a 1h40+ for a build which on that machine should take 30-35
minutes these days

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Budgeting & investing in features ...

[Norbert Thiebaud]

On Fri, Dec 9, 2016 at 9:45 AM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
> On Thu, Dec 8, 2016 at 6:34 AM, Noel Grandin <noelgran...@gmail.com> wrote:
>> Hi
>>
>> More CI hardware.
>>
>> With more people and people being keener to use the CI infrastructure, I'm
>> frequently seeing 3h+ turn-around times.
>
> what would help a lot that, without new hardware, it to improve the
> testing so that hung test are a thing of the past
> 3 hours is the current global timeout for gerrit build...
>
> I've just kill a linux tb build that was hung for 5 hours
> and this
> http://ci.libreoffice.org/job/lo_gerrit/Config=windows_msc_dbgutil_32/3921/console
> is already a 1h40+ for a build which on that machine should take 30-35


FYI the above is stuck in java test : sc_unoapi_4
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: distro-configs/Jenkins

[Norbert Thiebaud]

 distro-configs/Jenkins/linux_random.conf |1 +
 1 file changed, 1 insertion(+)

New commits:
commit 83c6c871db3eb11e284909902e0dc639e9753136
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Wed Dec 7 17:36:43 2016 -0600

add a linux_random jenkins's distro config to be able to have a tb on that

Change-Id: I203f5e7c4f78fe57a09373b944f0d58d8c3a5d05

diff --git a/distro-configs/Jenkins/linux_random.conf 
b/distro-configs/Jenkins/linux_random.conf
new file mode 100644
index 000..4e71d38
--- /dev/null
+++ b/distro-configs/Jenkins/linux_random.conf
@@ -0,0 +1 @@
+--enable-fuzz-options
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: Build Problem with Lode

[Norbert Thiebaud]

On Mon, Dec 5, 2016 at 1:48 PM, Jason Marshall
 wrote:
> Thank you for this information and advice.

While on the topic of information and advice, please read

http://mailformat.dan.info/quoting/bottom-posting.html

Norbert

---
A: Because it messes up the order in which people normally read text.
Q: Why is top-posting such a bad thing?
A: Top-posting.
Q: What is the most annoying thing in e-mail?
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Error in ./setup script

[Norbert Thiebaud]

On Mon, Nov 28, 2016 at 1:01 AM, Jan Iversen
 wrote:
>
> I have proxy in me internet service with user authentication. I also used
> export=http_proxy=http://usern...@userpass...etc etc command in cywgin but
> that not worked.
> Can anyone give me solution?
>
>
> Setup uses a ssh connection and not a http: connection, so you need to make
> sure you ssh works.

No, setup use git:// to clone repos.
and it is not configurable.
./setup assume you have a working internet connection.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Lode Build Problem

[Norbert Thiebaud]

On Sun, Nov 27, 2016 at 12:04 PM, Jan Iversen
 wrote:

> Only use a-z and 0-9 in the name, NO spaces or other special characters"

actually _ is fine too.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Lode Build Problem

[Norbert Thiebaud]

On Sun, Nov 27, 2016 at 2:49 PM, Jason Marshall
 wrote:
> Dear Norbert/ Jan
>
> Having followed your advice, the setup of the dev build environment has now
> succeeded. Hopefully the build should proceed with similar success.

glad to hear.

>
> Thanks again.

welcome.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Regarding ./setup for download or cloning source code

[Norbert Thiebaud]

On Sat, Nov 26, 2016 at 6:01 AM, Nikhil Kumar vats
 wrote:
> Hello,
> I have installed cywgin as mentioned here
> [https://wiki.documentfoundation.org/Development/lode]
>
> When I run ./setup its says connection time out.

That is surely not the only thing it says.

for example here is the output of ./setup on an already set-up lode
./setup
Check directory 'packages' ... : Exist
Check directory 'opt' ... : Exist
Check directory 'ext_tar' ... : Exist
Check directory 'adm' ... : Exist
Check directory 'tb' ... : Exist
git repo '/Volumes/GP/lode/adm/buildbot' exist
autoconf already installed
automake already installed
make already installed
ant already installed
junit Already Installed
cmake already installed
doxygen already installed
ccache already installed

   Done.


setup does quite a few things and print helpful message to inform at
which step it is.
so that one could figure out on 'what' exactly does it timeout on.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Lode Build Problem

[Norbert Thiebaud]

On Sun, Nov 27, 2016 at 10:52 AM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
> On Sun, Nov 27, 2016 at 9:29 AM, Jason Marshall

>
> I am not certain that it is the root cause of the problme.. but
> certainly that is
> likely to cause trouble.

actually thinking more about I'm quite certain that it was the root cause:

lode create a local mirror of core.git (in LODE_HOME/mirrors/core.git

and use it as an 'alternate source' to clone core.git for /dev/core
(and toher working checkout you way want to create
using ./setup -new 

This is to save disk when one has multiple dev environement (lode was
originaly created for Continuous Integration which can have half a
dozen of such workspace)

This works by having LODE_HOME/dev/core/.git/objects/info/alternate
containint the path tot he object of the local mirror

in you case that would have been
/home/Json and Emma/lode/mirrors/core.git/objects

and I'm pretty sure that confused the heck out of git to have spaces there

Norbert

PS: even if there is a way to unconfuse git there... there are bound
to be other place in the build and the many scripts involved that do
not play well with space in filename... so, let's not seek pain for
the sake of it and just work around the issue :-)
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Lode Build Problem

[Norbert Thiebaud]

On Sun, Nov 27, 2016 at 9:29 AM, Jason Marshall
 wrote:
> Dear Sir/ Madam
>
>
> I am currently attempting to build LibreOffice on a 32-bit Windows 7
> platform using Lode.  I have got as far as trying to set-up a dev build
> environment using the following command:
>
>
> ./setup --dev
>
>
> This runs with the following output but then terminates:
>
>
> Jason and Emma@JasonandEmma-PC ~/lode
> $ ./setup --dev
> Check directory 'packages' ... : Exist
> Check directory 'opt' ... : Exist
> Check directory 'ext_tar' ... : Exist
> Check directory 'adm' ... : Exist
> Check directory 'tb' ... : Exist
> git repo '/home/Jason and Emma/lode/adm/buildbot' exist
> ant already installed
> junit Already Installed
> Check directory 'mirrors' ... : Exist
> Cloning into bare repository 'core.git'...
> remote: Counting objects: 4708711, done.
> remote: Compressing objects: 100% (993620/993620), done.
> remote: Total 4708711 (delta 3586930), reused 4702903 (delta 3581848)
> Receiving objects: 100% (4708711/4708711), 1.44 GiB | 368.00 KiB/s, done.
> Resolving deltas: 100% (3586930/3586930), done.
> Checking connectivity... done.
> Cloned mirror /home/Jason and Emma/lode/mirrors/core.git
> Check directory 'dev' ... : Created
> Cloning into 'core'...
> remote: Counting objects: 9, done.
> remote: Compressing objects: 100% (3/3), done.
> remote: Total 5 (delta 4), reused 2 (delta 2)
> fatal: mmap failed: Resource temporarily unavailable
> fatal: index-pack failed
> Error Cloning /home/Jason and Emma/lode/dev/core
>
>
> I have attempted to search the archived dev mailing lists for similar such
> issues, but the error above implies that perhaps a resource beyond my
> control, for whatever reason, may not be available at the time that I am
> running this.
>
>
> As an aside, you will see that my home directory contains spaces (i. e.
> Jason and Emma) which seems to have been where upon installation, Cygwin
> simply named the home directory after the Windows 7 user account.  Would
> this likely be an issue aggravating the error above

I am not certain that it is the root cause of the problme.. but
certainly that is
likely to cause trouble.

if you follow lode install then cygwin was installed with your user..
so you _can_ create a new home directory

assuming you are in cygwin in your current home directory 'Jason and Emma'

mv lode /. # copy the existing lode stuff.. to the root (you can do
that in this case because your user own the root filesystem

make sure to edit your .bashrc so that LODE_HOME point now to /lode
and PATH contain /lode/opt/bin at the front (so that 'make' resolve to
the right one by default when you use it fromt he command line)

terminate you cygwin session and start it again.

then

cd /lode

remove /lode/dev/core ust in case it was in bad state.

rm -fr /lode/dev/core

and run ./setup --dev again

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: distro-configs/Jenkins distro-configs/screenshot_linux.conf distro-configs/screenshot_mac.conf distro-configs/screenshot_win.conf

[Norbert Thiebaud]

 distro-configs/Jenkins/screenshot_linux.conf |1 +
 distro-configs/Jenkins/screenshot_mac.conf   |1 +
 distro-configs/Jenkins/screenshot_win.conf   |1 +
 distro-configs/screenshot_linux.conf |1 -
 distro-configs/screenshot_mac.conf   |1 -
 distro-configs/screenshot_win.conf   |1 -
 6 files changed, 3 insertions(+), 3 deletions(-)

New commits:
commit 79b9aa430b0a9846fdcb9d5e5160559c5b2a0eba
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sun Nov 27 10:23:05 2016 -0600

screenshot config at the wrong place

Change-Id: Icb72cb89f2dcc890b82b3ea6df2cc06aa902cc13

diff --git a/distro-configs/screenshot_linux.conf 
b/distro-configs/Jenkins/screenshot_linux.conf
similarity index 100%
rename from distro-configs/screenshot_linux.conf
rename to distro-configs/Jenkins/screenshot_linux.conf
diff --git a/distro-configs/screenshot_mac.conf 
b/distro-configs/Jenkins/screenshot_mac.conf
similarity index 100%
rename from distro-configs/screenshot_mac.conf
rename to distro-configs/Jenkins/screenshot_mac.conf
diff --git a/distro-configs/screenshot_win.conf 
b/distro-configs/Jenkins/screenshot_win.conf
similarity index 100%
rename from distro-configs/screenshot_win.conf
rename to distro-configs/Jenkins/screenshot_win.conf
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

[Libreoffice-commits] core.git: distro-configs/screenshot_linux.conf distro-configs/screenshot_mac.conf distro-configs/screenshot_win.conf

[Norbert Thiebaud]

 distro-configs/screenshot_linux.conf |1 +
 distro-configs/screenshot_mac.conf   |1 +
 distro-configs/screenshot_win.conf   |1 +
 3 files changed, 3 insertions(+)

New commits:
commit 8799120b372c2d60b00ad0993d0a2c6e68b841c5
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sun Nov 27 10:14:06 2016 -0600

add screenshot jenkins config

Change-Id: Ic6e2a68f948c62cabc210a1873baffe2df300e8b

diff --git a/distro-configs/screenshot_linux.conf 
b/distro-configs/screenshot_linux.conf
new file mode 100644
index 000..f93c7a7
--- /dev/null
+++ b/distro-configs/screenshot_linux.conf
@@ -0,0 +1 @@
+--with-lang=ALL
diff --git a/distro-configs/screenshot_mac.conf 
b/distro-configs/screenshot_mac.conf
new file mode 100644
index 000..f93c7a7
--- /dev/null
+++ b/distro-configs/screenshot_mac.conf
@@ -0,0 +1 @@
+--with-lang=ALL
diff --git a/distro-configs/screenshot_win.conf 
b/distro-configs/screenshot_win.conf
new file mode 100644
index 000..f93c7a7
--- /dev/null
+++ b/distro-configs/screenshot_win.conf
@@ -0,0 +1 @@
+--with-lang=ALL
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: Firebird backward incompatibility

[Norbert Thiebaud]

On Fri, Nov 25, 2016 at 9:46 AM, Lionel Elie Mamane  wrote:
>
>> The shell script would be something like:
>
>> unzip -d oldFirebird.odb temp;
>> gbak -b temp/database/firebird.fdb temp/database/firebird.fbk;
> insert here:
>  rm temp/database/firebird.fdb
>> zip -FSr oldFirebird.odb temp;
>> rm -r temp;
>
>> It would be fun to do this, though I don't have much experience with
>> web services.
>
> I assume we can get some TDF hosting for that?

Does anyone has considered the glaringly obvious privacy issue of such scheme ?

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: sw/CppunitTest_sw_dialogs_test_2.mk sw/Module_sw.mk sw/qa

[Norbert Thiebaud]

 sw/CppunitTest_sw_dialogs_test_2.mk   |   71 +
 sw/Module_sw.mk   |1 
 sw/qa/unit/data/sw-dialogs-test_2.txt |   81 ++
 sw/qa/unit/sw-dialogs-test_2.cxx  |   71 +
 4 files changed, 224 insertions(+)

New commits:
commit f888d17127557a8fd3867896dcfa1abbb9f6a53d
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sun Nov 20 07:41:10 2016 -0600

split sw screenshots test

Change-Id: I32e6f1e11f9e245f8b67fa21799b1927bb209a0d
Reviewed-on: https://gerrit.libreoffice.org/31030
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/sw/CppunitTest_sw_dialogs_test_2.mk 
b/sw/CppunitTest_sw_dialogs_test_2.mk
new file mode 100644
index 000..7e6fb0b
--- /dev/null
+++ b/sw/CppunitTest_sw_dialogs_test_2.mk
@@ -0,0 +1,71 @@
+# -*- Mode: makefile-gmake; tab-width: 4; indent-tabs-mode: t -*-
+#*
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#*
+
+$(eval $(call gb_CppunitTest_CppunitScreenShot,sw_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_add_exception_objects,sw_dialogs_test2, \
+sw/qa/unit/sw-dialogs-test_2 \
+))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,sw_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_set_include,sw_dialogs_test2,\
+-I$(SRCDIR)/sw/source/ui/inc \
+-I$(SRCDIR)/sw/inc \
+$$(INCLUDE) \
+))
+
+$(eval $(call gb_CppunitTest_use_libraries,sw_dialogs_test2, \
+basegfx \
+comphelper \
+cppu \
+cppuhelper \
+drawinglayer \
+editeng \
+i18nlangtag \
+i18nutil \
+msfilter \
+oox \
+sal \
+salhelper \
+sax \
+sw \
+swui \
+sfx \
+sot \
+svl \
+svt \
+svx \
+svxcore \
+test \
+tl \
+tk \
+ucbhelper \
+unotest \
+utl \
+vcl \
+xo \
+$(gb_UWINAPI) \
+))
+
+$(eval $(call gb_CppunitTest_use_external,sw_dialogs_test2,boost_headers))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,sw_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_ure,sw_dialogs_test2))
+$(eval $(call 
gb_CppunitTest_use_vcl_non_headless_with_windows,sw_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_rdb,sw_dialogs_test2,services))
+
+# $(eval $(call gb_CppunitTest_use_configuration,sw_dialogs_test2))
+$(eval $(call gb_CppunitTest_use_instdir_configuration,sw_dialogs_test2))
+
+# vim: set noet sw=4 ts=4:
diff --git a/sw/Module_sw.mk b/sw/Module_sw.mk
index b5950bb..3b5d06f 100644
--- a/sw/Module_sw.mk
+++ b/sw/Module_sw.mk
@@ -107,6 +107,7 @@ endif
 # screenshots
 $(eval $(call gb_Module_add_screenshot_targets,sw,\
 CppunitTest_sw_dialogs_test \
+CppunitTest_sw_dialogs_test_2 \
 ))
 
 # vim: set noet sw=4 ts=4:
diff --git a/sw/qa/unit/data/sw-dialogs-test_2.txt 
b/sw/qa/unit/data/sw-dialogs-test_2.txt
new file mode 100644
index 000..797c410
--- /dev/null
+++ b/sw/qa/unit/data/sw-dialogs-test_2.txt
@@ -0,0 +1,81 @@
+# -*- Mode: C++; tab-width: 4; indent-tabs-mode: nil; c-basic-offset: 4 -*-
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+
+# see sw-dialogs-test.txt
+
+modules/swriter/ui/optcaptionpage.ui
+modules/swriter/ui/optcomparison.ui
+modules/swriter/ui/optcompatpage.ui
+modules/swriter/ui/optfonttabpage.ui
+modules/swriter/ui/optformataidspage.ui
+modules/swriter/ui/optgeneralpage.ui
+modules/swriter/ui/optredlinepage.ui
+modules/swriter/ui/opttablepage.ui
+modules/swriter/ui/opttestpage.ui
+modules/swriter/ui/outlinenumbering.ui
+modules/swriter/ui/outlinepositionpage.ui
+modules/swriter/ui/paradialog.ui
+modules/swriter/ui/previewzoomdialog.ui
+modules/swriter/ui/printeroptions.ui
+modules/swriter/ui/printmergedialog.ui
+modules/swriter/ui/printmonitordialog.ui
+modules/swriter/ui/printoptionspage.ui
+modules/swriter/ui/privateuserpage.ui
+modules/swriter/ui/querycontinuebegindialog.ui
+modules/swriter/ui/querycontinueenddialog.ui
+modules/swriter/ui/querydefaultcompatdialog.ui
+modules/swriter/ui/queryrotateintostandarddialog.ui
+modules/swriter/ui/querysavelabeldialog.ui
+modules/swriter/ui/queryshowchangesdialog.ui
+modules/swriter/ui/renameautotextdialog.ui
+modules/swriter/ui/renameentrydialog.ui
+modules/swriter/ui/renameobjectdialog.ui
+modules/swriter/ui/rowheight.ui
+modules/swriter/ui/saveashtmldialog.ui
+modules/swriter/ui/savelabeldialog.ui
+modules/swriter/ui/sectionpage.ui
+modules/swriter/ui/sele

[Libreoffice-commits] core.git: cui/CppunitTest_cui_dialogs_test_2.mk cui/CppunitTest_cui_dialogs_test_3.mk cui/CppunitTest_cui_dialogs_test_4.mk cui/Module_cui.mk cui/qa

[Norbert Thiebaud]

 cui/CppunitTest_cui_dialogs_test_2.mk   |   68 +
 cui/CppunitTest_cui_dialogs_test_3.mk   |   68 +
 cui/CppunitTest_cui_dialogs_test_4.mk   |   68 +
 cui/Module_cui.mk   |3 
 cui/qa/unit/cui-dialogs-test_2.cxx  |   71 ++
 cui/qa/unit/cui-dialogs-test_3.cxx  |   71 ++
 cui/qa/unit/cui-dialogs-test_4.cxx  |   71 ++
 cui/qa/unit/data/cui-dialogs-test.txt   |  218 ++--
 cui/qa/unit/data/cui-dialogs-test_2.txt |   62 +
 cui/qa/unit/data/cui-dialogs-test_3.txt |   44 ++
 cui/qa/unit/data/cui-dialogs-test_4.txt |   74 ++
 11 files changed, 643 insertions(+), 175 deletions(-)

New commits:
commit 9be04c6cc69b423727038e95a43afd3ad03c22d5
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sun Nov 20 17:00:23 2016 -0600

Re-apply "split cui screenshot test in manageable pieces""

Change-Id: I22f99f830a3b4b6ba50514b7db076ea1a41478de
Reviewed-on: https://gerrit.libreoffice.org/31032
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/cui/CppunitTest_cui_dialogs_test_2.mk 
b/cui/CppunitTest_cui_dialogs_test_2.mk
new file mode 100644
index 000..24bf50e
--- /dev/null
+++ b/cui/CppunitTest_cui_dialogs_test_2.mk
@@ -0,0 +1,68 @@
+# -*- Mode: makefile-gmake; tab-width: 4; indent-tabs-mode: t -*-
+#*
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#*
+
+$(eval $(call gb_CppunitTest_CppunitScreenShot,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_add_exception_objects,cui_dialogs_test2, \
+cui/qa/unit/cui-dialogs-test_2 \
+))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_set_include,cui_dialogs_test2,\
+-I$(SRCDIR)/cui/source/inc \
+-I$(SRCDIR)/cui/inc \
+$$(INCLUDE) \
+))
+
+$(eval $(call gb_CppunitTest_use_libraries,cui_dialogs_test2, \
+basegfx \
+comphelper \
+cppu \
+cppuhelper \
+drawinglayer \
+editeng \
+i18nlangtag \
+i18nutil \
+msfilter \
+oox \
+sal \
+salhelper \
+sax \
+sfx \
+sot \
+svl \
+svt \
+cui \
+test \
+tl \
+tk \
+ucbhelper \
+unotest \
+utl \
+vcl \
+xo \
+$(gb_UWINAPI) \
+))
+
+$(eval $(call gb_CppunitTest_use_external,cui_dialogs_test2,boost_headers))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_ure,cui_dialogs_test2))
+$(eval $(call 
gb_CppunitTest_use_vcl_non_headless_with_windows,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_rdb,cui_dialogs_test2,services))
+
+# $(eval $(call gb_CppunitTest_use_configuration,cui_dialogs_test2))
+$(eval $(call gb_CppunitTest_use_instdir_configuration,cui_dialogs_test2))
+
+# vim: set noet sw=4 ts=4:
diff --git a/cui/CppunitTest_cui_dialogs_test_3.mk 
b/cui/CppunitTest_cui_dialogs_test_3.mk
new file mode 100644
index 000..80511e4
--- /dev/null
+++ b/cui/CppunitTest_cui_dialogs_test_3.mk
@@ -0,0 +1,68 @@
+# -*- Mode: makefile-gmake; tab-width: 4; indent-tabs-mode: t -*-
+#*
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#*
+
+$(eval $(call gb_CppunitTest_CppunitScreenShot,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_add_exception_objects,cui_dialogs_test3, \
+cui/qa/unit/cui-dialogs-test_3 \
+))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_set_include,cui_dialogs_test3,\
+-I$(SRCDIR)/cui/source/inc \
+-I$(SRCDIR)/cui/inc \
+$$(INCLUDE) \
+))
+
+$(eval $(call gb_CppunitTest_use_libraries,cui_dialogs_test3, \
+basegfx \
+comphelper \
+cppu \
+cppuhelper \
+drawinglayer \
+editeng \
+i18nlangtag \
+i18nutil \
+msfilter \
+oox \
+sal \
+salhelper \
+sax \
+sfx \
+sot \
+svl \
+svt \
+cui \
+test \
+tl \
+tk \
+ucbhelper \
+unotest \
+utl \
+vcl \
+xo \
+$(gb_UWINAPI) \
+))
+
+$(eval $(call gb_CppunitTest_use_external,cui_dialogs_test3,boost_headers))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_use_ure,c

Re: Revert of "split cui screenshot test in manageable pieces"

[Norbert Thiebaud]

On Sun, Nov 20, 2016 at 4:02 AM, julien2412 [via Document Foundation
Mail Archive]  wrote:
> Hello Norbert/Thorsten
>
> Just for your information, I reverted "split cui screenshot test in
> manageable pieces" patch because it made all TBs red.

It was missing a prereq change that should have been in a patch
series... I've pushed the pre-req.. I'll re-revert later

Norbert
>
> Regards,
> Julien
>
> 
> If you reply to this email, your message will be added to the discussion
> below:
> http://nabble.documentfoundation.org/Revert-of-split-cui-screenshot-test-in-manageable-pieces-tp4200364.html
> This email was sent by julien2412 (via Nabble)
> To receive all replies by email, subscribe to this discussion




--
View this message in context: 
http://nabble.documentfoundation.org/Revert-of-split-cui-screenshot-test-in-manageable-pieces-tp4200364p4200385.html
Sent from the Dev mailing list archive at Nabble.com.___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: basctl/CppunitTest_basctl_dialogs_test.mk chart2/CppunitTest_chart2_dialogs_test.mk cui/CppunitTest_cui_dialogs_test.mk dbaccess/CppunitTest_dbaccess_dialogs_test.mk de

[Norbert Thiebaud]

 basctl/CppunitTest_basctl_dialogs_test.mk |2 +-
 chart2/CppunitTest_chart2_dialogs_test.mk |2 +-
 cui/CppunitTest_cui_dialogs_test.mk   |2 +-
 dbaccess/CppunitTest_dbaccess_dialogs_test.mk |2 +-
 desktop/CppunitTest_desktop_dialogs_test.mk   |2 +-
 filter/CppunitTest_filter_dialogs_test.mk |2 +-
 formula/CppunitTest_formula_dialogs_test.mk   |2 +-
 fpicker/CppunitTest_fpicker_dialogs_test.mk   |2 +-
 reportdesign/CppunitTest_reportdesign_dialogs_test.mk |2 +-
 sd/CppunitTest_sd_dialogs_test.mk |2 +-
 sfx2/CppunitTest_sfx2_dialogs_test.mk |2 +-
 solenv/gbuild/CppunitTest.mk  |   16 +++-
 starmath/CppunitTest_starmath_dialogs_test.mk |2 +-
 svtools/CppunitTest_svtools_dialogs_test.mk   |2 +-
 svx/CppunitTest_svx_dialogs_test.mk   |2 +-
 sw/CppunitTest_sw_dialogs_test.mk |2 +-
 uui/CppunitTest_uui_dialogs_test.mk   |2 +-
 vcl/CppunitTest_vcl_dialogs_test.mk   |2 +-
 xmlsecurity/CppunitTest_xmlsecurity_dialogs_test.mk   |2 +-
 19 files changed, 33 insertions(+), 19 deletions(-)

New commits:
commit d8a8ed04b14390d4469ea19b079778fa6f074bb2
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Fri Nov 18 16:33:11 2016 -0600

gbuild allow unitest to run once per lang in WITH_LANG_LIST

this is motivated by the new screenshot feature
the initial proposed solution involved running make screenshot once per lang
which took ~6 hours for --with-lang=ALL on tb68 a reasonnably big
windows slavebot.

with this patch, one can run make screenshot just once and get all the 
screenshot
the elapsed time is 36 inutes on the same box/same config a 10x improvement.

Change-Id: I4339caebf915c118aa455de2a7e56e1a4e413939
Reviewed-on: https://gerrit.libreoffice.org/30970
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/basctl/CppunitTest_basctl_dialogs_test.mk 
b/basctl/CppunitTest_basctl_dialogs_test.mk
index 03a0287..a358372 100644
--- a/basctl/CppunitTest_basctl_dialogs_test.mk
+++ b/basctl/CppunitTest_basctl_dialogs_test.mk
@@ -9,7 +9,7 @@
 #
 #*
 
-$(eval $(call gb_CppunitTest_CppunitTest,basctl_dialogs_test))
+$(eval $(call gb_CppunitTest_CppunitScreenShot,basctl_dialogs_test))
 
 $(eval $(call gb_CppunitTest_add_exception_objects,basctl_dialogs_test, \
 basctl/qa/unit/basctl-dialogs-test \
diff --git a/chart2/CppunitTest_chart2_dialogs_test.mk 
b/chart2/CppunitTest_chart2_dialogs_test.mk
index a58559c..b7dd867 100644
--- a/chart2/CppunitTest_chart2_dialogs_test.mk
+++ b/chart2/CppunitTest_chart2_dialogs_test.mk
@@ -9,7 +9,7 @@
 #
 #*
 
-$(eval $(call gb_CppunitTest_CppunitTest,chart2_dialogs_test))
+$(eval $(call gb_CppunitTest_CppunitScreenShot,chart2_dialogs_test))
 
 $(eval $(call gb_CppunitTest_add_exception_objects,chart2_dialogs_test, \
 chart2/qa/unit/chart2-dialogs-test \
diff --git a/cui/CppunitTest_cui_dialogs_test.mk 
b/cui/CppunitTest_cui_dialogs_test.mk
index fa5d4e0..223404a 100644
--- a/cui/CppunitTest_cui_dialogs_test.mk
+++ b/cui/CppunitTest_cui_dialogs_test.mk
@@ -9,7 +9,7 @@
 #
 #*
 
-$(eval $(call gb_CppunitTest_CppunitTest,cui_dialogs_test))
+$(eval $(call gb_CppunitTest_CppunitScreenShot,cui_dialogs_test))
 
 $(eval $(call gb_CppunitTest_add_exception_objects,cui_dialogs_test, \
 cui/qa/unit/cui-dialogs-test \
diff --git a/dbaccess/CppunitTest_dbaccess_dialogs_test.mk 
b/dbaccess/CppunitTest_dbaccess_dialogs_test.mk
index 6a3ee19..fbc4ebe 100644
--- a/dbaccess/CppunitTest_dbaccess_dialogs_test.mk
+++ b/dbaccess/CppunitTest_dbaccess_dialogs_test.mk
@@ -9,7 +9,7 @@
 #
 #*
 
-$(eval $(call gb_CppunitTest_CppunitTest,dbaccess_dialogs_test))
+$(eval $(call gb_CppunitTest_CppunitScreenShot,dbaccess_dialogs_test))
 
 $(eval $(call gb_CppunitTest_add_exception_objects,dbaccess_dialogs_test, \
 dbaccess/qa/unit/dbaccess-dialogs-test \
diff --git a/desktop/CppunitTest_desktop_dialogs_test.mk 
b/desktop/CppunitTest_desktop_dialogs_test.mk
index bffa591..a00620c 100644
--- a/desktop/CppunitTest_desktop_dialogs_test.mk
+++ b/desktop/CppunitTest_desktop_dialogs_test.mk
@@ -9,7 +9,7 @@
 #
 #*
 
-$(eval $(call gb_CppunitTest_CppunitTest,desktop_dialogs_test))
+$(eval $(call gb_CppunitTest_CppunitScreenShot,desktop_dialogs_test))
 
 $(eval $(call gb_CppunitTest_add_exception_objects,desktop_dialogs_test, \
  

[Libreoffice-commits] core.git: cui/CppunitTest_cui_dialogs_test_2.mk cui/CppunitTest_cui_dialogs_test_3.mk cui/CppunitTest_cui_dialogs_test_4.mk cui/Module_cui.mk cui/qa

[Norbert Thiebaud]

 cui/CppunitTest_cui_dialogs_test_2.mk   |   68 +
 cui/CppunitTest_cui_dialogs_test_3.mk   |   68 +
 cui/CppunitTest_cui_dialogs_test_4.mk   |   68 +
 cui/Module_cui.mk   |3 
 cui/qa/unit/cui-dialogs-test_2.cxx  |   71 ++
 cui/qa/unit/cui-dialogs-test_3.cxx  |   71 ++
 cui/qa/unit/cui-dialogs-test_4.cxx  |   71 ++
 cui/qa/unit/data/cui-dialogs-test.txt   |  218 ++--
 cui/qa/unit/data/cui-dialogs-test_2.txt |   62 +
 cui/qa/unit/data/cui-dialogs-test_3.txt |   44 ++
 cui/qa/unit/data/cui-dialogs-test_4.txt |   74 ++
 11 files changed, 643 insertions(+), 175 deletions(-)

New commits:
commit c3c0557801d8713fb99216bea37443471affc1f6
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sat Nov 19 10:34:47 2016 -0600

split cui screenshot test in manageable pieces

Change-Id: I7aa6e7afecc841aecbc1acaf1a7c14ba92c089ba
Reviewed-on: https://gerrit.libreoffice.org/30981
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Thorsten Behrens <thorsten.behr...@cib.de>

diff --git a/cui/CppunitTest_cui_dialogs_test_2.mk 
b/cui/CppunitTest_cui_dialogs_test_2.mk
new file mode 100644
index 000..24bf50e
--- /dev/null
+++ b/cui/CppunitTest_cui_dialogs_test_2.mk
@@ -0,0 +1,68 @@
+# -*- Mode: makefile-gmake; tab-width: 4; indent-tabs-mode: t -*-
+#*
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#*
+
+$(eval $(call gb_CppunitTest_CppunitScreenShot,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_add_exception_objects,cui_dialogs_test2, \
+cui/qa/unit/cui-dialogs-test_2 \
+))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_set_include,cui_dialogs_test2,\
+-I$(SRCDIR)/cui/source/inc \
+-I$(SRCDIR)/cui/inc \
+$$(INCLUDE) \
+))
+
+$(eval $(call gb_CppunitTest_use_libraries,cui_dialogs_test2, \
+basegfx \
+comphelper \
+cppu \
+cppuhelper \
+drawinglayer \
+editeng \
+i18nlangtag \
+i18nutil \
+msfilter \
+oox \
+sal \
+salhelper \
+sax \
+sfx \
+sot \
+svl \
+svt \
+cui \
+test \
+tl \
+tk \
+ucbhelper \
+unotest \
+utl \
+vcl \
+xo \
+$(gb_UWINAPI) \
+))
+
+$(eval $(call gb_CppunitTest_use_external,cui_dialogs_test2,boost_headers))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_ure,cui_dialogs_test2))
+$(eval $(call 
gb_CppunitTest_use_vcl_non_headless_with_windows,cui_dialogs_test2))
+
+$(eval $(call gb_CppunitTest_use_rdb,cui_dialogs_test2,services))
+
+# $(eval $(call gb_CppunitTest_use_configuration,cui_dialogs_test2))
+$(eval $(call gb_CppunitTest_use_instdir_configuration,cui_dialogs_test2))
+
+# vim: set noet sw=4 ts=4:
diff --git a/cui/CppunitTest_cui_dialogs_test_3.mk 
b/cui/CppunitTest_cui_dialogs_test_3.mk
new file mode 100644
index 000..80511e4
--- /dev/null
+++ b/cui/CppunitTest_cui_dialogs_test_3.mk
@@ -0,0 +1,68 @@
+# -*- Mode: makefile-gmake; tab-width: 4; indent-tabs-mode: t -*-
+#*
+#
+# This file is part of the LibreOffice project.
+#
+# This Source Code Form is subject to the terms of the Mozilla Public
+# License, v. 2.0. If a copy of the MPL was not distributed with this
+# file, You can obtain one at http://mozilla.org/MPL/2.0/.
+#
+#*
+
+$(eval $(call gb_CppunitTest_CppunitScreenShot,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_add_exception_objects,cui_dialogs_test3, \
+cui/qa/unit/cui-dialogs-test_3 \
+))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_set_include,cui_dialogs_test3,\
+-I$(SRCDIR)/cui/source/inc \
+-I$(SRCDIR)/cui/inc \
+$$(INCLUDE) \
+))
+
+$(eval $(call gb_CppunitTest_use_libraries,cui_dialogs_test3, \
+basegfx \
+comphelper \
+cppu \
+cppuhelper \
+drawinglayer \
+editeng \
+i18nlangtag \
+i18nutil \
+msfilter \
+oox \
+sal \
+salhelper \
+sax \
+sfx \
+sot \
+svl \
+svt \
+cui \
+test \
+tl \
+tk \
+ucbhelper \
+unotest \
+utl \
+vcl \
+xo \
+$(gb_UWINAPI) \
+))
+
+$(eval $(call gb_CppunitTest_use_external,cui_dialogs_test3,boost_headers))
+
+$(eval $(call gb_CppunitTest_use_sdk_api,cui_dialogs_test3))
+
+$(eval $(call gb_CppunitTest_use_ure,cui_dialogs_test3))
+$(eval $(call 
gb_CppunitTest_use_vcl_non_hea

Re: Firebird build fails on MacOs (tdf#101789)

[Norbert Thiebaud]

On Mon, Nov 7, 2016 at 8:36 AM, Alexander Thurgood
<alex.thurg...@gmail.com> wrote:
> Le 07/11/2016 à 15:27, Norbert Thiebaud a écrit :
>
>>
>> you own shell cannot live in /usr/bin or /bin since mac in anny-mode
>> will prevent even root from putting stuff there.
>> so you need your own shell _and_ make sure that nothing you use below
>> have a #!/bin/sh or #!/usr/bin/sh shebang in it
>>
>
>
> Well that kind of puts the nail in the coffin for having Firebird3 on
> Mac then or have I (hopefully) misunderstood ?

No that means that firebird 3 need to learn to do thing without
relying on DYLD_LIBRARY_PATH

most of the problem lies in it wanting to use its own 'just built'
binary during the build process, using them from a temporary place
where they are built, not where they are meant to be delivered to

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Firebird build fails on MacOs (tdf#101789)

[Norbert Thiebaud]

On Mon, Nov 7, 2016 at 7:53 AM, Christian Lohmaier
 wrote:
> On Mon, Nov 7, 2016 at 2:42 PM, Jan Iversen  
> wrote:
>>
>>> On 7 Nov 2016, at 14:40, julien2412  wrote:
>
> yahoo is a bad choice for mailing lists, since it uses DMARC policy
> that tells the receiver to reject the mail if it doesn't match the
> mail's signature. As the list modifies the message by adding a footer
> (and also replacing other headers), your message will either end up in
> spam or dropped completely for quite a number of users
>
>>> Firebird 3 build fails on MacOs (see
>>> https://bugs.documentfoundation.org/show_bug.cgi?id=101789).
>>> It seems due to a pb with ICU and even after upgrade to ICU 5.8, the pb is
>>> still here.
>>>
>>> Would someone have some idea?
>
> Use an older version of Mac OSX that doesn't prevent setting
> DYLD_PATH/using simillar stuff.
>
> (if you feel like it, you can build your own shell that doesn't filter
> out the environment variables, but unless you want to use/work on
> firedb, I'd not bother with it...)

sadly that is not enough.

you own shell cannot live in /usr/bin or /bin since mac in anny-mode
will prevent even root from putting stuff there.
so you need your own shell _and_ make sure that nothing you use below
have a #!/bin/sh or #!/usr/bin/sh shebang in it

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: config folders

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 2:41 PM, Beatriz Fonseca  wrote:
> I'm trying to compile cui/source/options/personalization.cxx but I got this
> error:
> cui/source/options/personalization.cxx:10:28: fatal error: config_folders.h:
> No such file or directory
>  #include 

config_folder.h
should be in ./config_host/config_folder.h

it is generated from config_folder.h.in during the autogen.sh phase


when you said 'I'm trying to compile .' what exactly do you do, form where ?

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: assert in vcl 'builder.hxx' line 404 somehow trigger a dialog on windows slaves -> hang

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 3:45 PM, Caolán McNamara <caol...@redhat.com> wrote:
> On Sat, 2016-11-05 at 04:20 -0500, Norbert Thiebaud wrote:
>> All the slave bot were hung this morning
>> it turned out that the assert aforementionned was triggering
>> and somehow pop-in up a dialog...
>
> Seeing as its windows specific I assume that the SvxColorListBox
> factory doesn't end up in libsvx, but libsvxcore and hence the dialog
> that it can't be found.

I have absolutely no idea... I've just been playing wack-a-mole with
the windows slavebot all day...

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: assert in vcl 'builder.hxx' line 404 somehow trigger a dialog on windows slaves -> hang

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 12:49 PM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
> On Sat, Nov 5, 2016 at 4:20 AM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
>> All the slave bot were hung this morning
>> it turned out that the assert aforementionned was triggering
>> and somehow pop-in up a dialog... there was dozen of these stack per box
>> the 'mechanism' to cancel hung build was not able to clean these up so
>> they stock-piled... messing up further attempt to build.
>>
>
> bisection gave me
>
> There are only 'skip'ped commits left to test.
> The first bad commit could be any of:
> 08d6cd788f2584ce10ab8fa10665245e953c59d9
> a19b18ad7c9eb0197c10e6d7e451ec4542e4bc9e
> a989a0b1f2b425f05b58d0e44ce2de31c842ed65
> df738e0f8ceedb4bad756960be14d9c41adc165d
> 351a97ce6bda3075677b59fa1387ba3d1ab17d7a
> 760a198e697f3070a5e0e029e4eff7be220eb9cd
> 604b35bf55351751a396e34dcca3f85e75860fd5
> 8bea644d6117a49405e6426dc97214220fc869d1
> 347c2c334589b18cc62af292674bb3df1dd54b71
> d2ce812f1d3a7a2aad89ca0bd11948b63d2db7b0
> 43bc3031483d172eccd72c3804e2d4fc2ef37de4
> f300754bb1c6a347c92bb9548be7a65237176542
> f9a2c1c12ecad833c63b894c89d6008907477eb5
> We cannot bisect more!
>
>
> of course a bunch of these do not even compile... so sorry cant narrow it 
> down.
>

so I revert the whole patch series... and that now build and pass
check on windows.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: chart2/source chart2/uiconfig cui/source cui/uiconfig extensions/source extras/source include/sfx2 include/svtools include/svx reportdesign/inc reportdesign/source sc/i

[Norbert Thiebaud]

   |2 
 142 files changed, 2725 insertions(+), 1488 deletions(-)

New commits:
commit db380aab1063e8a5e40111c40ee9f7921aa82601
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sat Nov 5 13:28:17 2016 -0500

Reverts a commit series that cripple windows ci.

Revert "SvxShadowTabPage::Construct was removed"
 f9a2c1c12ecad833c63b894c89d6008907477eb5.
Revert "replace OColorPopup with SvxColorWindow"
 f300754bb1c6a347c92bb9548be7a65237176542.
Revert "drop AutoColorInvalid/SID_ATTR_AUTO_COLOR_INVALID"
 347c2c334589b18cc62af292674bb3df1dd54b71.
Revert "replace last ColorLB use with a listbox of colors"
 604b35bf55351751a396e34dcca3f85e75860fd5.
Revert "simplify, its just a vector of colors"
 351a97ce6bda3075677b59fa1387ba3d1ab17d7a.
Revert "replace user draw with an Image of the color"
 df738e0f8ceedb4bad756960be14d9c41adc165d.
Revert "strip down to the used pieces"
 commit 08d6cd788f2584ce10ab8fa10665245e953c59d9.
Revert "move now combined ColorLB to location of last user"
 a19b18ad7c9eb0197c10e6d7e451ec4542e4bc9e.
Revert "fold ColorListBox and ColorLB together"
 a989a0b1f2b425f05b58d0e44ce2de31c842ed65.
Revert "move ColorListBox beside sole thing that uses it"
 760a198e697f3070a5e0e029e4eff7be220eb9cd.
Revert "extensions leaks out details of Color Selector, patch it up"
 8bea644d6117a49405e6426dc97214220fc869d1.
Revert "inherit FillAttrLB from ListBox not ColorListBox"
 d2ce812f1d3a7a2aad89ca0bd11948b63d2db7b0.
Revert "unify color selectors"
 43bc3031483d172eccd72c3804e2d4fc2ef37de4.

diff --git a/chart2/source/controller/dialogs/dlg_View3D.cxx 
b/chart2/source/controller/dialogs/dlg_View3D.cxx
index 7cfd50f..eb0150b 100644
--- a/chart2/source/controller/dialogs/dlg_View3D.cxx
+++ b/chart2/source/controller/dialogs/dlg_View3D.cxx
@@ -39,7 +39,7 @@ using namespace ::com::sun::star::chart2;
 
 sal_uInt16 View3DDialog::m_nLastPageId = 0;
 
-View3DDialog::View3DDialog(vcl::Window* pParent, const uno::Reference< 
frame::XModel > & xChartModel)
+View3DDialog::View3DDialog(vcl::Window* pParent, const uno::Reference< 
frame::XModel > & xChartModel, const XColorListRef  )
 : TabDialog(pParent, "3DViewDialog", "modules/schart/ui/3dviewdialog.ui")
 , m_pGeometry(nullptr)
 , m_pAppearance(nullptr)
@@ -51,7 +51,7 @@ View3DDialog::View3DDialog(vcl::Window* pParent, const 
uno::Reference< frame::XM
 uno::Reference< beans::XPropertySet > xSceneProperties( 
ChartModelHelper::findDiagram( xChartModel ), uno::UNO_QUERY );
 m_pGeometry   = 
VclPtr::Create(m_pTabControl,xSceneProperties,m_aControllerLocker);
 m_pAppearance = 
VclPtr::Create(m_pTabControl,xChartModel,m_aControllerLocker);
-m_pIllumination = 
VclPtr::Create(m_pTabControl,xSceneProperties,xChartModel);
+m_pIllumination = 
VclPtr::Create(m_pTabControl,xSceneProperties,xChartModel,pColorTable);
 
 m_pTabControl->InsertPage( TP_3D_SCENEGEOMETRY, 
SCH_RESSTR(STR_PAGE_PERSPECTIVE) );
 m_pTabControl->InsertPage( TP_3D_SCENEAPPEARANCE, 
SCH_RESSTR(STR_PAGE_APPEARANCE) );
diff --git a/chart2/source/controller/dialogs/tp_3D_SceneIllumination.cxx 
b/chart2/source/controller/dialogs/tp_3D_SceneIllumination.cxx
index 40f2857..bdd53ff 100644
--- a/chart2/source/controller/dialogs/tp_3D_SceneIllumination.cxx
+++ b/chart2/source/controller/dialogs/tp_3D_SceneIllumination.cxx
@@ -22,7 +22,6 @@
 #include "Bitmaps.hrc"
 #include "CommonConverters.hxx"
 
-#include 
 #include "svx/dialogs.hrc"
 #include 
 #include 
@@ -104,7 +103,7 @@ void LightSourceInfo::initButtonFromSource()
 
 namespace
 {
-OUString lcl_makeColorName(const Color& rColor)
+OUString lcl_makeColorName( Color rColor )
 {
 OUString aStr = SVX_RESSTR(RID_SVXFLOAT3D_FIX_R) +
 OUString::number(rColor.GetRed()) +
@@ -116,11 +115,15 @@ namespace
 OUString::number(rColor.GetBlue());
 return aStr;
 }
-
-void lcl_selectColor(SvxColorListBox& rListBox, const Color& rColor)
+void lcl_selectColor( ColorListBox& rListBox, const Color& rColor )
 {
 rListBox.SetNoSelection();
-rListBox.SelectEntry(std::make_pair(rColor, 
lcl_makeColorName(rColor)));
+rListBox.SelectEntry( rColor );
+if( rListBox.GetSelectEntryCount() == 0 )
+{
+const sal_Int32 nPos = rListBox.InsertEntry( rColor, 
lcl_makeColorName( rColor ) );
+rListBox.SelectEntryPos( nPos );
+}
 }
 
 ::chart::LightSource lcl_getLightSourceFromProperties(
@@ -212,7 +215,8 @@ namespace
 
 ThreeD_SceneIllumination_TabPage::ThreeD_SceneIllumination_TabPage( 
vcl::Window* pWindow
 , const uno::Refere

Re: slower build.... sc_functions_test is a long tail end...

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 12:34 PM, Markus Mohrhard
 wrote:
> Hey Norbert,
>
> In which configuration? The test suffers extremely from.missing optimization
> and should be less than a minute in an optimized bud but can run for several
> minutes in a digital build.

of course we run make check in dbgutil/debug mode.
if they should not be ran in these config then have them excluded then.

and several here = 9 minutes cpu... which is 1/3 of the total build
elapsed time on that box.

btw the second worse offender is char2_export ccpunnit which run at
almost 6 minutes...

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: assert in vcl 'builder.hxx' line 404 somehow trigger a dialog on windows slaves -> hang

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 4:20 AM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
> All the slave bot were hung this morning
> it turned out that the assert aforementionned was triggering
> and somehow pop-in up a dialog... there was dozen of these stack per box
> the 'mechanism' to cancel hung build was not able to clean these up so
> they stock-piled... messing up further attempt to build.
>

bisection gave me

There are only 'skip'ped commits left to test.
The first bad commit could be any of:
08d6cd788f2584ce10ab8fa10665245e953c59d9
a19b18ad7c9eb0197c10e6d7e451ec4542e4bc9e
a989a0b1f2b425f05b58d0e44ce2de31c842ed65
df738e0f8ceedb4bad756960be14d9c41adc165d
351a97ce6bda3075677b59fa1387ba3d1ab17d7a
760a198e697f3070a5e0e029e4eff7be220eb9cd
604b35bf55351751a396e34dcca3f85e75860fd5
8bea644d6117a49405e6426dc97214220fc869d1
347c2c334589b18cc62af292674bb3df1dd54b71
d2ce812f1d3a7a2aad89ca0bd11948b63d2db7b0
43bc3031483d172eccd72c3804e2d4fc2ef37de4
f300754bb1c6a347c92bb9548be7a65237176542
f9a2c1c12ecad833c63b894c89d6008907477eb5
We cannot bisect more!


of course a bunch of these do not even compile... so sorry cant narrow it down.


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Git Branches

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 8:23 AM, Beatriz Fonseca  wrote:
> Hi everyone.
>
> I have a question. Which branch I could use to fix Mozilla themes?

master.

push via gerrit for review.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

slower build.... sc_functions_test is a long tail end...

[Norbert Thiebaud]

sc_functions_test is taking about 9 minutes cpu to run on windows
that is way too long for a single process.
unittest should really not take longer than 2 minutes so that we can
parallelize them properly.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: assert in vcl 'builder.hxx' line 404 somehow trigger a dialog on windows slaves -> hang

[Norbert Thiebaud]

On Sat, Nov 5, 2016 at 4:20 AM, Norbert Thiebaud <nthieb...@gmail.com> wrote:
> All the slave bot were hung this morning
> it turned out that the assert aforementionned was triggering
> and somehow pop-in up a dialog... there was dozen of these stack per box
> the 'mechanism' to cancel hung build was not able to clean these up so
> they stock-piled... messing up further attempt to build.
>

Looks like we need somthing like

_CrtSetReportMode(_CRT_WARN, _CRTDBG_MODE_DEBUG|_CRTDBG_MODE_FILE);
_CrtSetReportFile(_CRT_WARN, _CRTDBG_FILE_STDERR);

which cppuunittester already do, but also for other tests.
iow probably builtin libreoffice when doing automated-ci build
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

assert in vcl 'builder.hxx' line 404 somehow trigger a dialog on windows slaves -> hang

[Norbert Thiebaud]

All the slave bot were hung this morning
it turned out that the assert aforementionned was triggering
and somehow pop-in up a dialog... there was dozen of these stack per box
the 'mechanism' to cancel hung build was not able to clean these up so
they stock-piled... messing up further attempt to build.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Björn Balazs license statement

[Norbert Thiebaud]

On Tue, Nov 1, 2016 at 2:35 PM, Wols Lists  wrote:
>>
> Because giving orders to volunteers isn't actually a very good idea?

there was no 'order' given.
what was said is:
"Please be aware this license cannot be used for development activities."
which is a factual statement regarding the current rules in force in
the project.

>
> Sorry, but I'm probably not unique in getting upset when told to do
> something "just because".

It is not 'just because'. That you choose to ignore the reason behind
the choice of license, and the unfortunate necessity to manage these
things does not make the decision whimsical.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: bin/check-elf-dynamic-objects

[Norbert Thiebaud]

 bin/check-elf-dynamic-objects |2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

New commits:
commit 91d406f5afb6e08f418d14761beb4a5af5783275
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Mon Oct 10 23:11:55 2016 -0500

check-elf whitelist libxml2 not libxml

Change-Id: If0179726861c867f9c25c13f9eb0b61faaee9524

diff --git a/bin/check-elf-dynamic-objects b/bin/check-elf-dynamic-objects
index fa46cb7..7d444ee 100755
--- a/bin/check-elf-dynamic-objects
+++ b/bin/check-elf-dynamic-objects
@@ -62,7 +62,7 @@ programfiles=$(basename -a $(echo ${files} | grep -o 
'/program/[^/]* '))
 
 # whitelists should contain only system libraries that have a good reputation
 # of maintaining ABI stability
-globalwhitelist="ld-linux-x86-64.so.2 libc.so.6 libm.so.6 libdl.so.2 
libpthread.so.0 librt.so.1 libutil.so.1 libnsl.so.1 libcrypt.so.1 libgcc_s.so.1 
libstdc++.so.6 libz.so.1 libfontconfig.so.1 libfreetype.so.6 libxml.so.2 
libxslt.so.1 libexslt.so.0"
+globalwhitelist="ld-linux-x86-64.so.2 libc.so.6 libm.so.6 libdl.so.2 
libpthread.so.0 librt.so.1 libutil.so.1 libnsl.so.1 libcrypt.so.1 libgcc_s.so.1 
libstdc++.so.6 libz.so.1 libfontconfig.so.1 libfreetype.so.6 libxml2.so.2 
libxslt.so.1 libexslt.so.0"
 x11whitelist="libX11.so.6 libXext.so.6 libSM.so.6 libICE.so.6 libXinerama.so.1 
libXrender.so.1 libXrandr.so.2 libcairo.so.2"
 openglwhitelist="libGL.so.1"
 giowhitelist="libgio-2.0.so.0 libgobject-2.0.so.0 libglib-2.0.so.0 
libdbus-glib-1.so.2 libdbus-1.so.3"
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

[Libreoffice-commits] core.git: bin/check-elf-dynamic-objects

[Norbert Thiebaud]

 bin/check-elf-dynamic-objects |   75 +++---
 1 file changed, 70 insertions(+), 5 deletions(-)

New commits:
commit 4189551d56b4c6f37e8aebb856411794ea98e44f
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Mon Oct 10 22:21:14 2016 -0500

Support parallelism for check-elf-dynamic-objects

using that and restricting the tests to the content of
instdir/program bring the runtime on out slave builder from
4 minutes+ to just above one second

added -d  to override INSTDIR
added -p to indicate parallelism (no limit for now, so be cautious
with the -d you point too

Change-Id: I14c140f54bea329b82960843a6db44bfaf6c7108

diff --git a/bin/check-elf-dynamic-objects b/bin/check-elf-dynamic-objects
index b34af3e..fa46cb7 100755
--- a/bin/check-elf-dynamic-objects
+++ b/bin/check-elf-dynamic-objects
@@ -12,7 +12,51 @@
 
 set -euo pipefail
 
-files=$(find "${INSTDIR}" -type f)
+PARA=1
+check_path="${INSTDIR:-.}/program"
+
+while [ "${1:-}" != "" ]; do
+parm=${1%%=*}
+arg=${1#*=}
+has_arg=
+if [ "${1}" != "${parm?}" ] ; then
+has_arg=1
+else
+arg=""
+fi
+
+case "${parm}" in
+--dir|-d)
+   if [ "$has_arg" ] ; then
+   check_path=$arg
+   else
+   shift
+   check_path=$1
+   fi
+;;
+-p)
+   # this sound counter intuitive. but the idea
+# is to possibly support -p 
+# in the mean time 0 = nolimit and -p 1 would mean
+# the current default: serialize
+PARA=0
+;;
+-*)
+die "Invalid option $1"
+;;
+*)
+if [ "$DO_NEW" = 1 ] ; then
+REPO="$1"
+else
+die "Invalid argument $1"
+fi
+;;
+esac
+shift
+done
+
+
+files=$(find "${check_path}" -type f)
 # all RPATHs should point to ${INSTDIR}/program so that's the files they find
 programfiles=$(basename -a $(echo ${files} | grep -o '/program/[^/]* '))
 
@@ -29,10 +73,10 @@ kde4whitelist="libkio.so.5 libkfile.so.4 libkdeui.so.5 
libkdecore.so.5 libQtNetw
 avahiwhitelist="libdbus-glib-1.so.2 libdbus-1.so.3 libgobject-2.0.so.0 
libglib-2.0.so.0 libavahi-common.so.3 libavahi-client.so.3"
 kerberoswhitelist="libgssapi_krb5.so.2 libcom_err.so.2 libkrb5.so.3"
 
-status=0
+check_one_file()
+{
+local file="$1"
 
-for file in ${files}
-do
 skip=0
 whitelist="${globalwhitelist}"
 case "${file}" in
@@ -132,7 +176,28 @@ do
 esac
 fi
 fi
-done
+}
+status=0
+
+if [ "$PARA" = "1" ] ; then
+for file in ${files}
+do
+   check_one_file $file
+done
+else
+rm -f check_elf.out
+for file in ${files}
+do
+   (
+   check_one_file $file
+   )>> check_elf.out &
+done
 
+if [ -s check_elf.out ] ; then
+   cat check_elf.out
+   status=1
+fi
+rm check_elf.out
+fi
 exit ${status}
 
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

bibisect on linux and libboost system libs -> orcus

[Norbert Thiebaud]

So,

with internal boost (iow not system-boost) build, we seems
to build/deliver only static boost lib not the *.so
but liborcus, although it is configured to use the internal boost,
still will pick up the system .so (somewhere else than told) for boost
and not use these static libs

in turn that means that when using bibisect on a linux box that does
not have these boost library installed (or too old I guess), we got
missing library failure.

I peeked at configure.ac in liborcus, and I was suspecting that adding
BOOST_STATIC may do the trick... but we use configure directly in
external/liborcus.. so patching autogen.ac directly there is prolly
not the way to go...
Note: I'm not sure that BOOST_STATIC is an answer, even less the
'right(tm)' answer

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Whitespace and other stylistic changes

[Norbert Thiebaud]

On Sun, Sep 25, 2016 at 1:14 PM, Maarten Bosmans  wrote:
>
> So my question is: how far away from a functional change is it
> desired/allowed to make stylistic changes for consistency? Is that the
> whole file, class, function, scope block, statement or line?

my two cents:

1/ it is ok to do such adjustment for line that would have been
modified otherwise (with the caveat about the 'local style')

2/ it is not ok when the signal/noise ration become annoying low, but
of course the exact level of annoyance is reviewer sensitive. it is an
absolute no go when some files in a (functional) patch have only
whitespace changes

3/ in doubt, split the whitespace changes in a separate commit.
personally I would not consider if (a) vs if ( a ) vs if(a) a
justifiable noise in and of itself. but your miles may vary

one day maybe we will have an agreed upon canonical style (not holding
my breath) at which point we probably will have automated tools to do
the initial clean-up and commit hook to enforce it.


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: gerrit changes depending one on another

[Norbert Thiebaud]

On Wed, Sep 21, 2016 at 12:15 PM, Maarten Bosmans  wrote:
> I just tried to push 3 changes to gerrit with a topic set.
> The core/logerrit script does not have an option to set a topic, so I
> did a manual git push.
>
> This did not work:
> `git push origin HEAD:refs/for/master -o topic=tdf53698`
> It failed with the message "fatal: the receiving end does not support
> push options"
>
> The other (deprecated?) way did work:
> `git push origin HEAD:refs/for/master/topic=tdf53698`

 $ git push origin HEAD:refs/for/master%topic=multi-master
https://gerrit-review.googlesource.com/Documentation/intro-user.html#topics
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: svtools/source

[Norbert Thiebaud]

 svtools/source/graphic/grfmgr2.cxx |6 --
 1 file changed, 4 insertions(+), 2 deletions(-)

New commits:
commit 91cc48f3d4b8e11f44a366e395b010203ff9c5bc
Author: Norbert Thiebaud <norb...@sqdata.com>
Date:   Tue Sep 13 12:35:43 2016 +0200

cid#1372878 modulo by zero

Change-Id: I37ed251597f44d382fde96c39ff9d07e4d14c50a
Reviewed-on: https://gerrit.libreoffice.org/28859
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
    Tested-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/svtools/source/graphic/grfmgr2.cxx 
b/svtools/source/graphic/grfmgr2.cxx
index 67ead4c..408bee5 100644
--- a/svtools/source/graphic/grfmgr2.cxx
+++ b/svtools/source/graphic/grfmgr2.cxx
@@ -1675,8 +1675,10 @@ bool GraphicObject::ImplRenderTempTile( VirtualDevice& 
rVDev,
 }
 
 // one less
-nMSBFactor /= nExponent;
-
+if(nMSBFactor > 1)
+{
+nMSBFactor /= nExponent;
+}
 ImplTileInfo aTileInfo;
 
 // #105229# Switch off mapping (converting to logic and back to
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: gerrit changes depending one on another

[Norbert Thiebaud]

On Fri, Sep 2, 2016 at 1:58 PM, Eike Rathke <er...@redhat.com> wrote:
> Hi,
>
> On Friday, 2016-09-02 12:24:54 +0200, Norbert Thiebaud wrote:
>
>> On Fri, Sep 2, 2016 at 11:12 AM, Stephan Bergmann <sberg...@redhat.com> 
>> wrote:
>> > On 09/01/2016 06:01 PM, Eike Rathke wrote:
>> >>
>> >> Or plain git: git push gerrit HEAD:refs/for/master/mytopic
>> >
>> > is is "/mytopic" or
>>
>> "%topic=mytopic"?
>> ^^^ this is the documented way ever since the version 2.6.
>
> Then both work.. just verified with
> git push logerrit HEAD:refs/for/master/testtopic

yes but that syntax is 2.5 an older and at some point it will stop working.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: gerrit changes depending one on another

[Norbert Thiebaud]

On Fri, Sep 2, 2016 at 11:12 AM, Stephan Bergmann  wrote:
> On 09/01/2016 06:01 PM, Eike Rathke wrote:
>>
>> Or plain git: git push gerrit HEAD:refs/for/master/mytopic
>
>
> is is "/mytopic" or

"%topic=mytopic"?
^^^ this is the documented way ever since the version 2.6.

norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Maarten Bosmans license statement

[Norbert Thiebaud]

>
> (unless it doesn't, and that means someone else has committed changes to the
> same chunk of code in the meantime and your patch no longer applies, but let's
> deal with that only when such situation arises)
>
>> What is the correct way to submit a patch series to gerrit? I'd like
>> to avoid squashing all patches into one commit, but there also has to
>> be a way for reviewers to see that all patches in a series are
>> related.
>
> Yours is the correct way, the reviewers are well aware of false merge 
> conflicts
> of dependent patches. I'm still not sure if this is a gerrit bug or some
> sophisticated feature *grin*

WAD
the second patch cannot be 'merged' until the first one land on
master.. or the second patch is reworked to not depend on the first
one (should the first one be abandonned for some reason)

Jenkins build each patches.. but it pull each patch with it's history.
jenkins check-out the commit related to a given patch, which means for
dependents patches the later pull the former too.

Note it is possible to push patch for master with 'topic' to visually
indicate that they are related see
https://gerrit-review.googlesource.com/Documentation/intro-user.html#topics

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Devart

[Norbert Thiebaud]

Done. these bogus 'extensions' are now removed

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Devart

[Norbert Thiebaud]

On Wed, Aug 17, 2016 at 8:33 PM, toki  wrote:
> My recommendation would be to remove _all_ extensions provided by Devart
> from the LibreOffice Extension website, and permanently ban them from
> contributing anything to the project.
>

Thanks for pointing that out.

The first issue is not even the license.. it is the fact that these
are not even 'extensions' to start with.
and yeah the lying on the license is cherry on the cake.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Services outage: gerrit, jenkins, bugzilla ...

[Norbert Thiebaud]

Due to a multiple disks failure event on falco these services are
currently offline

We are working on restoring them...


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: compiler plugin

[Norbert Thiebaud]

On Wed, Jul 13, 2016 at 7:00 AM, Stephan Bergmann  wrote:
> Not necessarily.  Consider a change being discussed between an author and a
> reviewer, leading to the author generating multiple revisions of the change.
> To track the differences between the revisions, it is best if there's no
> rebases in between.

this is not quite true anymore
see https://gerrit.libreoffice.org/#/c/27120/
which was based on a 3 weeks old commit at the time
it was rebased...
if you do a diff against version 1 in gerrit the diff presented does
not have all the changes due to the rebase.
That used to be the case, but they fixed that in gerrit.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: compiler plugin

[Norbert Thiebaud]

On Wed, Jul 13, 2016 at 4:18 AM, Stephan Bergmann  wrote:
>
> The idea behind the second choice is to periodically update and rebuild the
> bot's compilerplugins repo, so that sequences of LO builds on the bot are
> guaranteed to all use the same plugin.so and be able to reuse cached objects
> across those builds.  However, that would mean that Gerrit changes based on
> a relatively old code base could be built against a newer version of the
> compilerpluigns repo, running into warnings/errors from new plugins, the
> fixes for which across the LO code base not yet being included in the code
> base revision the change is based on.  So I think this approach isn't
> feasible.


Actually that is very feasible. since
1/ the solution for these old-based change is 'rebase'

2/ that has to be done anyway before the said patch is the be
integrated into master.


> (Another problem would be that e.g. the name of a class from the
> LO code base can be whitelisted in one of the plugins, to suppress
> warnings/errors from that class.  If the name of the class changes in the LO
> code base, the compilerplugins repo must be changed in sync.)

That sound like the whole contraption is quite fragile and random.. if
you have to 'whitelist' random 'class'
directly in the pluging... I'd suggest there is something
fondamentally wrong in the process.
whitelisting, or more eactly a way to say to the plugin: 'shut-up, I know'
should be done by annotating the source code not by patching the plugin.

>
>
> So my proposal would be as follows:  First, check whether enabling
> compiler_check=content or the compiler_check=string:X setup (and increasing
> the ccache size if necessary and possible)

the cache size is 100GB dedicated to clang... that is quite a chunk of
disk already.


> gives good-enough performance.
> If not, restrict commits to compilerplugins/ to be less frequent, and see
> whether that increases the ccache hit rate and results in good-enough
> performance.

I do not favor compiler_check=content... as this means calculating a
hash of the compiler _and the plugin every time.
the plugin.so alone is 150MB (which is quite insane btw considering
that clang itself is ~50MB)
I really do not need to waste too much time experimenting to know that
hashing 15,000 X 200MB = 3TB per build is going to be a waste.

compiler_check=string  is no better since that would actually run
these as external process for each ccache invocation
bearing in mind that a build is north of 15K of theses...

What I suggest is:

1/ allow the plugins to be built standalone and delivered (ideally,
spin it in a sub-module). Allow configure.ac to use a 'external
version of the compiler plugin.
2/ work on the plugin in a branch.. (or ideally in a sub-module), push
core.git needed fixes in advance of merging the pluging changes
3/ every so often merge the plugin changes... (if it is a submodule it
is just a matter of moving the submodule ref in core.git) (if 2/ was
followed that does not break master and master had been compatible
with if for some time
4/ at that point a jenkins job will use 1/ to deploy a new version of
the plugin... everything will be build with that from that point on
(again if it is a submodule.. that can be cloned and built
stand-alone.. which would be less wastefull thatn maintaining a full
core.git workspace for that purpose on each concerned slaves)
5/ too old gerrit patch could fails.. but that is fine, they need to
be rebased anyway.. and hopefully that will give an incentive to
people not to keep patches on old base...


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

compiler plugin

[Norbert Thiebaud]

so I ran some number... after upgrading ccache

clang+pluging+dbgutil  (time result in minutes.. elapsed/user/system)

cold: 33/840/50
hot: 9/79/14
no-op: 4/46/1

clang+dbgutil (no plugins)

cold: 26/605/46
hot: 9/79/14
no-op: 4/46/1

gcc-dbgutil

cold: 28/621/97
hot: 9/79/14
no-op: 4/45/1

note: none of these comprise make check

so the cost of the plugins on a full build is 7 minutes elapsed, ~240
minutes cpu.

ccache works fine... on the other hand any change in any of the
plugins invalidate the cache...

I've enabled an additional build for gerrit doing clang + plugins on linux
we will see how that perform in average.
preliminary observation is that there is way to much churn in the
plugins for this to be viable at this time

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: minutes of ESC call ...

[Norbert Thiebaud]

On Mon, Jul 4, 2016 at 3:04 PM, David Ostrovsky  wrote:
>> when I get a stream message about an approval +2 I do not know if
>> that
>> change was +1 verified by gerrit
>
> Nobody ever told you to use stream events, but GQL.

Yes somebody told me: vmiklos pointed me to a python example using
stream.. on face value that was not a bad idea since gerrit's doc is
saying for 'Change added:

approvals
All approval attributes granted.

but it turned out that was 'All approval attributes granted  in
_that_ specific comment', so really not all of them at all.

furthermore the json representation of a patcheset does not seems to
contain the 'is mergeable attribute'

That turned out not fitting the need. but streaming would have been a
great solution to be both reactive and cheap.

>
> Don't mix accounts and groups entities with change meta data.

I need both. The rule to be implemented is
"if a commit on master is verified _by jenkins_ AND is reviewed +2 by
_the author of the patch_ then auto-submit."

that require both accounts information _and_ change/patchsets information.


> Have you read my original mail? Here is the fixed SQL version as GQL:

Have you read the minutes you answered to ?
"if +2 __by author__ and +1 by jenkins -> auto-push"

>
> project:core
> branch:master
> is:open
> is:mergeable
> label:Code-Review+2
> label:Verified+1,c...@libreoffice.org
> NOT label:Code-Review-2
> NOT label:Verified-1

that does not implement the rule desired. see above
we want Code-Review+2 _from the author_.
that query return 4 records instead of 2 (at the time of testing)
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: minutes of ESC call ...

[Norbert Thiebaud]

On Sun, Jul 3, 2016 at 3:59 PM, David Ostrovsky  wrote:
> On Thu Jun 30 15:46:27 UTC 2016 Michael Meeks wrote:
>
> [...]
>>* Pending Action Items:
> [...]
>>+ investigate a cron job that queries & auto-merges (Norbert)
>> + if +2 by author and +1 by jenkins -> auto-push (on-going)
>>+ SQL query done.. but yield interesting result...
>>  https://gerrit.libreoffice.org/#/c/26347/  which 1/
> is marked
>>  'not mergeable' but I can't yet detect that
>>  and really should not be merged anyway.. based on
> comments.
>>  + Not found a way to detect this yet.
>
> I think you are confused.

No, although using 3 different back-end is indeed a confused design.
and yes H2 is off-limit.. but NoteDB hopefully will not be hidden
behind a wall of java crap, so presumably there will be way to access
the content efficiently.


>
> I do not have a better explanation how you came to the idea to mess
> around with backend details of the persistent layer for change meta
> data in Gerrit Code Review.

Because stream does not contain the necessary information to get the data...
when I get a stream message about an approval +2 I do not know if that
change was +1 verified by gerrit

>
> In the end you probably don't use emacs, to access content.xml from ODT
> file, if you want to read it, don't you?

I do not want to but I may have to.
for instance how exactly do you handle duplicate account using 'gerrit
front end' ?

>
> So the GQL query you are looking for, that is executed through SSH
> query command, is:
>
> $ ssh logerrit gerrit query "'project:core is:open is:mergeable
> label:Code-Review+2 label:Verified+1,c...@libreoffice.org NOT label:Code
> -Review-2 NOT label:Verified-1'"
>
> The result of it right now on gerrit.libreoffice.org are these two
> changes: 26566, 26308.
>
> You could use query --format json option and parse it with a couple of
> lines of python script, or alternatively you could add --current-patch
> -set option and grep for the "  revision:" line:
>
> $ ssh logerrit gerrit query --current-patch-set "'project:core ...'" |
> grep "  revision:" | awk '{print $2}'
> 50d05404589960804ebaba6d1ffa5afeddd1df6d
> d145feafa6c8c94b3848ecbbd624d2ceb1904b2f

select change_id || ',' || patch_set_id from patch_set_approvals where
change_id in (select c.change_id from changes c, patch_sets p,
patch_set_approvals a where status = 'n' and c.change_id = p.change_id
and c.change_id = a.change_id and p.patch_set_id =
c.current_patch_set_id and a.patch_set_id = c.current_patch_set_id and
p.uploader_account_id = a.account_id and a.category_id = 'Code-Review'
and a.value = 2) and category_id = 'Verified' and value = 1 and
account_id = 1000855

 time ./auto_merge_candidate
26347,1

real0m0.430s
user0m0.044s
sys0m0.056s

how do you suggest to write that in 'GQL' query ? (note: teh querry
above is missing a filter on the 'master' branch but that is trivial
to add )

>
> This would dump the commit ids to supply to the gerrit review command.
> Iterating over the result above and calling (untested):
>
> $ ssh logerrit gerrit review --submit 
>
> with a user that has required ACL, should do the job.

yeah except you need to find a way to analyze the result of this not
not try over and over again to submit the same patch that won't go for
some reason (merge conflict for instance). and yeah I have that
problem too.. that is the thing with production-grade automation... it
is not about the easy 95% bits.. it is about the killer corner-case...

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Voting page in wiki

[Norbert Thiebaud]

On Fri, Jun 17, 2016 at 6:28 AM, jan iversen
 wrote:
> Hi.
>
> During me work with the development pages in our Wiki, I came across this page
>
>
> https://wiki.documentfoundation.org/Vote_for_Enhancement
>
> Is that a page we actively use (if so, I need to give it more attention) ?

No.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Feature branch request

[Norbert Thiebaud]

On Fri, Jun 10, 2016 at 9:43 AM, jan iversen
 wrote:
>
> We have a slot in the weekly ESC meeting where new committers are suggested, 
> discussed and normally approved.

Khaled is a committers and the mentor of this GSOC student...
In the spirit of moving things along and not wasting a full week in
administrative trivia I have provisionally added him in the committer
list under Khaled patronage.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: configure.ac

[Norbert Thiebaud]

 configure.ac |5 -
 1 file changed, 5 deletions(-)

New commits:
commit 10fa03258fcc7d9e820bd68a25d9d781c6dd7397
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Thu Jun 9 10:34:14 2016 -0500

configure: we have enough options already no need to repeat them

Change-Id: Ie115a158a5a47265c21c49bde8e2fae5cbd60b6a
Reviewed-on: https://gerrit.libreoffice.org/26116
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Noel Grandin <noelgran...@gmail.com>

diff --git a/configure.ac b/configure.ac
index 49e7c75..3c5a130 100644
--- a/configure.ac
+++ b/configure.ac
@@ -1874,11 +1874,6 @@ AC_ARG_WITH(libloader-jar,
 [Specify path to jarfile manually.]),
 LIBLOADER_JAR=$withval)
 
-AC_ARG_WITH(libloader-jar,
-AS_HELP_STRING([--with-libloader-jar=JARFILE],
-[Specify path to jarfile manually.]),
-LIBLOADER_JAR=$withval)
-
 AC_ARG_WITH(libformula-jar,
 AS_HELP_STRING([--with-libformula-jar=JARFILE],
 [Specify path to jarfile manually.]),
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

[Libreoffice-commits] core.git: Branch 'libreoffice-5-2' - 2 commits - oox/source sd/qa solenv/gbuild vcl/source

[Norbert Thiebaud]

 oox/source/export/drawingml.cxx   |5 -
 sd/qa/unit/data/odp/tdf93883.odp  |binary
 sd/qa/unit/export-tests.cxx   |   12 
 solenv/gbuild/platform/unxgcc.mk  |3 +++
 solenv/gbuild/platform/windows.mk |3 +++
 vcl/source/app/svmain.cxx |5 -
 6 files changed, 26 insertions(+), 2 deletions(-)

New commits:
commit c720aeead20cd9f7a675970ae3810b9904dade84
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Thu May 19 13:35:13 2016 -0500

shortcut the signal handler during batch test

Reviewed-on: https://gerrit.libreoffice.org/25178
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
    Tested-by: Norbert Thiebaud <nthieb...@gmail.com>
(cherry picked from commit 9d2f7be4e65595241db3cf5135b69bd9e4ce6a30)

try to avoid 'by-design' solar-mutex deadlock during ci
Reviewed-on: https://gerrit.libreoffice.org/25839
Tested-by: Jenkins <c...@libreoffice.org>
    Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
(cherry picked from commit 6b3b352b06d92ef20194b9a992a521af2ef07b48)

Change-Id: I7848555743e8d54e755dc82cbf0148eeed54110f
Reviewed-on: https://gerrit.libreoffice.org/25978
Reviewed-by: Michael Stahl <mst...@redhat.com>
Tested-by: Jenkins <c...@libreoffice.org>

diff --git a/solenv/gbuild/platform/unxgcc.mk b/solenv/gbuild/platform/unxgcc.mk
index 278006b..bc44906 100644
--- a/solenv/gbuild/platform/unxgcc.mk
+++ b/solenv/gbuild/platform/unxgcc.mk
@@ -17,6 +17,9 @@
 #   the License at http://www.apache.org/licenses/LICENSE-2.0 .
 #
 
+# to block heavy exception handling that try to acquire the solarmutex
+export LO_LEAN_EXCEPTION=1
+
 include $(GBUILDDIR)/platform/com_GCC_defs.mk
 
 gb_MKTEMP := mktemp -t gbuild.XX
diff --git a/solenv/gbuild/platform/windows.mk 
b/solenv/gbuild/platform/windows.mk
index a2e621d..ebd2a3c 100644
--- a/solenv/gbuild/platform/windows.mk
+++ b/solenv/gbuild/platform/windows.mk
@@ -8,6 +8,9 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 #
 
+# to block heavy exception handling that try to acquire the solarmutex
+export LO_LEAN_EXCEPTION=1
+
 # to avoid flashing windows during tests
 export VCL_HIDE_WINDOWS=1
 
diff --git a/vcl/source/app/svmain.cxx b/vcl/source/app/svmain.cxx
index 2b315b2..69ab53b 100644
--- a/vcl/source/app/svmain.cxx
+++ b/vcl/source/app/svmain.cxx
@@ -95,6 +95,8 @@
 
 using namespace ::com::sun::star;
 
+static bool g_bIsLeanException;
+
 static bool isInitVCL();
 
 oslSignalAction SAL_CALL VCLExceptionSignal_impl( void* /*pData*/, 
oslSignalInfo* pInfo)
@@ -102,7 +104,7 @@ oslSignalAction SAL_CALL VCLExceptionSignal_impl( void* 
/*pData*/, oslSignalInfo
 static volatile bool bIn = false;
 
 // if we crash again, bail out immediately
-if ( bIn )
+if ( bIn  || g_bIsLeanException)
 return osl_Signal_ActCallNextHdl;
 
 ExceptionCategory nVCLException = ExceptionCategory::NONE;
@@ -302,6 +304,7 @@ bool InitVCL()
 pSVData->maGDIData.mpScreenFontCache= new ImplFontCache;
 pSVData->maGDIData.mpGrfConverter   = new GraphicConverter;
 
+g_bIsLeanException = getenv("LO_LEAN_EXCEPTION") ? true : false;
 // Set exception handler
 pExceptionHandler = osl_addSignalHandler(VCLExceptionSignal_impl, nullptr);
 
commit 02a5941c6ea747fa26dcbb27aa45ef9faac9c857
Author: Mark Hung <mark...@gmail.com>
Date:   Tue May 31 01:22:11 2016 +0800

tdf#93883 don't add space to empty paragraph when bullet is turned off.

Change-Id: Ia1bad61cb5585dae0501e8cd657fc84bfea60ab2
Reviewed-on: https://gerrit.libreoffice.org/25679
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Michael Stahl <mst...@redhat.com>
(cherry picked from commit 005ae2fa195bddd1073f3d883cf041d9ac80fad1)
Reviewed-on: https://gerrit.libreoffice.org/25984

diff --git a/oox/source/export/drawingml.cxx b/oox/source/export/drawingml.cxx
index 880351d..dfa46d7 100644
--- a/oox/source/export/drawingml.cxx
+++ b/oox/source/export/drawingml.cxx
@@ -1583,6 +1583,9 @@ void DrawingML::WriteRun( const Reference< XTextRange >& 
rRun )
 sal_Int16 nLevel = -1;
 GET( nLevel, NumberingLevel );
 
+bool bNumberingIsNumber = true;
+GET( bNumberingIsNumber, NumberingIsNumber );
+
 bool bIsURLField = false;
 OUString sFieldValue = GetFieldValue( rRun, bIsURLField );
 bool bWriteField  = !( sFieldValue.isEmpty() || bIsURLField );
@@ -1590,7 +1593,7 @@ void DrawingML::WriteRun( const Reference< XTextRange >& 
rRun )
 OUString sText = rRun->getString();
 
 //if there is no text following the bullet, add a space after the bullet
-if (nLevel !=-1 && sText.isEmpty() )
+if (nLevel !=-1 && bNumberingIsNumber && sText.isEmpty() )
  sText=" ";
 
 if ( bIsURLField )
diff --git a/sd/qa/unit/data/odp/tdf93883.odp b/sd/qa/unit/data/odp/tdf93883.odp
new

[Libreoffice-commits] core.git: solenv/gbuild

[Norbert Thiebaud]

 solenv/gbuild/platform/windows.mk |3 +++
 1 file changed, 3 insertions(+)

New commits:
commit 6b3b352b06d92ef20194b9a992a521af2ef07b48
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Thu Jun 2 14:02:30 2016 -0500

try to avoid 'by-design' solar-mutex deadlock during ci

Change-Id: I9e1d79613adf6184f76c2e07aca0b78a9329838f
Reviewed-on: https://gerrit.libreoffice.org/25839
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/solenv/gbuild/platform/windows.mk 
b/solenv/gbuild/platform/windows.mk
index a2e621d..ebd2a3c 100644
--- a/solenv/gbuild/platform/windows.mk
+++ b/solenv/gbuild/platform/windows.mk
@@ -8,6 +8,9 @@
 # file, You can obtain one at http://mozilla.org/MPL/2.0/.
 #
 
+# to block heavy exception handling that try to acquire the solarmutex
+export LO_LEAN_EXCEPTION=1
+
 # to avoid flashing windows during tests
 export VCL_HIDE_WINDOWS=1
 
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: Jenkins coverage (Android and clang plugins)

[Norbert Thiebaud]

On Mon, May 30, 2016 at 3:04 AM, Miklos Vajna <vmik...@collabora.co.uk> wrote:
> Hi Norbert,
>
> On Sun, May 29, 2016 at 02:32:52PM -0500, Norbert Thiebaud 
> <nthieb...@gmail.com> wrote:
>> I installed clang and llvm
>> clang comes as version 3.4.2 (tags/RELEASE_34/dot2-final) on centos 7
>>
>> saddly the plugings thing use api that are not in there
>> namely
>> FunctionDecl::params
>>
>> which lead to
>> build CXX] compilerplugins/clang/unusedfields.cxx
>> /home/tdf/lode/dev/core/compilerplugins/clang/unusedfields.cxx:127:58:
>> error: no member named 'params' in 'clang::FunctionDecl'
>> for (const ParmVarDecl *pParmVarDecl : functionDecl->params()) {
>>  ^
>>
>>
>> It is one thing to activate clang-plugin, it is another to maintains a
>> set of manual packages on all build slaves
>
> Thanks for checking this, indeed I haven't tried the plugins on clang
> <3.5.
>
> One more idea before abandoning this for now: the Ubsan build uses
> /home/tdf/lode/opt_private/bin/clang++ as the compiler, I guess that's a
> self-built, newer version -- could you please give that a try?

Ubsan build is a once a day build, tied to specific slave, that I
manually customized
furthermore it is not on the critical path, meaning that if it is red
that does not impact massively the workflow.

The clang plugin ideas is for gerrit build... that is a 40-80 builds a
day proposition
that need to run on all possible (linux) machines I can use as slave.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Jenkins coverage (Android and clang plugins)

[Norbert Thiebaud]

On Fri, May 20, 2016 at 8:03 AM, Miklos Vajna  wrote:
> Hi Norbert,
>
> - building with the clang plugins enabled (on Linux, the OS X one is not
>   capable for running the plugins; dbgutil, product would not bring much
>   benefit)

I installed clang and llvm
clang comes as version 3.4.2 (tags/RELEASE_34/dot2-final) on centos 7

saddly the plugings thing use api that are not in there
namely
FunctionDecl::params

which lead to
build CXX] compilerplugins/clang/unusedfields.cxx
/home/tdf/lode/dev/core/compilerplugins/clang/unusedfields.cxx:127:58:
error: no member named 'params' in 'clang::FunctionDecl'
for (const ParmVarDecl *pParmVarDecl : functionDecl->params()) {
     ^


It is one thing to activate clang-plugin, it is another to maintains a
set of manual packages on all build slaves

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: solenv/gbuild vcl/source

[Norbert Thiebaud]

 solenv/gbuild/platform/unxgcc.mk |3 +++
 vcl/source/app/svmain.cxx|5 -
 2 files changed, 7 insertions(+), 1 deletion(-)

New commits:
commit 9d2f7be4e65595241db3cf5135b69bd9e4ce6a30
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Thu May 19 13:35:13 2016 -0500

shortcut the signal handler during batch test

Change-Id: I7848555743e8d54e755dc82cbf0148eeed54110f
Reviewed-on: https://gerrit.libreoffice.org/25178
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
    Tested-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/solenv/gbuild/platform/unxgcc.mk b/solenv/gbuild/platform/unxgcc.mk
index 278006b..bc44906 100644
--- a/solenv/gbuild/platform/unxgcc.mk
+++ b/solenv/gbuild/platform/unxgcc.mk
@@ -17,6 +17,9 @@
 #   the License at http://www.apache.org/licenses/LICENSE-2.0 .
 #
 
+# to block heavy exception handling that try to acquire the solarmutex
+export LO_LEAN_EXCEPTION=1
+
 include $(GBUILDDIR)/platform/com_GCC_defs.mk
 
 gb_MKTEMP := mktemp -t gbuild.XX
diff --git a/vcl/source/app/svmain.cxx b/vcl/source/app/svmain.cxx
index 2b315b2..69ab53b 100644
--- a/vcl/source/app/svmain.cxx
+++ b/vcl/source/app/svmain.cxx
@@ -95,6 +95,8 @@
 
 using namespace ::com::sun::star;
 
+static bool g_bIsLeanException;
+
 static bool isInitVCL();
 
 oslSignalAction SAL_CALL VCLExceptionSignal_impl( void* /*pData*/, 
oslSignalInfo* pInfo)
@@ -102,7 +104,7 @@ oslSignalAction SAL_CALL VCLExceptionSignal_impl( void* 
/*pData*/, oslSignalInfo
 static volatile bool bIn = false;
 
 // if we crash again, bail out immediately
-if ( bIn )
+if ( bIn  || g_bIsLeanException)
 return osl_Signal_ActCallNextHdl;
 
 ExceptionCategory nVCLException = ExceptionCategory::NONE;
@@ -302,6 +304,7 @@ bool InitVCL()
 pSVData->maGDIData.mpScreenFontCache= new ImplFontCache;
 pSVData->maGDIData.mpGrfConverter   = new GraphicConverter;
 
+g_bIsLeanException = getenv("LO_LEAN_EXCEPTION") ? true : false;
 // Set exception handler
 pExceptionHandler = osl_addSignalHandler(VCLExceptionSignal_impl, nullptr);
 
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: multi-threading task under SolarMutex -> deadlock

[Norbert Thiebaud]

On Wed, May 18, 2016 at 3:43 AM, Armin Le Grand  wrote:
> Hi Norbert,
>
> thanks for also having an eye on this - I am looking for the failure reports
> on ci.libreoffice.org currently, too.
> Last is from http://ci.libreoffice.org/job/lo_tb_master_linux_dbg/7195/, so
> last is from Friday, 13th (uhhh...)
>
> Have you seen such or similar stacks anywhere else? In the meantime I tried
> ChartTest massively locally on Linux and Win, but could never locally
> reproduce.

I noticed, and I still had to cancel a hung job few minutes ago, that
things started to hang regularely.
I only look at that particular case I sent the backtrace about.

>
> I knew that and made sure that the multithreaded 3DRenderer WorkerThreads do
> not need the SolarMutex for their work. I did not know yet that the memory
> fail handler tries to get the SolarMutex, too, but is logic when it wants to
> bring up a dialog in some form.

yeah but that _is_ a major flaw. code under a signal handler are only
allow to call async-safe-signal functions.
Ignoring that _will_ cause trouble. it is not your doing. it just
happen that your multithreading case seems to cause memory starvation
which in turn trigger a signal that exhibit clearly why all the thing
we try to do in a signal handler are not a good idea.
Not to mention that, even if that was allowed, trying to bring about a
dialog when we have already ran out of memory is never going to end
well anyway.

for reference:

  Async-signal-safe functions
   A signal handling routine established by sigaction(2) or
signal(2) must be very careful, since processing elsewhere may be
interrupted at some arbitrary point in the execution of the program.
POSIX has the concept of "safe function".  If a signal interrupts the
execution of an unsafe function, and handler calls an  unsafe
function,  then
   the behavior of the program is undefined.

   POSIX.1-2004 (also known as POSIX.1-2001 Technical Corrigendum
2) requires an implementation to guarantee that the following
functions can be safely called inside a signal handler:

   _Exit()
   _exit()
   abort()
   accept()
   access()
   aio_error()
   aio_return()
   aio_suspend()
   alarm()
   bind()
   cfgetispeed()
   cfgetospeed()
   cfsetispeed()
   cfsetospeed()
   chdir()
   chmod()
   chown()
   clock_gettime()
   close()
   connect()
   creat()
   dup()
   dup2()
   execle()
   execve()
   fchmod()
   fchown()
   fcntl()
   fdatasync()
   fork()
   fpathconf()
   fstat()
   fsync()
   ftruncate()
   getegid()
   geteuid()
   getgid()
   getgroups()
   getpeername()
   getpgrp()
   getpid()
   getppid()
   getsockname()
   getsockopt()
   getuid()
   kill()
   link()
   listen()
   lseek()
   lstat()
   mkdir()
   mkfifo()
   open()
   pathconf()
   pause()
   pipe()
   poll()
   posix_trace_event()
   pselect()
   raise()
   read()
   readlink()
   recv()
   recvfrom()
   recvmsg()
   rename()
   rmdir()
   select()
   sem_post()
   send()
   sendmsg()
   sendto()
   setgid()
   setpgid()
   setsid()
   setsockopt()
   setuid()
   shutdown()
   sigaction()
   sigaddset()
   sigdelset()
   sigemptyset()
   sigfillset()
   sigismember()
   signal()
   sigpause()
   sigpending()
   sigprocmask()
   sigqueue()
   sigset()
   sigsuspend()
   sleep()
   sockatmark()
   socket()
   socketpair()
   stat()
   symlink()
   sysconf()
   tcdrain()
   tcflow()
   tcflush()
   tcgetattr()
   tcgetpgrp()
   tcsendbreak()
   tcsetattr()
   tcsetpgrp()
   time()
   timer_getoverrun()
   timer_gettime()
   timer_settime()
   times()
   umask()
   uname()
   unlink()
   utime()
   wait()
   waitpid()
   write()

   POSIX.1-2008 removes fpathconf(), pathconf(), and sysconf()
from the above list, and adds the following functions:

   execl()
   execv()
   faccessat()
   fchmodat()
   fchownat()
   fexecve()
   fstatat()
   futimens()
   linkat()
   mkdirat()
   mkfifoat()
   mknod()
   mknodat()
 

Re: multi-threading task under SolarMutex -> deadlock

[Norbert Thiebaud]

On Tue, May 17, 2016 at 6:44 AM, Thorsten Behrens <t...@libreoffice.org> wrote:
> Norbert Thiebaud wrote:
>> The threaded work then raise() due to some memory problem and out
>> signal handler try to acquire the solar mutex ->deadlock
>>
> Eek, that's ugly. Then again, at the core is the OOM condition, which
> needs solving independently. Per chance, is that happening on a box
> with massive amounts of CPU threads?

it is on the ci builder, so yeah 32 thread or so.

but I disagree that it is _at the core_

at the core this exhibit 2 things:
1/ we do a lot of thing that is verboten in a signal handler.
2/ taking a lock that rely on other thread to move forward while
holding the solarmutex is begging for deadlock.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

multi-threading task under SolarMutex -> deadlock

[Norbert Thiebaud]

Recently I have had linux dbgutil build hang on occasion

The issue is that recently drawinlayer is starting using threadpool
( 
https://cgit.freedesktop.org/libreoffice/core/commit/?id=657413b5deea11a850970f23cba2cf34a5bdf8ea
)
and is issuing a waitUntilEmpty() on a threadpool, while holding the
solar mutex...


The threaded work then raise() due to some memory problem
and out signal handler try to acquire the solar mutex ->deadlock

relevant backtrace:

#0  0x2af85e71c6d5 in pthread_cond_wait@@GLIBC_2.3.2 () at
/lib64/libpthread.so.0
#1  0x2af85d8cd744 in osl_waitCondition(oslCondition, TimeValue
const*) (Condition=0x35fff90, pTimeout=0x0) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/sal/osl/unx/conditn.cxx:228
#2  0x2af866bf12b6 in osl::Condition::wait(TimeValue const*)
(this=0x3669e78, pTimeout=0x0) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/include/osl/conditn.hxx:84
#3  0x2af866c556a1 in comphelper::ThreadPool::waitUntilEmpty()
(this=0x3669e60) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/comphelper/source/misc/threadpool.cxx:202
#4  0x2af878ef7a4d in
drawinglayer::primitive2d::ScenePrimitive2D::create2DDecomposition(drawinglayer::geometry::ViewInformation2D
const&) const (this=0x2af8961bacd0, rViewInformation=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/primitive2d/sceneprimitive2d.cxx:439

^^^ wait for threadpool

#5  0x2af878eae474 in
drawinglayer::primitive2d::BufferedDecompositionPrimitive2D::get2DDecomposition(drawinglayer::geometry::ViewInformation2D
const&) const (this=0x2af8961bacd0, rViewInformation=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/primitive2d/baseprimitive2d.cxx:99
#6  0x2af878ef9009 in
drawinglayer::primitive2d::ScenePrimitive2D::get2DDecomposition(drawinglayer::geometry::ViewInformation2D
const&) const (this=0x2af8961bacd0, rViewInformation=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/primitive2d/sceneprimitive2d.cxx:695
#7  0x2af878f63eb6 in
drawinglayer::processor2d::VclPixelProcessor2D::processBasePrimitive2D(drawinglayer::primitive2d::BasePrimitive2D
const&) (this=0x3b59700, rCandidate=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/vclpixelprocessor2d.cxx:1251
#8  0x2af878f4192c in
drawinglayer::processor2d::BaseProcessor2D::process(drawinglayer::primitive2d::Primitive2DContainer
const&) (this=0x3b59700, rSource=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/baseprocessor2d.cxx:63
#9  0x2af878f63ecf in
drawinglayer::processor2d::VclPixelProcessor2D::processBasePrimitive2D(drawinglayer::primitive2d::BasePrimitive2D
const&) (this=0x3b59700, rCandidate=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/vclpixelprocessor2d.cxx:1251
#10 0x2af878f4192c in
drawinglayer::processor2d::BaseProcessor2D::process(drawinglayer::primitive2d::Primitive2DContainer
const&) (this=0x3b59700, rSource=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/baseprocessor2d.cxx:63
#11 0x2af878f63ecf in
drawinglayer::processor2d::VclPixelProcessor2D::processBasePrimitive2D(drawinglayer::primitive2d::BasePrimitive2D
const&) (this=0x3b59700, rCandidate=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/vclpixelprocessor2d.cxx:1251
#12 0x2af878f4192c in
drawinglayer::processor2d::BaseProcessor2D::process(drawinglayer::primitive2d::Primitive2DContainer
const&) (this=0x3b59700, rSource=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/baseprocessor2d.cxx:63
#13 0x2af878f63ecf in
drawinglayer::processor2d::VclPixelProcessor2D::processBasePrimitive2D(drawinglayer::primitive2d::BasePrimitive2D
const&) (this=0x3b59700, rCandidate=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/vclpixelprocessor2d.cxx:1251
#14 0x2af878f4192c in
drawinglayer::processor2d::BaseProcessor2D::process(drawinglayer::primitive2d::Primitive2DContainer
const&) (this=0x3b59700, rSource=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/drawinglayer/source/processor2d/baseprocessor2d.cxx:63
#15 0x2af87e332e5a in paintUsingPrimitivesHelper(OutputDevice&,
drawinglayer::primitive2d::Primitive2DContainer const&,
basegfx::B2DRange const&, basegfx::B2DRange const&)
(rOutputDevice=..., rSequence=..., rSourceRange=..., rTargetRange=...)
at 
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/sw/source/core/doc/notxtfrm.cxx:744
#16 0x2af87e334068 in SwNoTextFrame::PaintPicture(OutputDevice*,
SwRect const&) const (this=0x3252b80, pOut=0x3d97410, rGrfArea=...) at
/home/tdf/lode/jenkins/workspace/lo_tb_master_linux_dbg/sw/source/core/doc/notxtfrm.cxx:1023
#17 0x2af87e330f21 in 

Re: Problem with multiple identities on Gerrit

[Norbert Thiebaud]

On Mon, May 9, 2016 at 1:43 AM, David Ostrovsky <d.ostrov...@gmx.de> wrote:
>
>> On Sat May 7 18:11:11 UTC 2016 Norbert Thiebaud wrote:
>
>>>On Fri, May 6, 2016 at 6:31 PM, Mike Kaganski  hotmail.com> wrote:
>>> Hello!
>>>
>>> Currently I have a gerrit account with two identities:
> [...]
>
>> I not sure if you really try to link (iow log
>>on with an existing identity -> setting -> add ana identity) and there
>>is a bug in OAuth + openid that recognized your old gerrit openid and
>>link Oauth to that.
>
> There is no way for him to link Google OpenID account, that he cannot
> use to log in, to Google OAuth2, obviously. That why the linking is
> working out of the box, so it's enough to just log in. It's funny you
> are talking about "a bug in OAuth + openid"

you are quote mining and dropped from the above the all important:

"or you did not and it worked as designed and
attached your google OAuth to your old expired goggle openid"

I offered 2 alternative:

Either he was logged on an existing account (not the one that had the
old google openid, but some other that was created using a different
identity)
and he tried to 'add a new identity'

Or he simply just tryied to log-in using google OAuth straight up

in the later case I expect indeed things to work the way you describe
(which is way I said: "or you did not and it worked as designed and
attached your google OAuth to your old expired goggle openid"

in the former case I do not expect that the new Oauth identity to be
attached to another account than the one he was logged in while trying
to 'add a new identity'

Since I could not be sure of which case it was I mentioned the 2 possibilities.
The only thing I'm sure of is the outcome I saw in gerrit database.


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Problem with multiple identities on Gerrit

[Norbert Thiebaud]

On Fri, May 6, 2016 at 6:31 PM, Mike Kaganski  wrote:
> Hello!
>
> Currently I have a gerrit account with two identities:
> mikekagan...@hotmail.com (using openid provider
> http://pip.verisignlabs.com/) and mike.kagan...@collabora.com.

actually you had 3

one created 2013-07-05 (gmail openid)
one created 2015-04-26 (http:verisign)
one created 2015-11-09 (https:verisign)


your google OAuth got attached to the original account from 2013

>
> Since linking the second identity some time ago,


Since the first verisign account was created 6 days _after_ google
shutdown its openid, there is no way you could have 'linked the second
identity'

so at that point you had 2 account google openid and verisign http:

Apparently gerrit's openid consider http and https variant has
separate identity. Arguably it is a bug that http: be allowed and even
if it is it would not be considered the same than https, in any case
again login with what gerrit consider another identity created a 3rd
one for you, 6 months ago.

later you tried AOuth. I not sure if you really try to link (iow log
on with an existing identity -> setting -> add ana identity) and there
is a bug in OAuth + openid that recognized your old gerrit openid and
link Oauth to that.. or you did not and it worked as designed and
attached your google OAuth to your old expired goggle openid
either way.. you ended up with 3 account http:verisign, https:verisign
and OAuth:google

>
> Could someone provide some guidance what steps should I follow to solve the
> two small issues I desccribed:
>
> 1. To join two existing accounts (mikekagan...@hotmail.com using
> http://pip.verisignlabs.com/ and https://pip.verisignlabs.com/) into one
> that is currently linked with mike.kagan...@collabora.com?
>
> 2. To link new google identity mikekagan...@gmail.com to that account using
> OAuth2? (it seems that the new identity is already linked to some *third*
> account.)

There is no steps for you to follow. this housekeeping require
tweaking gerrit underlying sql database.
I have done it for you... you now have one combined account.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: libre5.0 + win10 configure error

[Norbert Thiebaud]

On Tue, May 3, 2016 at 4:09 AM, jemmy  wrote:
> Hello, everyone!
> I'm a new developer of libreoffice. I have followed the instructions from
> https://wiki.documentfoundation.org/Development/BuildingOnWindows and used
> the recommended settings and paths. I use Win10 x64 system 、 64-bit
> Cygwin、32-bit Java JDK、VS2013_RTM_PRO.
>
> ./autogen.sh --with-external-tar=/cygdrive/d/sources/lo-externalsrc
> --with-junit=/cygdrive/d/sources/junit-4.10.jar
> --with-ant-home=/cygdrive/d/sources/apache-ant-1.9.5 --enable-pch
> --disable-ccache --enable-debug --disable-atl
>
> When I was trying to configure LibreOffice,  autogen.sh was not able to
> finish its job, with message below:
>
> ..
> checking for install dirname... libreoffice
> checking for prefix... /usr/local
> checking for Google Drive client id and secret... set
> checking for Alfresco Cloud client id and secret... set
> checking for OneDrive client id and secret... set
> checking build verbosity... not set
> checking whether to enable dependency tracking... yes
> checking for number of processors to use... 4
> checking whether C:/vs_net/VC/bin/cl.exe -arch:SSE has broken static
> initializer_list support... no
> checking whether to create huge library... no
> checking whether to create hardlinks for delivering files... no
> checking whether to use icerun wrapper... no
> checking MPL subset... no restrictions
> setting up the build environment variables...
> configure: creating ./config.status
> .in'ig.status: error: cannot find input file: `config_host.mk
> Error running configure at ./autogen.sh line 269.
>
> I'm thinking this problem all days, and can't resolve it. Is there anyone
> can help me ?

what is your core.git repo path ?
if it is directly at the root of a device like D:/core.. can you try
to move it one directory down
like D:/lo/core and see if that change something


do you see a config.log there ? if so look into it.. do you see place
where it talk about config_host
is there any error reported there ?
maybe some acl issue ?

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: LibreOffice Hackfest Ankara, Turkey 2016

[Norbert Thiebaud]

On Tue, May 3, 2016 at 5:15 PM, Gülşah Köse  wrote:
>
> Hello,
>
> LibreOffice Hackfest Ankara was held for the first time in Turkey between on 
> 29th April and 1st May. I've published a blog post[1] about how was it. 
> Thanks to TDF, Michael Meeks, Markus Mohrhard and Jan-Marek Glogowski for 
> spearing their times for us.
>
> Hope to meet you all one day.

Congrats Gülşah for a nice and successful hackfest. It is a lot of
work to make these things happen, thank you for your dedication and
efforts.

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice-commits] core.git: external/glew

[Norbert Thiebaud]

 external/glew/UnpackedTarball_glew.mk |4 
 external/glew/glew-tmpdir.patch.1 |   13 +
 2 files changed, 17 insertions(+)

New commits:
commit 7aa67d528b9f543fb61527cae7f0f487426cb414
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Sun May 1 23:51:47 2016 -0500

glew: honor TMPDIR

Change-Id: Id5c20ec78dcde0144bd7718b7e9825b92ed70f3f
Reviewed-on: https://gerrit.libreoffice.org/24582
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>
    Tested-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/external/glew/UnpackedTarball_glew.mk 
b/external/glew/UnpackedTarball_glew.mk
index 3b1a5a1..c47b7a0 100644
--- a/external/glew/UnpackedTarball_glew.mk
+++ b/external/glew/UnpackedTarball_glew.mk
@@ -41,4 +41,8 @@ $(eval $(call gb_UnpackedTarball_add_patches,glew,\
 ))
 endif
 
+$(eval $(call gb_UnpackedTarball_add_patches,glew,\
+   external/glew/glew-tmpdir.patch.1 \
+))
+
 # vim: set noet sw=4 ts=4:
diff --git a/external/glew/glew-tmpdir.patch.1 
b/external/glew/glew-tmpdir.patch.1
new file mode 100644
index 000..9a2847b
--- /dev/null
+++ b/external/glew/glew-tmpdir.patch.1
@@ -0,0 +1,13 @@
+diff -ur glew.org/Makefile glew/Makefile
+--- glew.org/Makefile  2016-05-02 06:25:53.521291677 +0200
 glew/Makefile  2016-05-02 06:31:10.325035494 +0200
+@@ -55,7 +55,8 @@
+ DIST_SRC_TGZ ?= $(shell pwd)/$(DIST_NAME).tgz
+ DIST_WIN32   ?= $(shell pwd)/$(DIST_NAME)-win32.zip
+ 
+-DIST_DIR := $(shell mktemp -d /tmp/glew.XX)/$(DIST_NAME)
++TMPDIR ?= /tmp
++DIST_DIR := $(shell mktemp -d $(TMPDIR)/glew.XX)/$(DIST_NAME)
+ 
+ # To disable stripping of binaries either:
+ #   - use STRIP= on gmake command-line
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

[Libreoffice-commits] core.git: solenv/bin

[Norbert Thiebaud]

 solenv/bin/macosx-codesign-app-bundle |2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

New commits:
commit 5892c6b2268b5b2290faca9cbdb2807f6a810d96
Author: Norbert Thiebaud <nthieb...@gmail.com>
Date:   Fri Apr 22 12:26:03 2016 -0500

gengal is not s script anymore so it is already signed on mac

69858a99c9b566303551a1bff91172917c5d624f removed the script layer
around a gengal.bin
so there is no gengal 'script' anymore and gengal is a binary
that was signed at link already.

remove the exception that force a sign of gengal as a 'script'
which caused an error due to re-signing.

Change-Id: Ie22efceb8e5ab20018be784968487917c6f0a95e
Reviewed-on: https://gerrit.libreoffice.org/24306
Tested-by: Jenkins <c...@libreoffice.org>
Reviewed-by: Norbert Thiebaud <nthieb...@gmail.com>

diff --git a/solenv/bin/macosx-codesign-app-bundle 
b/solenv/bin/macosx-codesign-app-bundle
index ff6397a..afd3e68 100755
--- a/solenv/bin/macosx-codesign-app-bundle
+++ b/solenv/bin/macosx-codesign-app-bundle
@@ -60,7 +60,7 @@ done
 # Code Signing In Depth" suggests we should get rid of them rather sooner than
 # later, but they appear to be OK for now):
 
-for i in gengal python senddoc unoinfo
+for i in python senddoc unoinfo
 do
 if [ -f "$APP_BUNDLE/Contents/MacOS/$i" ]
 then
___
Libreoffice-commits mailing list
libreoffice-comm...@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice-commits

Re: Broken master and newcomers.

[Norbert Thiebaud]

On Thu, Apr 21, 2016 at 6:14 AM, jan iversen
 wrote:
> Sorry for double mail, to this theme, but speaking with a good friend of mine 
> (who runs a very big ci system), I got an idea, which might solve the problem 
> for good.
>
> Whenever jenkins builds a gerrit job, jenkins master knows when all platforms 
> for that patch is compiled

Nope.

as I explained earlier.. tb job are per platform/config and are not
linked... so there is no garanteed that a given commit will get a tb
build on every config

one reason among other is the huge disparity in build time... so the
setup favor quick turn around per platform/config and maximal coverage
rather than big blocking check-all on the same commit setup

But ultimately the solution is to concentrate on gerrit build
verificatio, ot add the same coverage than tb there.. and leave the tb
as they are..
that way if something pass gerrit build verification it _will_ pass tb
so always green tb become a 'always gerrit workflow'


Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Building LODE on Mac fails with "Too many open files"

[Norbert Thiebaud]

On Thu, Apr 21, 2016 at 3:11 AM, Ralf S.  wrote:
> Hi there,
>
> I'm trying to set up LODE on the latest Mac OS X with newest Xcode, following
> the instructions here ("Quick Pre-canned Setup"):
> https://wiki.documentfoundation.org/Development/BuildingOnMac
>
> After running "make", the build process fails after a couple of hours with the
> following error:
>
> [build DEP] SRS:avmedia/res
> [build DEP] SRS:acc/res
> [...]/lode/dev/core/external//apache-commons/Module_apache-commons.mk:12:
> [...]/lode/dev/core/external/apache-
> commons/ExternalPackage_apache_commons_logging.mk: Too many open files
> [...]/lode/dev/core/external//apache-commons/Module_apache-commons.mk:12: ***
> No target registered while reading [...]/lode/dev/core/external/apache-
> commons/ExternalPackage_apache_commons_logging.mk!.  Stop.
> make: *** [build] Error 2

There is something fundamentally wrong if after a couple of hours you
are still in the [build Dep]
section.

iow the too many files is the symptom of something else going wrong
and increasing the file limit won't help

note that the mac build bot run with a vanilla config wrt to open files...

tb66:~ tdf$ launchctl limit maxfiles
maxfiles256unlimited

tb66:~ tdf$ ulimit -a
core file size  (blocks, -c) 0
data seg size   (kbytes, -d) unlimited
file size   (blocks, -f) unlimited
max locked memory   (kbytes, -l) unlimited
max memory size (kbytes, -m) unlimited
open files  (-n) 256
pipe size(512 bytes, -p) 1
stack size  (kbytes, -s) 8192
cpu time   (seconds, -t) unlimited
max user processes  (-u) 709
virtual memory  (kbytes, -v) unlimited


how do you start the build.. did you add $LODE_HOME/opt/bin in the
path (iow do you use the gnu-make setup by lode or accidentally use
Apple's provided one ?

also check you TEMP directory... sometimes there a serious leak of tempfiles...

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Building with Clang trunk on Mac OS X

[Norbert Thiebaud]

On Wed, Apr 20, 2016 at 3:09 AM, Stephan Bergmann  wrote:
> * Also, for reasons unclear to me, "make check" often fails when during
> JunitTests soffice crashes in what smells like threading violations in using
> the system's GUI APIs.  Again, other people seem unaffected by this, and I'm
> not sure what part of my setup is the root cause for this.

10.11 maybe related.
It is already known to cause quite a few problem with the
'DYLD_LIBRARY_PATH' 'sanitizing' maybe there are other 'security
feature'

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Broken master and newcomers.

[Norbert Thiebaud]

On Wed, Apr 6, 2016 at 6:14 PM, Anthonys Lists <antli...@youngman.org.uk> wrote:
> On 07/04/2016 00:08, Norbert Thiebaud wrote:
>>>
>>> Imho, IF we do something like this, we should use the git facilities to
>>> >create a shallow clone backup, that people can then use ftp or rsync or
>>> > some
>>> >other interruptible protocol to download.
>>
>> We have been doing that for years:
>>
>> http://dev-www.libreoffice.org/bundles/
>>
> That should then make this whole discussion moot then? Just modify the
> instructtions to warn newcomers that master *can* be broken, and if they
> want a known-good build to start with, go and get a bundle rather than using
> clone. Once thye've got a good build they can "git pull" and go from there.

the bundle is not guarantee to build anymore than a random master...

It is just that we regularly (in fact is a cron iirc) tar-gz core.git
and other repo, precisely for people that have trouble doing a git
clone on a big repo

Norbert
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Broken master and newcomers.

[Norbert Thiebaud]

On Wed, Apr 6, 2016 at 5:58 PM, Anthonys Lists  wrote:
> On 06/04/2016 08:21, Bjoern Michaelsen wrote:
>>>
>>> An alternative (which I have seen used in another project), would be to
>>> let
>>> >jenkins generate a source tar ball, incl. the .git directory, of the
>>> > latest
>>> >sane build.
>>
>> No, thats horrible -- it make onboarding to git even harder.
>>
> What quite do you mean? If the tarball includes the .git directory, then
> it's effectively a clone, so they get a "known good" clone.
>
> That said, I'd rather not do that ...
>
> If we're going down that sort of route (and I had this discussion on the
> firebird list) then remember that "git clone" does not work well over a
> flaky internet connection, and if it bombs you typically have to restart
> from scratch. And flaky connections tend to be pay connections, making it an
> expensive business :-(
>
> Imho, IF we do something like this, we should use the git facilities to
> create a shallow clone backup, that people can then use ftp or rsync or some
> other interruptible protocol to download.


We have been doing that for years:

http://dev-www.libreoffice.org/bundles/
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Broken master and newcomers.

[Norbert Thiebaud]

On Wed, Apr 6, 2016 at 1:04 AM, jan iversen  wrote:
> Hi.
>
> I experience quite frequently that new contributors follow the instructions, 
> but end up with a master that will not compile,

these day that is unlucky:


here are the stats for the last 14 days:

from:Wed Mar 23 21:30:24 2016
master linux rel  jobs: 296 ok: 282 ko:  11 fail ratio:  3.72 %
break:   9 broken duration: 2.12%
master linux dbg  jobs: 242 ok: 226 ko:  15 fail ratio:  6.20 %
break:  12 broken duration: 2.98%
master mac reljobs: 281 ok: 274 ko:   7 fail ratio:  2.49 %
break:   4 broken duration: 0.86%
master mac dbgjobs: 285 ok: 282 ko:   3 fail ratio:  1.05 %
break:   2 broken duration: 1.95%
master win reljobs: 222 ok: 212 ko:  10 fail ratio:  4.50 %
break:   4 broken duration: 1.00%
master win dbgjobs: 229 ok: 215 ko:  14 fail ratio:  6.11 %
break:   7 broken duration: 2.64%
master win64 dbg  jobs: 232 ok: 216 ko:  15 fail ratio:  6.47 %
break:   7 broken duration: 3.19%

broken duration is giving you the following info:
if you randomly pull master at any time in the period considered: what
was the odd that it was broken at that time
during the last 14 days your odd varied from 1 in a 115 for mac
releases build to 1 in 30 for windows64 dbg

let's look at windows dbg:

the last break right this minute on windows dbg was 1 day 9 hours ago
http://ci.libreoffice.org/job/lo_tb_master_win_dbg/5935/
"Device or resource busy"
either a parallelism problem in our build system or a bug in Windows.
master was not really red because of that.

the previous break was
http://ci.libreoffice.org/job/lo_tb_master_win_dbg/5915/
2 days 5 hours ago.. direct push by sberg

the one before was 2 hour earlier
http://ci.libreoffice.org/job/lo_tb_master_win_dbg/5913/

that one seems attributable to Windows crappy FS.

the one before that was
http://ci.libreoffice.org/job/lo_tb_master_win_dbg/5878/
5 days 13 hours ago, this one failed on win-dbg tb but passed Jenkins.


the one before that was 6 days 11 hours ago
http://ci.libreoffice.org/job/lo_tb_master_win_dbg/5851/
that was due to 41bdaa37cc62f656cc164992c4c7d39bec7e57e2 a direct push by noel


the one before that was 7 days 16 hr ago
and was not due to the commit.. but some jenkins <-> windows interaction problem
Note that these kind of 'break' would not actually impact someone
pulling master.
iow it is red in ci but not really red

so of the win dbg break... 3 where due to windows glitches. 1 was
properly vetted by Jenkins but still failed in that tinderbox
2 could have been avoided by using gerrit.

but in the end that is 3 real break in 2 weeks for a real red duration
of 2 hours 38 minutes that is 1 chance in 127 to get a random red
state.


so :"I experience quite frequently " seems to be a 'count the
his/ignore the misses' kind of thing.

IOW: my advice... let's keep nudging people to avail themselves of the
CI facility and avoid direct push...
that would have change the odd of a random red pull from 1 to 127 to 1
in 305 in this particular case.
but let's not forget that not that long ago it was easier to count how
often windows was buildable rather than hunt the commit where it was
not.

I also have another data point: the bibisect build: these are less
sensitive since I purposefully build them without Werror and I
purposfully do not run the checks
(the goal being to have binaries.. not to validate the build quality)


here are the result: this are the commit in bibisect that cover more
than 1 source commit. the number at the end of each line is the number
of unbuildable commit that are folded in that one bibisect commit

1dee09a4aa086524fd7704ad3dd6dc7e9ba201f4:1
1ebeb8fb59bfff5b9bdb3ac6db158c7c13e85e70:1
2a019e2b222ad49446c730bdbdea6c6651e36d91:1
3229e0f25f5d55d566208e42b276339a9c3c63c5:1
4c4af51cf6105e635feb206ea193f150f05e0edf:1
6d6a571cd180ee828ec8c8a5bd21a46d571d47ee:1
81168aee40fd18d7fcc475406f4c58ff7ba63455:1
8a70721672f5a1bce10547849435d7a0e286fb45:1
a54383ef92868298a368bcec871e5f2f2b2343b6:1
a9cca36433706038da0a51d6311da1bb77eb6112:1
e5c3bed99f2508b10833ccada394ce1abc37bf07:1
e72fd72eaea986ba41b8afff31ae9b77177fc685:1
f3ed988856fe00c5826781d97aaf2cae61f1abb7:1
4498e0cb3277ce65738acbc7c1faac43e9066430:2
6d46b870384fbfce0293a8fce41e373087ff681f:2
7b7852a9877e1123c60d5872ecf8ff0d78da45d7:2
ac5a57d70ad872f2dfe9d1bff73baa5318042e0e:2
b89bf27413eaf5cfab9a501a297756b8ef22eaa2:2
c6f1dc4f2407945334adef66c989f3878abc7043:2
cc822d9c6af1e8d6e937e23145471daf52ebd472:2
f73273c14261004eae364d06509b029c3a3883f2:2
45efc63eb004ebd2643cea9ebcdd810d0935bfa6:3
5de2c9cdc2e246e98ab43dde0baf14ca0947dd47:3
90fbb1101724c057d3c269f3b05f5584f701bb2f:3
b1d96f0e8acd28c6536896ed9aa70a88daa3f8fc:3
ba10b24c5f679cfcace557a920c83264b374e054:4
cebf63930f184875b7a04b3c95ad182554df5aa6:4
1d25fbbb38f0973147eda92bde66872b7cc974c0:5
1cb794851d2cbb56563a82b763da6c8a696dd465:6
326d9741e8236d3912ca55b0e9a7d4cee7b5a882:6

Re: linux dbgutil tinderbox stuck -> backtrace

[Norbert Thiebaud]

On Fri, Apr 1, 2016 at 8:23 AM, Stephan Bergmann  wrote:
>
>
> I don't claim Jenkins is a good infrastructure for this.  (But I don't want
> to start any discussion about that.  Just wanted to point out that adding
> timeouts to test code is the wrong solution to the underlying problem of
> having reliable build bots.)
not 'to the test code'.. to wrap around the execution of tests.
I fail to see what is wrong with that.

>
>> 2/ the linux debug build, once a week also rebuild the doc.. which
>> takes a long time.. so I had to bump that global level deadlock to the
>> max time a full build + build the doc and upload it can take... which
>> make the deadlock kick in the 5-6 hours range... not great.
>
>
> A single, fixed timeout, independent of the kind of build you're doing,
> indeed doesn't look very useful.

notwithstanding the generate doc case... if you have a trick to figure
out in advance how long an incremental build will take.. please let me
know :-)
the conservative tiemout is there to limit the damage.. but it is
indeed suboptimal. thankfully we do not have _that_ many hangs
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice