Re: Compiling LO 6.2.0.3 get undefined reference to `hb_graphite2_face_get_gr_face'

2019-02-26 Thread Alex Kempshall 

On 25/02/2019 14:42, Michael Stahl wrote:

On 25.02.19 15:32, Alex Kempshall wrote:

On 25/02/2019 14:17, Michael Stahl wrote:

On 25.02.19 14:52, Alex Kempshall wrote:

On 25/02/2019 12:28, Michael Stahl wrote:

On 25.02.19 12:10, Alex Kempshall wrote:

The error I get is

/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collectGraphite()':
FeatureCollector.cxx:(.text+0x34): undefined reference to 
`hb_graphite2_face_get_gr_face'


S=/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3 && I=$S/instdir && 
W=$S/workdir && /usr/bin/ccache g++ -shared -Wl,-z,noexecstack 
-Wl,-z,origin '-Wl,-rpath,$ORIGIN' -Wl,-rpath-link,$I/program 
-Wl,-z,defs -fstack-protector-strong -Wl,-rpath-link,/lib:/usr/lib 
-Wl,-z,combreloc  -Wl,--hash-style=gnu -Wl,-Bsymbolic-functions 
-L$W/LinkTarget/StaticLibrary -L$I/sdk/lib  -L$I/program 
-L$I/program -Wl,-rpath,/usr/lib64/libreoffice/program/ -L/usr/lib64 


^ this -L/usr/lib64 is your problem; it is pointless (because that is 
in the default search path anyway) and harmful (because it overrides 
the -L$W/UnpackedTarball/harfbuzz/src/.libs that comes later:


$W/CxxObject/vcl/opengl/x11/X11DeviceInfo.o -Wl,--start-group 
$W/LinkTarget/StaticLibrary/libgraphite.a 
-L$W/UnpackedTarball/libjpeg-turbo/.libs -ljpeg 
$W/UnpackedTarball/libeot/.libs/libeot.a  -lgio-2.0 -lgobject-2.0 
-lglib-2.0 -L$W/UnpackedTarball/harfbuzz/src/.libs -lharfbuzz 
-lharfbuzz-icu -licuuc -licuuc -L$W/UnpackedTarball/lcms2/src/.libs 
-llcms2 -lcairo -lcups -ldbus-1  -lfontconfig -lfreetype -lfreetype 
-lm -ldl -lpthread -lX11 -lXext  -Wl,--end-group -Wl,--no-as-needed 
-lpdfiumlo -lsvllo -ltllo -lutllo -lsotlo -lucbhelper -lbasegfxlo 
-lcomphelper -luno_cppuhelpergcc3 -li18nlangtag -li18nutil 
-lopencllo -luno_cppu -luno_sal -luno_salhelpergcc3 -lxmlreaderlo 
-ljvmaccesslo -lepoxy  -o $I/program/libvcllo.so


now the question is where does that come from, try:
 "grep /lib64[^/] config_host.mk"


grep /lib64[^/] config_host.mk

returns

export LDFLAGS= -Wl,-rpath,/usr/lib64/libreoffice/program/ -L/usr/lib64 


I elected to remove the -L/usr/lib64 from config_host.mk and run the 
make again. This time the make completed successfully.


Then had a detailed look at my script. Bit of trial and error. Then 
changed the 64-bit section of the code below  to remove the contents 
from the SLKLDFLAGS and LIBDIRSUFFIX  as below



case "$ARCH" in
  i?86)  SLKCFLAGS="-march=${ARCH} -mtune=i686"
 SLKLDFLAGS=""; LIBDIRSUFFIX=""
 ;;
  x86_64)    SLKCFLAGS="-fPIC"
# SLKLDFLAGS="-L/usr/lib64"; LIBDIRSUFFIX="64"
 SLKLDFLAGS=""; LIBDIRSUFFIX=""
 ;;
  armv7hl)   SLKCFLAGS="-O2 -march=armv7-a -mfpu=vfpv3-d16 -std=c++98"
 SLKLDFLAGS=""; LIBDIRSUFFIX=""
 ;;
  *) SLKCFLAGS=""
 SLKLDFLAGS=""; LIBDIRSUFFIX=""
 ;;
esac


I can now compile successfully.

Michael thanks for your help and suggestions. Learnt a lot.

Alex





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

Re: Compiling LO 6.2.0.3 get undefined reference to `hb_graphite2_face_get_gr_face'

2019-02-25 Thread Alex Kempshall 

On 25/02/2019 14:17, Michael Stahl wrote:

On 25.02.19 14:52, Alex Kempshall wrote:

On 25/02/2019 12:28, Michael Stahl wrote:

On 25.02.19 12:10, Alex Kempshall wrote:

The error I get is

/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collectGraphite()':
FeatureCollector.cxx:(.text+0x34): undefined reference to 
`hb_graphite2_face_get_gr_face'


can you try that again with "make PARALLELISM=1 verbose=" to see the 
command line...



Hi Michael

Not sure if I understood correctly. This is what I did with the 
results obtained


make PARALLELISM=1 verbose=


yeah sorry i made a typo, try "make PARALLELISM=1 verbose=t"



Didn't know whether you wanted this is a more friendly way


make PARALLELISM=1 verbose=t



Automatic fetching of external tarballs is disabled.
mkdir -p /tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/solenv/bin/install-gdb-printers 
-a /tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir -c
make -j 1  -r -f 
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/Makefile.gbuild all
make[1]: Entering directory 
'/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3'

[DEP] LNK:Library/libvcllo.so
[LNK] Library/libvcllo.so
S=/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3 && I=$S/instdir && 
W=$S/workdir &&  mkdir -p $W/Dep/LinkTarget/Library/ && 
RESPONSEFILE=/tmp/gbuild.1Ym5Di && 
LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program" 
$W/LinkTarget/Executable/concat-deps ${RESPONSEFILE} > 
$W/Dep/LinkTarget/Library/libvcllo.so.d.tmp && rm -f ${RESPONSEFILE}
mv 
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/Dep/LinkTarget/Library/libvcllo.so.d.tmp 
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/Dep/LinkTarget/Library/libvcllo.so.d
S=/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3 && I=$S/instdir && 
W=$S/workdir &&  /usr/bin/ccache g++ -shared -Wl,-z,noexecstack   
-Wl,-z,origin '-Wl,-rpath,$ORIGIN' -Wl,-rpath-link,$I/program 
-Wl,-z,defs -fstack-protector-strong -Wl,-rpath-link,/lib:/usr/lib 
-Wl,-z,combreloc  -Wl,--hash-style=gnu  -Wl,-Bsymbolic-functions 
-L$W/LinkTarget/StaticLibrary -L$I/sdk/lib  -L$I/program -L$I/program 
-Wl,-rpath,/usr/lib64/libreoffice/program/ -L/usr/lib64   
$W/CObject/vcl/source/filter/jpeg/transupp.o 
$W/CxxObject/vcl/source/window/errinf.o 
$W/CxxObject/vcl/source/window/settings.o 
$W/CxxObject/vcl/source/window/paint.o 
$W/CxxObject/vcl/source/window/abstdlg.o 
$W/CxxObject/vcl/source/window/accel.o 
$W/CxxObject/vcl/source/window/accmgr.o 
$W/CxxObject/vcl/source/window/brdwin.o 
$W/CxxObject/vcl/source/window/accessibility.o 
$W/CxxObject/vcl/source/window/legacyaccessibility.o 
$W/CxxObject/vcl/source/window/clipping.o 
$W/CxxObject/vcl/source/window/stacking.o 
$W/CxxObject/vcl/source/window/debug.o 
$W/CxxObject/vcl/source/window/globalization.o 
$W/CxxObject/vcl/source/window/builder.o 
$W/CxxObject/vcl/source/window/commandevent.o 
$W/CxxObject/vcl/source/window/cursor.o 
$W/CxxObject/vcl/source/window/debugevent.o 
$W/CxxObject/vcl/source/window/decoview.o 
$W/CxxObject/vcl/source/window/dialog.o 
$W/CxxObject/vcl/source/window/dlgctrl.o 
$W/CxxObject/vcl/source/window/dndeventdispatcher.o 
$W/CxxObject/vcl/source/window/dndlistenercontainer.o 
$W/CxxObject/vcl/source/window/dockingarea.o 
$W/CxxObject/vcl/source/window/dockmgr.o 
$W/CxxObject/vcl/source/window/dockwin.o 
$W/CxxObject/vcl/source/window/event.o 
$W/CxxObject/vcl/source/window/floatwin.o 
$W/CxxObject/vcl/source/window/introwin.o 
$W/CxxObject/vcl/source/window/keycod.o 
$W/CxxObject/vcl/source/window/keyevent.o 
$W/CxxObject/vcl/source/window/layout.o 
$W/CxxObject/vcl/source/window/menu.o 
$W/CxxObject/vcl/source/window/menubarwindow.o 
$W/CxxObject/vcl/source/window/menufloatingwindow.o 
$W/CxxObject/vcl/source/window/menuitemlist.o 
$W/CxxObject/vcl/source/window/menuwindow.o 
$W/CxxObject/vcl/source/window/mnemonic.o 
$W/CxxObject/vcl/source/window/mnemonicengine.o 
$W/CxxObject/vcl/source/window/mouse.o 
$W/CxxObject/vcl/source/window/msgbox.o 
$W/CxxObject/vcl/source/window/popupmenuwindow.o 
$W/CxxObject/vcl/source/window/printdlg.o 
$W/CxxObject/vcl/source/window/scrwnd.o 
$W/CxxObject/vcl/source/window/seleng.o 
$W/CxxObject/vcl/source/window/split.o 
$W/CxxObject/vcl/source/window/splitwin.o 
$W/CxxObject/vcl/source/window/status.o 
$W/CxxObject/vcl/source/window/syschild.o 
$W/CxxObject/vcl/source/window/syswin.o 
$W/CxxObject/vcl/source/window/tabdlg.o 
$W/CxxObject/vcl/source/window/tabpage.o 
$W/CxxObject/vcl/source/window/taskpanelist.o 
$W/CxxObject/vcl/source/window/toolbox2.o 
$W/CxxObject/vcl/source/window/toolbox.o 
$W/CxxObject/vcl/source/window/window2.o 
$W/CxxObject/vcl/source/window/window3.o 
$W/CxxObject/vcl/source/window/window.o 
$W/CxxObject/vcl/source/window/winproc.o

Re: Compiling LO 6.2.0.3 get undefined reference to `hb_graphite2_face_get_gr_face'

2019-02-25 Thread Alex Kempshall 

On 25/02/2019 12:28, Michael Stahl wrote:

On 25.02.19 12:10, Alex Kempshall wrote:

The error I get is

/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collectGraphite()':
FeatureCollector.cxx:(.text+0x34): undefined reference to 
`hb_graphite2_face_get_gr_face'


can you try that again with "make PARALLELISM=1 verbose=" to see the 
command line...



Hi Michael

Not sure if I understood correctly. This is what I did with the results 
obtained


make PARALLELISM=1 verbose=


Automatic fetching of external tarballs is disabled.
mkdir -p /tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/solenv/bin/install-gdb-printers 
-a /tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir -c
make -j 1  -rs -f 
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/Makefile.gbuild all

[DEP] LNK:Library/libvcllo.so
[LNK] Library/libvcllo.so
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collectGraphite()':
FeatureCollector.cxx:(.text+0x34): undefined reference to 
`hb_graphite2_face_get_gr_face'
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collect()':
FeatureCollector.cxx:(.text+0xadb): undefined reference to 
`hb_graphite2_face_get_gr_face'

collect2: error: ld returned 1 exit status
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/vcl/Library_vcl.mk:20: 
recipe for target 
'/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir/program/libvcllo.so' 
failed
make[1]: *** 
[/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir/program/libvcllo.so] 
Error 1

Makefile:286: recipe for target 'build' failed
make: *** [build] Error 2


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

Compiling LO 6.2.0.3 get undefined reference to `hb_graphite2_face_get_gr_face'

2019-02-25 Thread Alex Kempshall 

I'm  having problems compiling 6.2.0.3 from the downloads

libreoffice-help-6.2.0.3.tar.xz
libreoffice-dictionaries-6.2.0.3.tar.xz
libreoffice-6.2.0.3.tar.xz
libreoffice-translations-6.2.0.3.tar.xz

The error I get is


[build CXX] svx/source/accessibility/ChildrenManager.cxx
[build CXX] svx/source/accessibility/ChildrenManagerImpl.cxx
[build CXX] svx/source/accessibility/DescriptionGenerator.cxx
[build CXX] svx/source/accessibility/GraphCtlAccessibleContext.cxx
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collectGraphite()':
FeatureCollector.cxx:(.text+0x34): undefined reference to 
`hb_graphite2_face_get_gr_face'
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/workdir/CxxObject/vcl/source/font/FeatureCollector.o: 
In function `vcl::font::FeatureCollector::collect()':
FeatureCollector.cxx:(.text+0xadb): undefined reference to 
`hb_graphite2_face_get_gr_face'

collect2: error: ld returned 1 exit status
/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/vcl/Library_vcl.mk:20: 
recipe for target 
'/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir/program/libvcllo.so' 
failed
make[1]: *** 
[/tmp/build/tmp-libreoffice/libreoffice-6.2.0.3/instdir/program/libvcllo.so] 
Error 1

make[1]: *** Waiting for unfinished jobs
Makefile:286: recipe for target 'build' failed
make: *** [build] Error 2


I've come across this problem before and it went away somehow, not even 
sure if it was with LO. Inspecting my system I'd downloaded 
harfbuzz-1.8.8.tar.bz2 back in August 2018 and was comparing this with 
harfbuzz-1.8.4.tar.bz2. Can't remember what conclusion I came to! I 
firmly believe I don't use harfbuzz-1.8.8.tar.bz2 anywhere.


On my system I have harfbuzz-1.2.7-x86_64-1. However, I'm using the 
flags --with-system-icu and --without-system-harfbuzz which to me means 
I should be using the package 
https://dev-www.libreoffice.org/src/harfbuzz-1.8.4.tar.bz2.


What am I doing wrong or what point have I missed?

Alex

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

Re: [Libreoffice-qa] Using bibisect

2018-11-12 Thread Alex Kempshall 

Ilmari

Thanks for the reply where you wrote

Branches are not used. Tagged commits are used, but usually only 
"oldest" and "latest" are tagged. 

So this is maybe why I did this the last time I did a bibisect


git checkout f69ad1e0f0cfcfc6488e3739c4ef7efcc1445b72
git bisect start
git bisect good
git bisect bad master
./opt/program/soffice --base /home/alex/Documents/aprog_mysql.odb
git bisect bad


and so on

I shall have a look at the bibisect documentation that you suggested. 
I'm sure I'll wake up and remember what I did the last time.


Thanks again

Alex

On 12/11/2018 17:16, Ilmari Lauhakangas wrote:

Alex Kempshall kirjoitti 12.11.2018 klo 18.54:

I'm sure I've done this before, but for the moment my brain's gone dead!

...


If do

$git status

I get


On branch master
Your branch is up-to-date with 'origin/master'.

Untracked files:
  (use "git add ..." to include in what will be committed)

    instdir/user/

nothing added to commit but untracked files present (use "git add" 
to track)


At the moment I'm not worried about the untracked files.


instdir/user is the user profile, don't worry about it.

For bisect, on source code, I would normally do a git branch -a to 
see what I've got and then start bisecting.


git branch -a

gives me


* master
  remotes/origin/HEAD -> origin/master
  remotes/origin/master


I'm expecting a long list of branches which I could use to kick start 
my good and bad bisect steps.


Branches are not used. Tagged commits are used, but usually only 
"oldest" and "latest" are tagged.

Use
git tag -l
to list the tagged commits.

Using tags is not necessary. You can also just use commit hashes (of 
the bibisect repo commits.


The wiki article has some tips added this year, so do check it out: 
https://wiki.documentfoundation.org/QA/Bibisect


Ilmari
___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: 
https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? 
http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/

Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

[Libreoffice-qa] Using bibisect

2018-11-12 Thread Alex Kempshall 

I'm sure I've done this before, but for the moment my brain's gone dead!

I've changed to what I regard as my bibisect working directory and done

$git pull

Seemed to behave as I expected. I've subsequently done another pull and 
the results the second time are again to me as expected -



 instdir/share/registry/impress.xcd | 4 +-
 instdir/share/registry/main.xcd | 8 +-
 instdir/share/registry/writer.xcd | 2 +-
 216 files changed, 13933 insertions(+), 1646 deletions(-)
 create mode 100644 instdir/share/config/images_breeze_svg.zip
 create mode 100644 instdir/share/config/images_colibre_svg.zip
 create mode 100644 instdir/share/config/images_elementary_svg.zip
 create mode 100644 
instdir/share/config/soffice.cfg/modules/scalc/ui/notebookbar_compact.ui


If do

$git status

I get


On branch master
Your branch is up-to-date with 'origin/master'.

Untracked files:
  (use "git add ..." to include in what will be committed)

    instdir/user/

nothing added to commit but untracked files present (use "git add" to 
track)


At the moment I'm not worried about the untracked files.

For bisect, on source code, I would normally do a git branch -a to see 
what I've got and then start bisecting.


git branch -a

gives me


* master
  remotes/origin/HEAD -> origin/master
  remotes/origin/master


I'm expecting a long list of branches which I could use to kick start my 
good and bad bisect steps.


Is somebody able to kick my brain back into bibisect mode?

Alex




___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

No SDBC driver was found for the URL

2018-09-11 Thread Alex Kempshall 

I've upgraded to 6.2 using the daily build

I then start a base application which uses mariadb as a back end with 
the mysql-connector-java-8.0.12.jar


After entering the details in the "Authentication required" dialog I get 
a message box pop up with the following details.


The connection to the data source "_mysql" could not be 
established



The connection to the external data source could not be established. 
No SDBC driver was found for the URL 
'sdbc:mysql:jdbc:localhost:3306/x'




If I drop back to 6.1 my application works as expected. What have I missed?

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

Re: Segmentation fault when compiling

2018-08-02 Thread Alex Kempshall 
I've thrown some more memory at the problem. Increased actual memory 
from 3.5GB to 4GB. Seems to have fixed the problem. I already had 4GB of 
swap which I would have thought would have rolled in when the compiler 
needed it.


For the record I'm running Slackware 14.2 - 64bit, Kernel 4.4.144 with 
4GB  of physical memory and 4GB  of swap. All in VirtualBox 5.0.40.


I shall drop back to 3.5GB and see if I can create the problem again.

Alex




On 01/08/18 11:34, Alex Kempshall wrote:
I'm getting a segmentation fault when compiling. I've tried my usual 
techniques to overcome the problem such as - waiting till the next day 
and/or complete refresh. All to no avail.


I've been compiling for several years without major incident. Last 
compiled successfully on July 6th 2018.


Have I missed something  that I should have done recently to cater for 
some change. I'm not aware of other people encountering this problem. 
Should I raise it as a bug?


Alex


[build DEP] LNK:Library/libavmedialo.so
[build LNK] Library/libavmedialo.so
[build DEP] LNK:Library/libdbplo.so
[build LNK] Library/libdbplo.so
/bin/sh: line 1:  1727 Segmentation fault  ( 
LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program":$W/UnpackedTarball/cppunit/src/cppunit/.libs 
MALLOC_CHECK_=2 MALLOC_PERTURB_=153 
$W/LinkTarget/Executable/cppunittester 
$W/LinkTarget/CppunitTest/libtest_vcl_lifecycle.so --headless 
"-env:BRAND_BASE_DIR=file://$S/instdir" 
"-env:BRAND_SHARE_SUBDIR=share" 
"-env:BRAND_SHARE_RESOURCE_SUBDIR=program/resource" 
"-env:UserInstallation=file://$W/CppunitTest/vcl_lifecycle.test.user" 
"-env:CONFIGURATION_LAYERS=xcsxcu:file://$I/share/registry 
xcsxcu:file://$W/unittest/registry" 
"-env:UNO_TYPES=file://$I/program/types.rdb 
file://$I/program/types/offapi.rdb" 
"-env:UNO_SERVICES=file://$W/Rdb/ure/services.rdb 
file://$W/ComponentTarget/configmgr/source/configmgr.component 
file://$W/ComponentTarget/i18npool/util/i18npool.component 
file://$W/ComponentTarget/ucb/source/core/ucb1.component 
file://$W/ComponentTarget/ucb/source/ucp/file/ucpfile1.component 
file://$W/ComponentTarget/framework/u/bin/sh: line 1:  1713 
Segmentation fault  ( 
LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program":$W/UnpackedTarball/cppunit/src/cppunit/.libs 
MALLOC_CHECK_=2 MALLOC_PERTURB_=153 
$W/LinkTarget/Executable/cppunittester 
$W/LinkTarget/CppunitTest/libtest_toolkit.so --headless 
"-env:BRAND_BASE_DIR=file://$S/instdir" 
"-env:BRAND_SHARE_SUBDIR=share" 
"-env:BRAND_SHARE_RESOURCE_SUBDIR=program/resource" 
"-env:UserInstallation=file://$W/CppunitTest/toolkit.test.user" 
"-env:CONFIGURATION_LAYERS=xcsxcu:file://$I/share/registry 
xcsxcu:file://$W/unittest/registry" 
"-env:UNO_TYPES=file://$I/program/types/offapi.rdb 
file://$I/program/types.rdb" 
"-env:UNO_SERVICES=file://$W/Rdb/ure/services.rdb 
file://$W/ComponentTarget/comphelper/util/comphelp.component 
file://$W/ComponentTarget/configmgr/source/configmgr.component 
file://$W/ComponentTarget/framework/util/fwk.component 
file://$W/ComponentTarget/i18npool/util/i18npool.component 
file://$W/ComponentTarget/sfx2/util/sfx.component 
file://$W/ComponentTarget/toolkit/util/tk.component 
file://$W/ComponentTarget/ucb/source/core/ucb1.component 
file://$W/ComponentTarget/ucb/source/ucp/file/ucpfile1.component" 
-env:URE_INTERNAL_LIB_DIR=file://$I/program 
-env:LO_LIB_DIR=file://$I/program 
-env:LO_JAVA_DIR=file://$I/program/classes --protector 
$W/LinkTarget/Library/unoexceptionprotector.so unoexceptionprotector 
--protector $W/LinkTarget/Library/unobootstrapprotector.so 
unobootstrapprotector --protector 
$W/LinkTarget/Library/libvclbootstrapprotector.so 
vclbootstrapprotector 
"-env:CPPUNITTESTTARGET=$W/CppunitTest/toolkit.test" ) > 
$W/CppunitTest/toolkit.test.log 2>&1
til/fwk.component file://$W/ComponentTarget/sfx2/util/sfx.component" 
-env:URE_INTERNAL_LIB_DIR=file://$I/program 
-env:LO_LIB_DIR=file://$I/program 
-env:LO_JAVA_DIR=file://$I/program/classes --protector 
$W/LinkTarget/Library/unoexceptionprotector.so unoexceptionprotector 
--protector $W/LinkTarget/Library/unobootstrapprotector.so 
unobootstrapprotector --protector 
$W/LinkTarget/Library/libvclbootstrapprotector.so 
vclbootstrapprotector 
"-env:CPPUNITTESTTARGET=$W/CppunitTest/vcl_lifecycle.test" ) > 
$W/CppunitTest/vcl_lifecycle.test.log 2>&1
warn:vcl.opengl:1713:1713:vcl/opengl/x11/X11DeviceInfo.cxx:356: 
unknown vendor => blocked

Window position and size in pixel:
X: 100
Y: 100
Width: 640
Height: 480
Window position:
X: 2646 1/100mm
Y: 2646 1/100mm
Window size:
Width: 16933 1/100mm
Height: 12700 1/100mm
Window size:
Width: 1693 1/10mm
Height: 1270 1/10mm
Window size:
Width: 640 pixel
Height: 480 pixel
Window size:
Width: 332 appfont
Height: 295 appfont
Window si

Segmentation fault when compiling

2018-08-01 Thread Alex Kempshall 
I'm getting a segmentation fault when compiling. I've tried my usual 
techniques to overcome the problem such as - waiting till the next day 
and/or complete refresh. All to no avail.


I've been compiling for several years without major incident. Last 
compiled successfully on July 6th 2018.


Have I missed something  that I should have done recently to cater for 
some change. I'm not aware of other people encountering this problem. 
Should I raise it as a bug?


Alex


[build DEP] LNK:Library/libavmedialo.so
[build LNK] Library/libavmedialo.so
[build DEP] LNK:Library/libdbplo.so
[build LNK] Library/libdbplo.so
/bin/sh: line 1:  1727 Segmentation fault  ( 
LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program":$W/UnpackedTarball/cppunit/src/cppunit/.libs 
MALLOC_CHECK_=2 MALLOC_PERTURB_=153 
$W/LinkTarget/Executable/cppunittester 
$W/LinkTarget/CppunitTest/libtest_vcl_lifecycle.so --headless 
"-env:BRAND_BASE_DIR=file://$S/instdir" 
"-env:BRAND_SHARE_SUBDIR=share" 
"-env:BRAND_SHARE_RESOURCE_SUBDIR=program/resource" 
"-env:UserInstallation=file://$W/CppunitTest/vcl_lifecycle.test.user" 
"-env:CONFIGURATION_LAYERS=xcsxcu:file://$I/share/registry 
xcsxcu:file://$W/unittest/registry" 
"-env:UNO_TYPES=file://$I/program/types.rdb 
file://$I/program/types/offapi.rdb" 
"-env:UNO_SERVICES=file://$W/Rdb/ure/services.rdb 
file://$W/ComponentTarget/configmgr/source/configmgr.component 
file://$W/ComponentTarget/i18npool/util/i18npool.component 
file://$W/ComponentTarget/ucb/source/core/ucb1.component 
file://$W/ComponentTarget/ucb/source/ucp/file/ucpfile1.component 
file://$W/ComponentTarget/framework/u/bin/sh: line 1:  1713 
Segmentation fault  ( 
LD_LIBRARY_PATH=${LD_LIBRARY_PATH:+$LD_LIBRARY_PATH:}"$I/program:$I/program":$W/UnpackedTarball/cppunit/src/cppunit/.libs 
MALLOC_CHECK_=2 MALLOC_PERTURB_=153 
$W/LinkTarget/Executable/cppunittester 
$W/LinkTarget/CppunitTest/libtest_toolkit.so --headless 
"-env:BRAND_BASE_DIR=file://$S/instdir" 
"-env:BRAND_SHARE_SUBDIR=share" 
"-env:BRAND_SHARE_RESOURCE_SUBDIR=program/resource" 
"-env:UserInstallation=file://$W/CppunitTest/toolkit.test.user" 
"-env:CONFIGURATION_LAYERS=xcsxcu:file://$I/share/registry 
xcsxcu:file://$W/unittest/registry" 
"-env:UNO_TYPES=file://$I/program/types/offapi.rdb 
file://$I/program/types.rdb" 
"-env:UNO_SERVICES=file://$W/Rdb/ure/services.rdb 
file://$W/ComponentTarget/comphelper/util/comphelp.component 
file://$W/ComponentTarget/configmgr/source/configmgr.component 
file://$W/ComponentTarget/framework/util/fwk.component 
file://$W/ComponentTarget/i18npool/util/i18npool.component 
file://$W/ComponentTarget/sfx2/util/sfx.component 
file://$W/ComponentTarget/toolkit/util/tk.component 
file://$W/ComponentTarget/ucb/source/core/ucb1.component 
file://$W/ComponentTarget/ucb/source/ucp/file/ucpfile1.component" 
-env:URE_INTERNAL_LIB_DIR=file://$I/program 
-env:LO_LIB_DIR=file://$I/program 
-env:LO_JAVA_DIR=file://$I/program/classes --protector 
$W/LinkTarget/Library/unoexceptionprotector.so unoexceptionprotector 
--protector $W/LinkTarget/Library/unobootstrapprotector.so 
unobootstrapprotector --protector 
$W/LinkTarget/Library/libvclbootstrapprotector.so 
vclbootstrapprotector 
"-env:CPPUNITTESTTARGET=$W/CppunitTest/toolkit.test" ) > 
$W/CppunitTest/toolkit.test.log 2>&1
til/fwk.component file://$W/ComponentTarget/sfx2/util/sfx.component" 
-env:URE_INTERNAL_LIB_DIR=file://$I/program 
-env:LO_LIB_DIR=file://$I/program 
-env:LO_JAVA_DIR=file://$I/program/classes --protector 
$W/LinkTarget/Library/unoexceptionprotector.so unoexceptionprotector 
--protector $W/LinkTarget/Library/unobootstrapprotector.so 
unobootstrapprotector --protector 
$W/LinkTarget/Library/libvclbootstrapprotector.so 
vclbootstrapprotector 
"-env:CPPUNITTESTTARGET=$W/CppunitTest/vcl_lifecycle.test" ) > 
$W/CppunitTest/vcl_lifecycle.test.log 2>&1
warn:vcl.opengl:1713:1713:vcl/opengl/x11/X11DeviceInfo.cxx:356: 
unknown vendor => blocked

Window position and size in pixel:
X: 100
Y: 100
Width: 640
Height: 480
Window position:
X: 2646 1/100mm
Y: 2646 1/100mm
Window size:
Width: 16933 1/100mm
Height: 12700 1/100mm
Window size:
Width: 1693 1/10mm
Height: 1270 1/10mm
Window size:
Width: 640 pixel
Height: 480 pixel
Window size:
Width: 332 appfont
Height: 295 appfont
Window size:
Width: 332 sysfont
Height: 295 sysfont
Window size:
Width: 169 mm
Height: 127 mm
Window size:
Width: 17 cm
Height: 13 cm
Window size:
Width: 6667 1/1000inch
Height: 5000 1/1000inch
Window size:
Width: 667 1/100inch
Height: 500 1/100inch
Window size:
Width: 67 1/10inch
Height: 50 1/10inch
Window size:
Width: 7 inch
Height: 5 inch
Window size:
Width: 480 point
Height: 360 point
Window size:
Width: 9600 twip
Height: 7200 twip
Window size:
Width: 640 pixel
Height: 480 pixel
(anonymous namespace)::ToolkitTest::testXUnitConversion finished in: 728ms
OK (1)
warn:fwk.desktop:1713:1713:framework/source/services/desktop.cxx:1069: 
Desktop disposed before terminating it


No core file identified

tdf#115718 Fix Watch Window EditBox

2018-07-03 Thread Alex Kempshall 
I submitted a patch through logerrit a couple of weeks ago. It was 
suggested that I make a small change to the comment.


What I did was

1. abandon the change, though was suggested I needn't have done this
2. redo the change with the improved comment
3. resubmitted the change with logerrit
4. I initially got a Jenkins "FAILURE", which I suspected was because
   of another issue. So I did nothing.
5. a few hours late Jenkins sent a  "Build Successful" message. The
   link can be found here Build #9097
   

How do I get to move on to the next phase of the process?

Is there any documentation that describes the process?

Alex

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

Re: [Libreoffice-qa] Next level of QA and how to get there

2018-06-25 Thread Alex Kempshall 

Hi Xisco

I would have liked to have listened in on tomorrows meeting, 
unfortunately I have to be elsewhere.


Back in 2016 Buovjaga wrote this to me


mailmerge testing is very valuable as it is a complex feature and typically not
many QA members like to tackle it.

On IRC, there are usually people active during European waking hours. Jan-Marek
Glogowski is a developer doing a lot for mailmerge (working for City of Munich).
He also hangs out at the QA IRC channel, using the nickname jmux. He is
currently on vacation (a week still), but still seems to be joining IRC now and
then.

I have triaged a lot of mailmerge issues myself, even though I don't use the
feature.

On IRC you could get guidance on how to bisect the regressions you find.
Bisecting is very helpful as it pinpoints the offending commit.
Here is our info page on (bi)bisecting aka binary bisecting:
https://wiki.documentfoundation.org/QA/Bibisect


Would like to help where I can, but my time is limited.

I am keeping an eye on MailMerge and database activity, mainly because 
I'm an active user of such things. Have in the past found problems in 
Stable that have take time to sort out. Never quite managed to do much 
testing in Alphas or Betas.


I've just got a successful compile of LO with --enable-kde5 and 
--enable-qt5. It looks quite a mess - is this expected? If not I will 
raise a bug report. I knew that Jan-Marek was taking an active interest 
in kde5 integration, but he is not active at the moment. Who's looking 
after kde integration at the moment. Had this message, a few days ago, 
from Katarina Behrens



I'm looking into KDE5 implementation every day :D
KDE5 filepicker just landed in master earlier this week. Sadly I don't have
any screenshots at hand but Haiku developers do [1]

[1]https://www.haiku-os.org/blog/waddlesplash/
2018-06-06_haiku_monthly_activity_report_-_052018/


I suppose what I'm saying is that when looking in from the outside it's 
very difficult to understand what's going on and where we are with 
LibreOffice


Alex


On 22/06/18 17:00, Xisco Fauli wrote:

Hello everyone,

In line with buovjaga's proposal, I would like to propose the next QA
meeting ( June 26th at 15:00 UTC) to talk about more concrete actions
and ideas...

For instance, we could run the week/month of QA, similar to the Month Of
LibreOffice, where we could try to put the list of unconfirmed bugs to
150/200 and engage new contributors in the project...

It's just an idea but it could help the team to reach the steps
described by Buovjaga in a reasonable amount of time.

What do you think?

Regards

El 14/06/18 a les 17:37, Ilmari Lauhakangas ha escrit:

Sure, the ultimate goal is that no one has to burn themselves out. If
you feel like it, use the "last month" query found here:
https://wiki.documentfoundation.org/QA/GetInvolved#Find_unconfirmed_bugs

Ilmari

On 14.06.2018 18:33, Sophia Schröder wrote:

Hi *,

well, I could try to step my feeds in the field of "confirming bugs"
too,
given that Bugzilla and me are no best friends (too complicated by now).

So providing a BZ query in this regard would help me.
OTOH I am full with other stuff to do
even RL and/or LO related so I cannot promise so much time.

I am on Windows 10 64 Bit using LO 6.0.5.1 64 Bit too ATM.
Printer: Canon Pixma MX925 4-in-1


Am 14.06.2018 um 16:58 schrieb Ilmari Lauhakangas:

The idea is simply:
- what would it take to get to next levels
- how things would change in practice

I just thought I would lay that out in public, so it will hopefully
inspire others to continue or get involved.

Everyone is free to draft their own plans. Personally, I am focusing
on VolunteerMatch. Over the years I have tried so many ways to
recruit that I've lost count and VM feels the most promising so far.

Ilmari

On 14.06.2018 17:36, dgp-m...@gmx.de wrote:

I totally agree to all ideas and concrete aims, but is there a
specific plan to reach the different milestones? If this is the
case, it would be great to know something more about it. If there
is now specific plan: What is the idea behind your email?

Dieter


Am 13.06.2018 um 15:05 schrieb Ilmari Lauhakangas:

The impact of the QA team has been roughly on the same level for
four years and it is getting boring. Now that we have new
contributors trickling in through VolunteerMatch, we have some
hope of leveling up.

What does getting to the next level mean? Freedom. We will be able
to learn new skills and complete projects that have been
languishing in our TODOs for ages. Maybe even have time to read a
book or watch a movie (I know, preposterous).

I will use unconfirmeds as a yardstick as that is something we can
all easily grasp.
The assumption is that we continue to get 500-800 reports per month.
A prerequisite to leveling up is that the triaging work is spread
evenly between contributors. In practice: only needing an average
rate of one triaged report per day per regular tester. This would
signal sustainability.

When we manage to keep the

Re: Compiler error: Libraries must be registered in Repository.mk

2018-06-22 Thread Alex Kempshall 



On 22/06/18 09:43, Stephan Bergmann wrote:

On 22/06/18 09:49, Alex Kempshall wrote:
I've now set up a test bed with Slackware Current (kde5) and have 
tried to compile with the following flags



--disable-ccache \
--disable-gtk3 \
--enable-kde5 \
--with-parallelism=5 \
--without-junit \
--without-gssapi \
--without-krb5 


Almost immediately it fails with

[info  ALL] currently known libraries are: OGLTrans 
PresentationMinimizer PresenterScreen abp acc active_native 
affine_uno_uno analysis animcore avmedia avmediagst basctl basegfx 
basprov bib binaryurp bootstrap cached1 cairocanvas calc 
canvasfactory canvastools chartcontroller chartcore chartopengl clew 
clucene cmdmail collator_data comphelper configmgr cppcanvas cppu 
cppuhelper ctl cui date dba dbahsql dbase dbaxml dbmm dbp dbpool2 
dbtools dbu deployment deploymentgui deploymentmisc desktop_detector 
desktopbe1 dict_ja dict_zh dlgprov drawinglayer editeng embobj 
emboleobj emfio epoxy evtatt expwrap file filterconfig firebird_sdbc 
flash flat for forui fps_office frm fsstorage fwe fwi fwk fwl fwm 
gcc3_uno gie graphicfilter guesslang helplinker hsqldb hwp hyphen 
i18nlangtag i18npool i18nsearch i18nutil icg index_data 
introspection invocadapt invocation io java_uno javaloader javavm 
jdbc jpipe juh juhx jvmaccess jvmfwk kde5be1 ldapbe2 lng lnth 
localebe1 localedata_en localedata_es localedata_euro 
localedata_others log log_uno_uno losessioninstall lwpft mcnttype 
migrationoo2 migrationoo3 mork mozbootstrap msfilter msforms msword 
mtfrenderer mysql mysqlc mysqlcppconn namingservice neon numbertext 
odbc odfflatxml offacc officebean oglcanvas oox opencl package2 
passive_native passwordcontainer pcr pdffilter pdfimport pdfium 
postgresql-sdbc postgresql-sdbc-impl pricing protocolhandler 
proxyfac purpenvhelper pythonloader pyuno pyuno_wrapper recentfile 
reflection reg rpt rptui rptxml sal sal_textenc salhelper sax sb sc 
scd scfilt scn scqahelper scriptframe scui sd sdbc2 sdbt sdd sdfilt 
sdui sfx simplecanvas slideshow sm smd smoketest sofficeapp solver 
sot spell spl srtrs1 stocservices storagefd store stringresource 
subsequenttest svgfilter svgio svl svt svx svxcore sw swd swui syssh 
t602filter test test-setupvcl testtools_bridgetest 
testtools_bridgetest-common testtools_constructors testtools_cppobj 
textconv_dict textconversiondlgs textfd tk tl ucb1 ucbhelper 
ucpchelp1 ucpcmis1 ucpdav1 ucpexpand1 ucpext ucpfile1 ucpftp1 
ucpgio1 ucphier1 ucpimage ucppkg1 ucptdoc1 unobootstrapprotector 
unoexceptionprotector unoidl unopkgapp unordf unotest unowinreg 
unoxml unsafe_uno_uno updatefeed utl uui uuresolver vbaevents 
vbahelper vbaobj vbaswobj vcl vclbootstrapprotector vclcanvas 
vclplug_gen vclplug_gtk vclplug_kde5 wpftcalc wpftdraw wpftimpress 
wpftqahelper wpftwriter writer writerfilter writerperfect xmlfa 
xmlfd xmlreader xmlscript xmlsecurity xo xof xsec_xmlsec xsltdlg 
xsltfilter xstor
/home/alex/master/vcl/Library_vclplug_kde5.mk:36: *** Cannot link 
against library/libraries vclplug_qt5. Libraries must be registered 
in Repository.mk or RepositoryExternal.mk.  Stop.

make: *** [Makefile:286: build] Error 2


vcl/Library_vclplug_kde5.mk depends on vcl/Library_vclplug_qt5.mk, but 
--enable-kde5 presumably doesn't set ENABLE_QT5=TRUE, so that 
dependency doesn't get built.  A stopgap measure until that's fixed 
properly would be to also configure --with-qt5 (which will add a 
instdir/program/libvclplug_qt5lo.so that you don't really need, but 
which shouldn't hurt, either).



Hi Stephan

I added --enable-qt5 which got me a clean compile.

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

Compiler error: Libraries must be registered in Repository.mk

2018-06-22 Thread Alex Kempshall 
Historically I've compiled the git clone master in Slackware 14.2 (kde4) 
with the following flags -



--disable-ccache \
--disable-gtk3 \
--enable-kde4 \
--with-parallelism=5 \
--without-junit \
--without-gssapi \
--without-krb5


never given me a problem

I've now set up a test bed with Slackware Current (kde5) and have tried 
to compile with the following flags



--disable-ccache \
--disable-gtk3 \
--enable-kde5 \
--with-parallelism=5 \
--without-junit \
--without-gssapi \
--without-krb5 


Almost immediately it fails with

[info  ALL] currently known libraries are: OGLTrans 
PresentationMinimizer PresenterScreen abp acc active_native 
affine_uno_uno analysis animcore avmedia avmediagst basctl basegfx 
basprov bib binaryurp bootstrap cached1 cairocanvas calc canvasfactory 
canvastools chartcontroller chartcore chartopengl clew clucene cmdmail 
collator_data comphelper configmgr cppcanvas cppu cppuhelper ctl cui 
date dba dbahsql dbase dbaxml dbmm dbp dbpool2 dbtools dbu deployment 
deploymentgui deploymentmisc desktop_detector desktopbe1 dict_ja 
dict_zh dlgprov drawinglayer editeng embobj emboleobj emfio epoxy 
evtatt expwrap file filterconfig firebird_sdbc flash flat for forui 
fps_office frm fsstorage fwe fwi fwk fwl fwm gcc3_uno gie 
graphicfilter guesslang helplinker hsqldb hwp hyphen i18nlangtag 
i18npool i18nsearch i18nutil icg index_data introspection invocadapt 
invocation io java_uno javaloader javavm jdbc jpipe juh juhx jvmaccess 
jvmfwk kde5be1 ldapbe2 lng lnth localebe1 localedata_en localedata_es 
localedata_euro localedata_others log log_uno_uno losessioninstall 
lwpft mcnttype migrationoo2 migrationoo3 mork mozbootstrap msfilter 
msforms msword mtfrenderer mysql mysqlc mysqlcppconn namingservice 
neon numbertext odbc odfflatxml offacc officebean oglcanvas oox opencl 
package2 passive_native passwordcontainer pcr pdffilter pdfimport 
pdfium postgresql-sdbc postgresql-sdbc-impl pricing protocolhandler 
proxyfac purpenvhelper pythonloader pyuno pyuno_wrapper recentfile 
reflection reg rpt rptui rptxml sal sal_textenc salhelper sax sb sc 
scd scfilt scn scqahelper scriptframe scui sd sdbc2 sdbt sdd sdfilt 
sdui sfx simplecanvas slideshow sm smd smoketest sofficeapp solver sot 
spell spl srtrs1 stocservices storagefd store stringresource 
subsequenttest svgfilter svgio svl svt svx svxcore sw swd swui syssh 
t602filter test test-setupvcl testtools_bridgetest 
testtools_bridgetest-common testtools_constructors testtools_cppobj 
textconv_dict textconversiondlgs textfd tk tl ucb1 ucbhelper ucpchelp1 
ucpcmis1 ucpdav1 ucpexpand1 ucpext ucpfile1 ucpftp1 ucpgio1 ucphier1 
ucpimage ucppkg1 ucptdoc1 unobootstrapprotector unoexceptionprotector 
unoidl unopkgapp unordf unotest unowinreg unoxml unsafe_uno_uno 
updatefeed utl uui uuresolver vbaevents vbahelper vbaobj vbaswobj vcl 
vclbootstrapprotector vclcanvas vclplug_gen vclplug_gtk vclplug_kde5 
wpftcalc wpftdraw wpftimpress wpftqahelper wpftwriter writer 
writerfilter writerperfect xmlfa xmlfd xmlreader xmlscript xmlsecurity 
xo xof xsec_xmlsec xsltdlg xsltfilter xstor
/home/alex/master/vcl/Library_vclplug_kde5.mk:36: *** Cannot link 
against library/libraries vclplug_qt5. Libraries must be registered in 
Repository.mk or RepositoryExternal.mk.  Stop.

make: *** [Makefile:286: build] Error 2


I've also tried compiling on Kubuntu 17.10 and get the same error.

Can anyone advise on how this can be fixed.

Alex




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

Re: [Libreoffice-qa] Correct use of bibisect to identify a particular commit

2018-06-21 Thread Alex Kempshall 

Hi Xisco

I've move to the max series.

Thanks

Alex



On 20/06/18 14:34, Xisco Fauli wrote:

Hello Alex,


El 20/06/18 a les 13:47, Alex Kempshall ha escrit:

Hi Xisco

Still don't understand what the  technique is to get from a bibisect
log to a bisect log.

If you remember I identified, by using bibisect, when a regression
occurred. Then you went on to identify the actual commit see

https://bugs.documentfoundation.org/show_bug.cgi?id=118220#c3

So, taking this as example.

I used bibisect-linux64-6.2 [1] as mentioned in the comment. This is a
'max' repository which means there's one build per commit, thus, if I
bisect that bug, I get:

670b6459ca52221d8c2a314dd55133c8dada35d4 is the first bad commit
commit 670b6459ca52221d8c2a314dd55133c8dada35d4
Author: Jenkins Build User 
Date:   Tue Jun 5 00:24:31 2018 +0200

     source sha:b69457c1a5cb31c8b0fb3d047e2826bc3d61fea4

     source sha:b69457c1a5cb31c8b0fb3d047e2826bc3d61fea4


which points directly to the commit in source sha.

OTOH, as you used a gdbutil daily repo to bibisect it, you got:

# good: [072a924b4800a972e66f4f2c043d91b79749a19c] 2018-05-29:
source-hash-63ce2fa4835109eeab172ab167dafe4a752b6764
git bisect good 072a924b4800a972e66f4f2c043d91b79749a19c
# first bad commit: [26517de6f6c3dc7472175ca36e5a959d7b9bbf21]
2018-05-30: source-hash-7725eff80a7bc3f3e422e56e1cd7bfa531f19a70

which returns you a range of commits [2]. At this point you could check
manually with commit introduced the problem, but it's much easier to use
the 'max' repositories instead.

Regards

[1] https://gerrit.libreoffice.org/#/admin/projects/bibisect-linux-64-6.2
[2]
https://cgit.freedesktop.org/libreoffice/core/log/?qt=range=63ce2fa4835109eeab172ab167dafe4a752b6764..7725eff80a7bc3f3e422e56e1cd7bfa531f19a70




___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

Re: [Libreoffice-qa] Correct use of bibisect to identify a particular commit

2018-06-20 Thread Alex Kempshall 

Hi Xisco

Still don't understand what the  technique is to get from a bibisect log 
to a bisect log.


If you remember I identified, by using bibisect, when a regression 
occurred. Then you went on to identify the actual commit see


https://bugs.documentfoundation.org/show_bug.cgi?id=118220#c3

Just want to know how you got from my bibisect.

Alex


On 12/06/18 11:24, Alex Kempshall wrote:

I'm using the dbgutil series. Storage is cheap so I've cloned

git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till61.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till60.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till54.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till53.git

So I'm covered from May 2016 through to May 2018.

Alex


On 11/06/18 15:11, Xisco Fauli wrote:

Hello Alex,


El 11/06/18 a les 14:53, Alex Kempshall ha escrit:

Hi all

I'm more of a bisect person, but towards the end of last week I did a
bibisect looking for a regression I've encountered. Will post details
later today.

As far as my understanding goes I can use bibisect to identify which
day the commit that caused the regression was made. How do I then use
this information to identify the actual commit?

it depends on which kind of repository you're using. For linux [1], we
have the  'max' ones, covering from 4.2 to master, which contain one
commit per build. OTOH, we also have the dbgutil daily repos which are
build once a day. Finally, the 'all' repository contains 64 commits per
build.

Could you please tell us which repo you are using?

Regards

[1] https://wiki.documentfoundation.org/QA/Bibisect/Linux#Versions

Alex
___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings:
https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems?
http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: 
http://wiki.documentfoundation.org/Netiquette

List archive: http://lists.freedesktop.org/archives/libreoffice-qa/




___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

Re: What does weld do for libreoffice

2018-06-19 Thread Alex Kempshall 

How does kde fit into this?

Would Tools-> Macros -> Organise Macros -> LibreOffice Basic -> Edit 
become a candidate for "welding"



On 17/06/18 21:03, Caolán McNamara wrote:

On Fri, 2018-06-15 at 22:35 +0200, Michael Stahl wrote:

see
http://caolanm.blogspot.com/2018/04/some-native-gtk-dialogs-in-libreo
ffice.html

Yes, so this "welding" makes it possible to replace the vcl
implementation of dialogs with an alternative toolkit implementation,
namely GTK. The dialogs are already described by the GtkBuilder file
format and the translations are already in compatible gettext format so
the GTK implementation can use those directly. Non gtk platforms
continue to use the vcl implementation with the vcl parser of the
GtkBuilder file format which maps them to equivalent vcl widgets, so
other platforms should generally remain as they are.

Writer's "insert->table dialog", "table->split cells" and
"insert->special characters" are some concrete examples of this at the
moment.

Under GTK you can tell if its a welded dialog because the themeing is,
by definition I suppose, fully correct. Stuff like animated checkboxes
for example and fade in of entry borders on focus-in now work. Password
entries that indicate if the caps lock is on, and the semi-transparent
overlay scrollbar are some other indicators that a dialog has been
converted over.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice


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

What does weld do for libreoffice

2018-06-15 Thread Alex Kempshall 

I saw this commit the other day


commit 6dc4745117d93cea45a8e02721e4850d4cb17243
Author: Caolán McNamara 
Date:   Wed Jun 13 21:07:53 2018 +0100

    weld SvxAsianTabPage


Wondering what this weld is?

Is it a new method of generating forms?

Will eventually all forms be created this way.

Is there a program for this work?

Alex

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

Re: [Libreoffice-qa] Correct use of bibisect to identify a particular commit

2018-06-12 Thread Alex Kempshall 

I'm using the dbgutil series. Storage is cheap so I've cloned

git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till61.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till60.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till54.git
git://dev-downloads.libreoffice.org/lo-linux-dbgutil-daily-till53.git

So I'm covered from May 2016 through to May 2018.

Alex


On 11/06/18 15:11, Xisco Fauli wrote:

Hello Alex,


El 11/06/18 a les 14:53, Alex Kempshall ha escrit:

Hi all

I'm more of a bisect person, but towards the end of last week I did a
bibisect looking for a regression I've encountered. Will post details
later today.

As far as my understanding goes I can use bibisect to identify which
day the commit that caused the regression was made. How do I then use
this information to identify the actual commit?

it depends on which kind of repository you're using. For linux [1], we
have the  'max' ones, covering from 4.2 to master, which contain one
commit per build. OTOH, we also have the dbgutil daily repos which are
build once a day. Finally, the 'all' repository contains 64 commits per
build.

Could you please tell us which repo you are using?

Regards

[1] https://wiki.documentfoundation.org/QA/Bibisect/Linux#Versions

Alex
___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings:
https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems?
http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/


___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

[Libreoffice-qa] Correct use of bibisect to identify a particular commit

2018-06-11 Thread Alex Kempshall 

Hi all

I'm more of a bisect person, but towards the end of last week I did a 
bibisect looking for a regression I've encountered. Will post details 
later today.


As far as my understanding goes I can use bibisect to identify which day 
the commit that caused the regression was made. How do I then use this 
information to identify the actual commit?


Alex
___
List Name: Libreoffice-qa mailing list
Mail address: Libreoffice-qa@lists.freedesktop.org
Change settings: https://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

Re: clang compiler gives warnings expecting errors

2017-09-18 Thread Alex Kempshall 

Hi Stephan

That worked a treat.

Thanks

Alex



On 15/09/17 09:18, Stephan Bergmann wrote:
You need to configure LO with --enable-werror.  (Which developers 
should do anyway, independent of Clang/loplugin.) 


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

clang compiler gives warnings expecting errors

2017-09-15 Thread Alex Kempshall 

Hi Stephen

Thanks for the pointer on how to compile with clang. Done an autogen -

Running ./configure with 'CC=clang CXX=clang++ --enable-debug 
--disable-ccache --disable-gtk3 --with-parallelism=5 --without-junit 
--without-gssapi --without-krb5 --enable-dbgutil 
--enable-compiler-plugins --enable-option-checking=fatal 
--srcdir=/home/alex/master --enable-option-checking=fatal'


The make gives me a clean compile, with warnings,  which was not what I 
was hoping for!


/home/alex/master/sw/source/ui/inc/mmresultdialogs.hxx:213:5: 
*warning*: VclReferenceBase subclass dispose() function MUST call 
dispose() of its superclass as the last thing it does 
[loplugin:vclwidgets]
/home/alex/master/sw/source/ui/dbui/mmoutputtypepage.cxx:303:1: 
*warning*: VclReferenceBase subclass dispose() function MUST call 
dispose() of its superclass as the last thing it does 
[loplugin:vclwidgets]


I would have liked the compile to have failed in a similar way to 
Jenkins where the *warnings* above became *errors*-


In file included from 
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/dbui/mmoutputtypepage.cxx:37:
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/inc/mmresultdialogs.hxx:213:5:***error*: 
VclReferenceBase subclass dispose() function MUST call dispose() of 
its superclass as the last thing it does [loplugin:vclwidgets]

 virtual void    dispose() override;
 ^~
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/dbui/mmoutputtypepage.cxx:303:1: 
*error*: VclReferenceBase subclass dispose() function MUST call 
dispose() of its superclass as the last thing it does 
[loplugin:vclwidgets]

void SwSendMailDialog::dispose()
^~~~
2 errors generated.
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/solenv/gbuild/LinkTarget.mk:288: 
recipe for target 
'/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/workdir/CxxObject/sw/source/ui/dbui/mmoutputtypepage.o' 
failed
make[1]: *** 
[/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/workdir/CxxObject/sw/source/ui/dbui/mmoutputtypepage.o] 
Error 1

make[1]: *** Waiting for unfinished jobs
Makefile:266: recipe for target 'build' failed
make: *** [build] Error 2
Build step 'Execute shell' marked build as failure
Finished: FAILURE 


What have I missed?

Alex




On 14/09/17 11:40, Stephan Bergmann wrote:

On 09/13/2017 11:27 AM, Alex Kempshall wrote:
Is there a way that I can pick up these type of errors before 
submitting them to gerrit?


By building with Clang and configuring --enable-compiler-plugins. See 
<https://wiki.documentfoundation.org/Development/Building_LibreOffice_with_Clang> 
for some pointers (and 
<https://wiki.documentfoundation.org/Development/Clang_plugins> for 
more info on loplugin).




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

Build Failed - 18186

2017-09-14 Thread Alex Kempshall 

Hi all

I missed the initial Jenkins linux_clang_dbgutil_64 failure, got 
distracted by the patch going forward for code review even with the 
compile failure. Since then the patch has been submitted twice since 
then. Don't know why.


Anyway, I'm looking at the failure which is -


In file included from 
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/dbui/mmoutputtypepage.cxx:37:
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/inc/mmresultdialogs.hxx:213:5:
 error: VclReferenceBase subclass dispose() function MUST call dispose() of its 
superclass as the last thing it does [loplugin:vclwidgets]
 virtual voiddispose() override;
 ^~
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/sw/source/ui/dbui/mmoutputtypepage.cxx:303:1:
 error: VclReferenceBase subclass dispose() function MUST call dispose() of its 
superclass as the last thing it does [loplugin:vclwidgets]
void SwSendMailDialog::dispose()
^~~~
2 errors generated.
/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/solenv/gbuild/LinkTarget.mk:288:
 recipe for target 
'/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/workdir/CxxObject/sw/source/ui/dbui/mmoutputtypepage.o'
 failed
make[1]: *** 
[/home/tdf/lode/jenkins/workspace/lo_gerrit/Config/linux_clang_dbgutil_64/workdir/CxxObject/sw/source/ui/dbui/mmoutputtypepage.o]
 Error 1
make[1]: *** Waiting for unfinished jobs
Makefile:266: recipe for target 'build' failed
make: *** [build] Error 2
Build step 'Execute shell' marked build as failure
Finished: FAILURE


I think I understand what it's telling me in that an object or two 
aren't being disposed of at the appropriate time. Will take me a bit of 
time to work out how this can be resolved.


Is there a way that I can pick up these type of errors before submitting 
them to gerrit?


Alex






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

Subject: Alex Kempshall license statement

2016-12-06 Thread Alex Kempshall 

According to the Credits I joined in 2011-08-14 and did 1 commit.

More than five years later my second commit is on it's way. Thought it 
best to renew my licence statement so here it is -



All of my past & future contributions to LibreOffice may be
   licensed under the MPLv2/LGPLv3+ dual license.


Alex Kempshall

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

Re: Test File: sc/qa/unit/data/functions/fods/chiinv.fods: fails with Assertion

2016-09-23 Thread Alex Kempshall 


On 23/09/16 12:34, Stephan Bergmann wrote:
Should we want to force those flags for 32-bit x68 Linux (and thus 
drop support for x86 processors not supporting SSE2), my proposed 
patch would be:


Stephan

Do you want me to test this patch before moving on?

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

Re: Test File: sc/qa/unit/data/functions/fods/chiinv.fods: fails with Assertion

2016-09-22 Thread Alex Kempshall 

On 16/09/16 11:05, Stephan Bergmann wrote:


Turns out, the relevant GCC switches are

  -mfpmath=sse -msse2



On this basis I ran

   make clean && make ENVCFLAGS="-mfpmath=sse -msse2"
   ENVCFLAGSCXX="-mfpmath=sse -msse2" && make sc.check

The compile and tests all completed successfully on an up to date 32-bit 
Lubuntu 14.04 and 32-bit Slackware-Current from November 2015. I haven't 
pulled any new code since 1st September so haven't inadvertently picked 
up the work around.


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

Re: internal compiler error: in connect_traces, at dwarf2cfi.c:2676

2013-12-28 Thread Alex Kempshall 
Looks like setting the correct compiler flags fixed this problem for me. 


Alex



On Saturday 28 Dec 2013 09:52:43 mcmurchy1917-libreoff...@yahoo.co.uk wrote:
 Quite early on in compilling LO I get the following error. Any suggestions
 on how to resolve. It seems to suggest it's a problem with gcc not with LO.
  /usr/bin/make -j 2 -rs -f /home/libreoffice/Downloads/core/Makefile.gbuild
  \ 
   \
   \
   \
  
  all
  
  [build DEP] LNK:Library/libbasegfxlo.so
  [build DEP] LNK:Library/libucbhelper.so
  [build LNK] Library/libucbhelper.so
  [build CXX] basegfx/source/polygon/b2dsvgpolypolygon.cxx
  [build CXX] basegfx/source/polygon/b3dpolygontools.cxx
  [build CXX] basegfx/source/polygon/b3dpolypolygon.cxx
  /home/libreoffice/Downloads/core/basegfx/source/polygon/b2dsvgpolypolygon.
  cx x: In function 'rtl::OUString basegfx::tools::exportToSvgD(const
  basegfx::B2DPolyPolygon, bool, bool, bool)':
  /home/libreoffice/Downloads/core/basegfx/source/polygon/b2dsvgpolypolygon.
  c
  xx:935:9: internal compiler error: in connect_traces, at dwarf2cfi.c:2676
  
   }
   ^
  
  Please submit a full bug report,
  with preprocessed source if appropriate.
  See http://gcc.gnu.org/bugs.html for instructions.
  make[1]: ***
  [/home/libreoffice/Downloads/core/workdir/CxxObject/basegfx/source/polygon
  /
  b2dsvgpolypolygon.o] Error 1 make[1]: *** Waiting for unfinished jobs
  make: *** [build] Error 2
 
 Alex
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Preserve CollateSingleJobs between calls to print dialog (Bug 32841)

2011-08-15 Thread Alex Kempshall 
Yes, this patch is contributed under the LGPLv3+ / MPL license.

Regards

Alex Kempshall





On Sunday 14 Aug 2011 18:26:55 Norbert Thiebaud wrote:
 On Sun, Aug 14, 2011 at 3:08 AM,  mcmurchy1917-libreoff...@yahoo.co.uk 
wrote:
  This patch will preserve the contents of Create separate print jobs for
  collated output Checkbox on the Options tab between calls to the Print
  Dialog.
 
 btw: can you confirm the license: GPLv3+/MPL 1.1
 
 Thanks
 
  Alex Kempshall
  ___
  LibreOffice mailing list
  LibreOffice@lists.freedesktop.org
  http://lists.freedesktop.org/mailman/listinfo/libreoffice
 
 ___
 LibreOffice mailing list
 LibreOffice@lists.freedesktop.org
 http://lists.freedesktop.org/mailman/listinfo/libreoffice
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice