Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Stephan Bergmann]

On 01/04/2012 11:02 PM, Alex Thurgood wrote:

Unfortunately, I'm not on Windows. I can only build on Linux and Mac. Am
looking to test with gdb on Linux to see what is being called (or not)
as the case may be.


But the bug as described only happens on Windows, right?  (Just tested 
on a recent unxlngx6 master build, and 
mysql-connector-ooo-1.0.1-linux-x86_64.oxt as available from 
 indeed 
installs without problem.)


In that case, I doubt that inspecting Linux behaviour gives a clue to 
what fails on Windows.  I suspect it rather has something to do with the 
Windows MSVC runtime libraries or somesuch.


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

[Libreoffice] Problem building the 3.5 branch on Windows, postgresql related.

[Kohei Yoshida]

Hi there,

I'm having trouble building the latest 3.5 branch on Windows.  First,
the postgresql module failed to build due to it not finding ldap.h
from mozilla.  I had mozilla entirely disabled (I always had).  So I
decided to enable it to see if it would solve this.  But because I
didn't want to build it from the sources, I simply stashed the
pre-built mozilla zips into moz/zipped.

Now the postgresql module builds, but it fails in connectivity due to
unresolved symbols (as follows).

/MAP /OPT:NOREF -safeseh -nxcompat -dynamicbase -NODEFAULTLIB -RELEASE
-DEBUG -INCREMENTAL:NO /SUBSYSTEM:CONSOLE /DLL
-out:../../../wntmsci12.pro/bin/postgresql-sdbc-impl.uno.dll
-map:../../../wntmsci12.pro/misc/postgresql-sdbc-impl.uno.map
-def:../../../wntmsci12.pro/misc/postgresql-sdbc-impl.uno.def
-implib:../../../wntmsci12.pro/lib/ipostgresql_t2.lib
../../../wntmsci12.pro/slo/postgresql-sdbc-impl.uno_version.obj
../../../wntmsci12.pro/slb/postgresql-sdbc-impl.uno.lib icppu.lib
icppuhelper.lib isal.lib isalhelper.lib
C:/libo/libreoffice-3-5/solver/wntmsci12.pro/lib/libpq.lib ws2_32.lib
secur32.lib advapi32.lib shell32.lib ssleay32.lib libeay32.lib
msvcrt.lib msvcprt.lib uwinapi.lib kernel32.lib user32.lib
oldnames.lib ../../../wntmsci12.pro/misc/postgresql-sdbc-impl.uno.res

   Creating library ../../../wntmsci12.pro/lib/ipostgresql_t2.lib and
object ../../../wntmsci12.pro/lib/ipostgresql_t2.exp

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_value_free_len@4 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_get_values_len@12 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_first_entry@8 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_count_entries@8 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_unbind@4 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_err2string@4 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_search_st@32 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_msgfree@4 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_result@20 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_simple_bind@12 referenced in function _ldapServiceLookup

libpq.lib(fe-connect.obj) : error LNK2019: unresolved external symbol
_ldap_init@8 referenced in function _ldapServiceLookup

../../../wntmsci12.pro/bin/postgresql-sdbc-impl.uno.dll : fatal error
LNK1120: 11 unresolved externals

dmake:  Error code 96, while making
'../../../wntmsci12.pro/bin/postgresql-sdbc-impl.uno.dll'

Any idea how I could get this issue resolved?  Do I now need to build
mozilla from sources on Windows, or ...?  How do others build 3.5 on
Windows these days?

Help appreciated.

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

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Importing smartArt from

[Muthu Subramanian K]

Thank you so much!
So, I assume we have 3 reviews for this.
I will push this to 3.5 today, if there are no objections by then.

Thanks again for the reviews!
Muthu Subramanian

On 01/05/2012 02:29 AM, Miklos Vajna wrote:
> On Wed, Jan 04, 2012 at 04:24:22PM +0100, Jan Holesovsky  
> wrote:
>>> One more review needed, I'd love to get Thorsten's input of course.
>>
>> For the late feature - the ideal would be cross-affiliation, if
>> possible :-)  So if anybody non-SUSE can have a look, it would be great
>> - Caolan?  Miklos?  Ivan?  Anybody else?
>>
>> As explained earlier, this makes many non-working pptx presentations
>> suddenly working; and we will see even more over the lifetime of 3.5.
>> The patch itself is really short & self-contained, so I think no real
>> Impress or PPTX knowledge is necessary for the approval.  And we will
>> still have enough time to revert it in case of trouble if it gets into
>> Beta3.
> 
> Hi Kendy,
> 
> Sure, from a user's perspective, I agree to have it in 3.5, it can't
> make the situation worse. ;-)
> 
> Miklos

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Andrew Douglas Pitonyak]

For what it is worth, I would provide an argument so that you could 
enable or disable the generation. I don't really care what the default 
is, as long as I can enable or disable doxygen.


What can become easy hacks, though, is to clean up the warnings 
doxygen still produces ...


That seems like a good way to get someone involved.

--
Andrew Pitonyak
My Macro Document: http://www.pitonyak.org/AndrewMacro.odt
Info:  http://www.pitonyak.org/oo.php

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

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Improved copying between different calc documents

[Markus Mohrhard]

Hello Eike,

> We talked about that before Christmas (actually I thought that would be
> already in 3-5 now), and yes, this feature makes sense when copying
> between documents.

After I had to enable the external references to unsaved documents
through direct input I was not sure if it is not a new feature.

> So, approval from me, but apparently at least some
> quirk pointed out by Cor lately needs to be solved.
>

I think I finally have an idea how Cor managed to create that file. I
think we are still missing one warning message if you close an unsaved
file which is still referenced by another file. At this moment this
ref is invalid. I'll adda warning message in this case and will insert
Kohei's suggestions.

Thanks for your reviews.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] minor idl fixes

[Tomas Hlavaty]

Hi Michael,

> Until then I'll reverting the patch, in case I forget later; we
> wouldn't like to accidentally license your code against your wishes.

sorry for inconvenience, I have sent the License statement email to the
mailing list so please feel free to reapply the patch.

Also, I attach a few more patches.

Thank you,

Tomas

>From da9fb395502bb2066db60dd4e69a6c69a5d8bb8f Mon Sep 17 00:00:00 2001
From: Tomas Hlavaty 
Date: Thu, 15 Dec 2011 23:57:19 +0100
Subject: [PATCH 1/7] fix broken ifndef/define guard in idl

---
 .../star/frame/DoubleInitializationException.idl   |4 ++--
 .../sun/star/frame/IllegalArgumentIOException.idl  |4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

diff --git a/offapi/com/sun/star/frame/DoubleInitializationException.idl b/offapi/com/sun/star/frame/DoubleInitializationException.idl
index d225770..f716e3e 100644
--- a/offapi/com/sun/star/frame/DoubleInitializationException.idl
+++ b/offapi/com/sun/star/frame/DoubleInitializationException.idl
@@ -24,8 +24,8 @@
  * for a copy of the LGPLv3 License.
  *
  /
-#ifndef __com_sun_star_frame_TerminationVetoException_idl__
-#define __com_sun_star_frame_TerminationVetoException_idl__
+#ifndef __com_sun_star_frame_DoubleInitializationException_idl__
+#define __com_sun_star_frame_DoubleInitializationException_idl__
 
 #include 
 
diff --git a/offapi/com/sun/star/frame/IllegalArgumentIOException.idl b/offapi/com/sun/star/frame/IllegalArgumentIOException.idl
index a0bef81..9c0ccd8 100644
--- a/offapi/com/sun/star/frame/IllegalArgumentIOException.idl
+++ b/offapi/com/sun/star/frame/IllegalArgumentIOException.idl
@@ -24,8 +24,8 @@
  * for a copy of the LGPLv3 License.
  *
  /
-#ifndef __com_sun_star_frame_TerminationVetoException_idl__
-#define __com_sun_star_frame_TerminationVetoException_idl__
+#ifndef __com_sun_star_frame_IllegalArgumentIOException_idl__
+#define __com_sun_star_frame_IllegalArgumentIOException_idl__
 
 #include 
 
-- 
1.7.0.4

>From 2172e7435d5b067e1bf25c8906852c74a1c3322c Mon Sep 17 00:00:00 2001
From: Tomas Hlavaty 
Date: Sat, 17 Dec 2011 17:42:58 +0100
Subject: [PATCH 2/7] fix package in idl

---
 offapi/com/sun/star/chart/XAxisSupplier.idl |6 +++---
 1 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/offapi/com/sun/star/chart/XAxisSupplier.idl b/offapi/com/sun/star/chart/XAxisSupplier.idl
index f87d245..cfb42b6 100644
--- a/offapi/com/sun/star/chart/XAxisSupplier.idl
+++ b/offapi/com/sun/star/chart/XAxisSupplier.idl
@@ -27,7 +27,7 @@
 #ifndef com_sun_star_chart_XAxisSupplier_idl
 #define com_sun_star_chart_XAxisSupplier_idl
 
-#include 
+#include 
 
 module com {  module sun {  module star {  module chart {
 
@@ -43,7 +43,7 @@ interface XAxisSupplier : ::com::sun::star::uno::XInterface
 @param nDimensionIndex
 Parameter nDimensionIndex says wether it is a x, y or z-axis (0 for x).
  */
-com::sun::star::chart::XAxis getAxis( [in] long nDimensionIndex );
+com::sun::star::chart2::XAxis getAxis( [in] long nDimensionIndex );
 
 /** @returns
 the secondary axis of the specified dimension. The returned object supports service ChartAxis.
@@ -51,7 +51,7 @@ interface XAxisSupplier : ::com::sun::star::uno::XInterface
 @param nDimensionIndex
 Parameter nDimensionIndex says wether it is a x, y or z-axis (0 for x).
  */
-com::sun::star::chart::XAxis getSecondaryAxis( [in] long nDimensionIndex );
+com::sun::star::chart2::XAxis getSecondaryAxis( [in] long nDimensionIndex );
 };
 
 }; }; }; };
-- 
1.7.0.4

>From 287b3f61c9d4930c26a1235e0e2cd1c1ac3dded1 Mon Sep 17 00:00:00 2001
From: Tomas Hlavaty 
Date: Sat, 17 Dec 2011 17:45:13 +0100
Subject: [PATCH 3/7] idl fix package names

---
 offapi/com/sun/star/chart2/InterpretedData.idl |2 +-
 offapi/com/sun/star/chart2/ScaleData.idl   |2 +-
 offapi/com/sun/star/chart2/XChartDocument.idl  |2 +-
 offapi/com/sun/star/chart2/XChartTypeTemplate.idl  |4 ++--
 offapi/com/sun/star/chart2/XDataInterpreter.idl|4 ++--
 offapi/com/sun/star/chart2/XDiagram.idl|2 +-
 .../com/sun/star/chart2/XInternalDataProvider.idl  |2 +-
 .../com/sun/star/xml/sax/XFastSAXSerializable.idl  |8 
 offapi/com/sun/star/xml/sax/XSAXSerializable.idl   |4 ++--
 9 files changed, 15 insertions(+), 15 deletions(-)

diff --git a/offapi/com/sun/star/chart2/InterpretedData.idl b/offapi/com/sun/star/chart2/InterpretedData.idl
index 0eedd9c..281bac9 100644
--- a/offapi/com/sun/star/chart2/InterpretedData.idl
+++ b/offapi/com/sun/star/chart2/InterpretedData.idl
@@ -50,7 +50,7 @@ struct InterpretedData
 
 /**
  */
-data::XLabeledDataSequence  Categories;
+com::sun::star::chart2::data::XLabeledDataSequence  Categories;
 };
 
 } ; // chart2
diff --git a/offapi/com/s

Re: [Libreoffice] [PATCH] Replace ByteString with rtl::OString

[Chr. Rossmanith]

Am 02.01.2012 23:26, schrieb Eike Rathke:



  if( ! aToken.compareToAscii( "pdf=", 4 ) )
  {
  sal_Int32 nPos = 0;
  aDir = aToken.getToken( 1, '=', nPos );
-if( ! aDir.Len() )
-aDir = String( ByteString( getenv( "HOME" ) ), 
osl_getThreadTextEncoding() );
+if( aDir.isEmpty() )
+aDir = rtl::OUString( getenv( "HOME" ), 4, 
osl_getThreadTextEncoding() );

Trapped ;-)  The string isn't constructed of the word HOME, but the
content of the HOME environment variable instead, so assuming 4 here
isn't correct.

I've corrected that and checked for NULL.

Besides that, already the original code didn't deal with a possible NULL
pointer returned by getenv() in case a HOME variable wasn't set at all..
that should be handled instead of relying on some OUString (back then
String) magic, which it seems OUString does not silently do and would
crash instead if a length>0 was passed.

Otherwise the patch looks good.
I've finished genprnpsp.cxx and had to touch some related files as well 
(but didn't finish them yet). After a review and removing some 
unnecessary rtl:: prefices I'd like to push this modified patch.


Thank you for reviewing!

Christina
>From 807fc4c27fc8f8b80a1883dd0ef14f0b6749113b Mon Sep 17 00:00:00 2001
From: Christina Rossmanith 
Date: Mon, 2 Jan 2012 21:56:19 +0100
Subject: [PATCH] Replace (Byte)String with rtl::O(U)String

---
 vcl/generic/print/common_gfx.cxx  |   18 +++---
 vcl/generic/print/genprnpsp.cxx   |   42 
 vcl/generic/print/printerjob.cxx  |2 +-
 vcl/headless/svpprn.cxx   |2 +-
 vcl/inc/vcl/ppdparser.hxx |3 +-
 vcl/unx/generic/printer/ppdparser.cxx |4 +-
 6 files changed, 36 insertions(+), 35 deletions(-)

diff --git a/vcl/generic/print/common_gfx.cxx b/vcl/generic/print/common_gfx.cxx
index 43c2cbb..cc3ac02 100644
--- a/vcl/generic/print/common_gfx.cxx
+++ b/vcl/generic/print/common_gfx.cxx
@@ -1188,26 +1188,26 @@ PrinterGfx::DrawEPS( const Rectangle& rBoundingBox, void* pPtr, sal_uInt32 nSize
 // first search the BoundingBox of the EPS data
 SvMemoryStream aStream( pPtr, nSize, STREAM_READ );
 aStream.Seek( STREAM_SEEK_TO_BEGIN );
-ByteString aLine;
+rtl::OString aLine;
 
 rtl::OString aDocTitle;
 double fLeft = 0, fRight = 0, fTop = 0, fBottom = 0;
 bool bEndComments = false;
 while( ! aStream.IsEof()
&& ( ( fLeft == 0 && fRight == 0 && fTop == 0 && fBottom == 0 ) ||
-( aDocTitle.getLength() == 0 && bEndComments == false ) )
+( aDocTitle.isEmpty() && bEndComments == false ) )
)
 {
 aStream.ReadLine( aLine );
-if( aLine.Len() > 1 && aLine.GetChar( 0 ) == '%' )
+if( aLine.getLength() > 1 && aLine[0] == '%' )
 {
-char cChar = aLine.GetChar(1);
+char cChar = aLine[1];
 if( cChar == '%' )
 {
-if( aLine.CompareIgnoreCaseToAscii( "%%BoundingBox:", 14 ) == COMPARE_EQUAL )
+if( comphelper::string::matchIgnoreAsciiCaseL( aLine, "%%BoundingBox:", 14 ) == sal_True )
 {
 aLine = WhitespaceToSpace( comphelper::string::getToken(aLine, 1, ':') );
-if( aLine.Len() && aLine.Search( "atend" ) == STRING_NOTFOUND )
+if( !aLine.isEmpty() && aLine.indexOf( "atend" ) < 0 )
 {
 fLeft   = StringToDouble( GetCommandLineToken( 0, aLine ) );
 fBottom = StringToDouble( GetCommandLineToken( 1, aLine ) );
@@ -1215,9 +1215,9 @@ PrinterGfx::DrawEPS( const Rectangle& rBoundingBox, void* pPtr, sal_uInt32 nSize
 fTop= StringToDouble( GetCommandLineToken( 3, aLine ) );
 }
 }
-else if( aLine.CompareIgnoreCaseToAscii( "%%Title:", 8 ) == COMPARE_EQUAL )
-aDocTitle = WhitespaceToSpace( aLine.Copy( 8 ) );
-else if( aLine.CompareIgnoreCaseToAscii( "%%EndComments", 13 ) == COMPARE_EQUAL )
+else if( comphelper::string::matchIgnoreAsciiCaseL( aLine, "%%Title:", 8 ) == sal_True )
+aDocTitle = WhitespaceToSpace( aLine.copy( 0, 8 ) );
+else if( comphelper::string::matchIgnoreAsciiCaseL( aLine, "%%EndComments", 13 ) == sal_True )
 bEndComments = true;
 }
 else if( cChar == ' ' || cChar == '\t' || cChar == '\r' || cChar == '\n' )
diff --git a/vcl/generic/print/genprnpsp.cxx b/vcl/generic/print/genprnpsp.cxx
index 23a7bf5..25b21d9 100644
--- a/vcl/generic/print/genprnpsp.cxx
+++ b/vcl/generic/print/genprnpsp.cxx
@@ -47,6 +47,7 @@
 #endif
 
 #include "rtl/ustring.hxx"
+#include "comphelper/string.hxx"
 
 #include "osl/module.h"
 
@@ -86,19 +87,19 @@ typedef int(*faxFunction)(String&);
 static faxFuncti

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Alex Thurgood]

Le 04/01/2012 21:36, Stephan Bergmann a écrit :

Hi Stephan,


On 01/04/2012 06:40 PM, Alexander Thurgood wrote:

elsewhere then. The question is : where ?


Alexander, if you have experience with depends.exe, you could run 
soffice.exe from within depends.exe (be sure to check the option to 
also trace sub-processes) and see why it fails to load mysqlc.uno.dll.


Unfortunately, I'm  not on Windows. I can only build on Linux and Mac. 
Am looking to test with gdb on Linux to see what is being called (or 
not) as the case may be.


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

[Libreoffice] License statement

[Tomas Hlavaty]

Hi,

all my current and past contributions made to the LibreOffice project
are done under the LGPLv3 / MPL1.1 or later version of these licenses.
Until further notice, all my future contributions to the LibreOffice
project are available under the LGPLv3 / MPL1.1 or later version of
these licenses.

Thank you,

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

Re: [Libreoffice] Introducing Grande Latte

[Jesús Corrius]

> On Wed, Jan 04, 2012 at 05:52:12PM +0100, Jesús Corrius wrote:
>> Grande Latte is a small and standalone ...
>
> nominated for the best l10n-troll of the year 2012. Also: We should have
> codenames for our LibreOffice releases.

You know I chose the name on purpose ;)

I guess the program is not bad for something I wrote in about 3 hours,
but it's just a prototype made with spaghetti code and if people start
to use it (for some reason) I will have to rewrite it in a more decent
form (especially if I ever want to find a job as a developer again).

-- 
Jesús Corrius 
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Importing smartArt from

[Miklos Vajna]

On Wed, Jan 04, 2012 at 04:24:22PM +0100, Jan Holesovsky  wrote:
> > One more review needed, I'd love to get Thorsten's input of course.
> 
> For the late feature - the ideal would be cross-affiliation, if
> possible :-)  So if anybody non-SUSE can have a look, it would be great
> - Caolan?  Miklos?  Ivan?  Anybody else?
> 
> As explained earlier, this makes many non-working pptx presentations
> suddenly working; and we will see even more over the lifetime of 3.5.
> The patch itself is really short & self-contained, so I think no real
> Impress or PPTX knowledge is necessary for the approval.  And we will
> still have enough time to revert it in case of trouble if it gets into
> Beta3.

Hi Kendy,

Sure, from a user's perspective, I agree to have it in 3.5, it can't
make the situation worse. ;-)

Miklos


pgpfA0LqjBVvV.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] feature/gbuild_extensions

[Michael Stahl]

On 04/01/12 08:54, Norbert Thiebaud wrote:
> So,
> I've run test on MacOSX... I ran  in a couple of issues.. I pushed
> patches for them on the branch
> 
> The last commit ( 381bed81e43d7fdd541115e0ab4213b0609e7ac0 )  is a
> hack and is probably wrong... but it get me to the end of the build +
> package
> 
> Norbert

i've just pushed a new gb_Library_Bundle function that sets a special
TARGETTYPE for this case (which is currently the only one we have in
LO), so you can try building it again, perhaps it'll even work :)

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

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Stephan Bergmann]

On 01/04/2012 06:40 PM, Alexander Thurgood wrote:

elsewhere then. The question is : where ?


Alexander, if you have experience with depends.exe, you could run 
soffice.exe from within depends.exe (be sure to check the option to also 
trace sub-processes) and see why it fails to load mysqlc.uno.dll.


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

Re: [Libreoffice] Introducing Grande Latte

[Stephan Bergmann]

On 01/04/2012 08:32 PM, Eike Rathke wrote:

On Wednesday, 2012-01-04 19:38:53 +0100, Bjoern Michaelsen wrote:

Grande Latte is a small and standalone ...


nominated for the best l10n-troll of the year 2012. Also: We should have
codenames for our LibreOffice releases.


Such as Latte Plus ? ;-)


Viagra Latte, in case it is not standalone.

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 07:28 PM, Hanno Meyer-Thurow wrote:

On Wed, 04 Jan 2012 17:01:32 +0100
Stephan Bergmann  wrote:

On 01/04/2012 04:35 PM, Hanno Meyer-Thurow wrote:

On Wed, 04 Jan 2012 15:57:58 +0100
Stephan Bergmann   wrote:

On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:

[...]

Doxygen won't be shipped with LO source like all the other 3rd parties?! 
Interesting ...


God no!


Hmm, quite strange to me but o well ...


We do not routinely include copies of build-time tools (compilers, make, 
ant, junit, ...).



And what do you mean with "generate the documentation anyway"?


If doxygen is there, you can use it anyway.


Which happens with the implicit --with-doxygen=yes (i.e., pick doxygen
from path) option.


Hmm, that kind of "evil" automagic I do not like ...


What's evil and automagic here?  Building LO has a new dependency, and 
there is an "emergency" switch (--without-doxygen) people can explicitly 
use to build without that prerequisite but with consequential loss of 
functionality.



"... an SDK w/o docu makes little sense ..."

Sure, like I said in the beginning, enforce the tool(s) for docu generation;
always(case: odk enabled) and for any SDK (C++, Java, Python, etc ...).
In the sense of keeping it simple.


Yeah, might work just as well to have no additional switch to disable 
doxygen, and instead let configure tell people to use --disable-odk if 
they have no doxygen.  I'm somewhat indifferent here.



"... controlling generation of HTML docu for any SDK stuff via 
--with-help
  does not look right ..."

Mind the "... or so?!" part of mine. Feel free to use '--with-developer-help' or
something like that. I just did not like the flag named after one tool that is 
subject
to change in future and surely for other SDKs.


A question of how much to abstract here, indeed.

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

Re: [Libreoffice] Introducing Grande Latte

[Eike Rathke]

Hi Bjoern,

On Wednesday, 2012-01-04 19:38:53 +0100, Bjoern Michaelsen wrote:

> > Grande Latte is a small and standalone ...
> 
> nominated for the best l10n-troll of the year 2012. Also: We should have
> codenames for our LibreOffice releases.

Such as Latte Plus ? ;-)

  Eike

-- 
LibreOffice Calc developer. Number formatter stricken i18n transpositionizer.
GnuPG key 0x293C05FD : 997A 4C60 CE41 0149 0DB3  9E96 2F1A D073 293C 05FD


pgp0KapxYmyOh.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [libreoffice-l10n] Re: [Libreoffice-qa] Online update service up & running

[Jan Holesovsky]

Hi Olivier,

On 2012-01-03 at 14:01 -0200, Olivier Hallot wrote:

> Actually, in a controlled environment, you may want to download the new
> version from a internal server, sparing your expensive external
> bandwidth. That is why I would like to have the download URL
> configureable in a XML file (e.g. main.xcd)

Ah, I see.  OK, so after getting rid of inst:buildid, I introduced an
inst:gitid in the update .xml so that one can control the updates via a
static .xml file.  The inst:buildid was unreliable, because of the way
we reset buildid with every version bump (3.3 -> 3.4 etc.)

So, what you want to do is to put the new version to your server (let's
say http://yourserver/newersion.msi) together with the following xml
(located at http://yourserver/update.xml):


http://update.libreoffice.org/description";>
  LibreOffice 3.something
  3.something
  123456-abcdef-1a2b3c-4d5e6f
  Windows
  x86

  http://yourserver/newersion.msi"; />


where inst:gitid is the value of 'buildid' in program/versionrc of
newersion.msi when installed.

The last step is to point your installations to this update xml, and
that is by modifying UpdateURL to value http://yourserver/update.xml in
program/versionrc of your installations.

Does this work for you?

inst:gitid will be available with Beta3, so the first update that will
be possible to handle this way will be Beta3 -> Beta4.  Please test &
report back in case of trouble.

Regards,
Kendy

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

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Improved copying between different calc documents

[Eike Rathke]

Hi Markus,

On Wednesday, 2012-01-04 17:04:48 +0100, Markus Mohrhard wrote:

> I propose this patch as a late feature.

We talked about that before Christmas (actually I thought that would be
already in 3-5 now), and yes, this feature makes sense when copying
between documents. So, approval from me, but apparently at least some
quirk pointed out by Cor lately needs to be solved.

  Eike

-- 
LibreOffice Calc developer. Number formatter stricken i18n transpositionizer.
GnuPG key 0x293C05FD : 997A 4C60 CE41 0149 0DB3  9E96 2F1A D073 293C 05FD


pgpqAFOD3Jy8L.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Norbert Thiebaud]

On Wed, Jan 4, 2012 at 12:28 PM, Hanno Meyer-Thurow  wrote:
> On Wed, 04 Jan 2012 17:01:32 +0100
> Stephan Bergmann  wrote:
>
>> On 01/04/2012 04:35 PM, Hanno Meyer-Thurow wrote:
>> > On Wed, 04 Jan 2012 15:57:58 +0100
>> > Stephan Bergmann  wrote:
>> >> On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:
>> [...]
>> > Doxygen won't be shipped with LO source like all the other 3rd parties?! 
>> > Interesting ...
>>
>> God no!
>
> Hmm, quite strange to me but o well ...

Not really. this is a purely build-time dependency, completely
optional, on something readily available. iow there is no reason
whatsoever to burden our build with a 'internal' one.

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

[Libreoffice] [Bug 35673] LibreOffice 3.4 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=35673

--- Comment #245 from John Manko  2012-01-04 11:27:36 PST 
---
Nominate Bug 42073: Writer input fields are not populated using the Java UNO
API

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] What are the criteria to nominate bugs as most annoying bug?

[Eike Rathke]

Hi Winfried,

On Wednesday, 2012-01-04 11:44:25 +0100, Winfried Donkers wrote:

> https://bugs.freedesktop.org/show_bug.cgi?id=37987
> https://bugs.freedesktop.org/show_bug.cgi?id=44383

See my comments there.

  Eike

-- 
LibreOffice Calc developer. Number formatter stricken i18n transpositionizer.
GnuPG key 0x293C05FD : 997A 4C60 CE41 0149 0DB3  9E96 2F1A D073 293C 05FD


pgpwcWrKKxDWB.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Improved copying between different calc documents

[Kohei Yoshida]

On Wed, 2012-01-04 at 13:44 -0500, Kohei Yoshida wrote:
> On Wed, 2012-01-04 at 12:33 -0500, Kohei Yoshida wrote:

> > As far as the behavior goes, you have my approval.
> 
> Actually I take that back.  There are some weird behaviors that I don't
> like.  I'm playing with it right now using the master build...
> 
> I'll report back later.

Ok.  I've played with it in-depth.  The weird behavior that I noticed
(that turns absolute references into external reference) happens only
when the original reference consists of all absolute references (sheet,
row and column) and references another sheet.  Once you understand why
it does that, it becomes logical.

This new behavior may be complex to the unsuspected at first.  But
copying formula with references between two separate documents itself is
in fact quite complex operation with many corner cases, so I think the
complexity of this new behavior is perhaps warranted.

I'll give my approval to this feature once again.

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Improved copying between different calc documents

[Kohei Yoshida]

On Wed, 2012-01-04 at 12:33 -0500, Kohei Yoshida wrote:
> On Wed, 2012-01-04 at 17:04 +0100, Markus Mohrhard wrote:
> > 
> > I propose this patch as a late feature. 
> 
> Just as a matter of record, I have already provided some feedback in his
> proposed patch on IRC.
> 
> As far as the behavior goes, you have my approval.

Actually I take that back.  There are some weird behaviors that I don't
like.  I'm playing with it right now using the master build...

I'll report back later.

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

Re: [Libreoffice] Introducing Grande Latte

[Bjoern Michaelsen]

Hi Jesus,

On Wed, Jan 04, 2012 at 05:52:12PM +0100, Jesús Corrius wrote:
> Grande Latte is a small and standalone ...

nominated for the best l10n-troll of the year 2012. Also: We should have
codenames for our LibreOffice releases.

Best,

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Hanno Meyer-Thurow]

On Wed, 04 Jan 2012 17:01:32 +0100
Stephan Bergmann  wrote:

> On 01/04/2012 04:35 PM, Hanno Meyer-Thurow wrote:
> > On Wed, 04 Jan 2012 15:57:58 +0100
> > Stephan Bergmann  wrote:
> >> On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:
> [...]
> > Doxygen won't be shipped with LO source like all the other 3rd parties?! 
> > Interesting ...
> 
> God no!

Hmm, quite strange to me but o well ...

> >> And what do you mean with "generate the documentation anyway"?
> >
> > If doxygen is there, you can use it anyway.
> 
> Which happens with the implicit --with-doxygen=yes (i.e., pick doxygen 
> from path) option.

Hmm, that kind of "evil" automagic I do not like ...

> >>> If the generation has to be configurable, chain it to --with-help or 
> >>> --with-helppack-integration or so?!
> >
> >> I'd say these are rather unrelated, so would prefer not to combine them.
> >
> > One kind of help for a developer is what doxygen generates I would say.
> 
> But it would be odd if generation of HTML docu for SDK C++ stuff were 
> controlled by --with-help while generation of HTML docu for SDK Java 
> stuff were not.  And an SDK w/o docu makes little sense, so generally 
> controlling generation of HTML docu for any SDK stuff via --with-help 
> does not look right, either.

... and ending in this mess. That should be fixed in one or another unified way.


"... an SDK w/o docu makes little sense ..."

Sure, like I said in the beginning, enforce the tool(s) for docu generation;
always(case: odk enabled) and for any SDK (C++, Java, Python, etc ...).
In the sense of keeping it simple.


"... controlling generation of HTML docu for any SDK stuff via 
--with-help
  does not look right ..."

Mind the "... or so?!" part of mine. Feel free to use '--with-developer-help' or
something like that. I just did not like the flag named after one tool that is 
subject
to change in future and surely for other SDKs.

I still do not like the extra flag. Well, it is just me ... ;)


Or leave it to pure automagic with configuration notice.

Or add that doxygen flag, just another funny flag.

Should the configure flags not be reduced and/or simplified?
I thought there was something like that on the list before ...


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

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Alexander Thurgood]

Le 04/01/12 18:02, Stephan Bergmann a écrit :

Hi Stephan,

> At least mysql-connector-ooo.oxt built in module myslqc apparently still
> uses active registration, so that should not be a reason to let it no
> longer work in old installations.  (Plus, passive component registration
> was already available in 3.4.)
> 

Well that's good to know at least, so the problem apparently lies
elsewhere then. The question is : where ?


Alex


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

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

Alex Thurgood  changed:

   What|Removed |Added

 Depends on||44075

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

--- Comment #73 from Alex Thurgood  2012-01-04 
09:35:25 PST ---
As per discussion with Michael Meeks, nominating bug 44075 - mysql connector
1.0.1 extension from Oracle no longer works with 3.5 and throws a UNO shared
object/component error.


Alex

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Alexander Thurgood]

Le 04/01/12 17:20, Michael Meeks a écrit :
>
>   So that sounds like a pretty serious bug; can you make a most-annoying
> out of it and provide all the relevant details: the old download, what
> versions it worked with etc. and comment on it for the "Most annoying"
> bug ?
>

There was a bug report here :

https://bugs.freedesktop.org/show_bug.cgi?id=44075

I had originally closed this as not our problem, because of my
experience with the builds of the connectors from master. I have now
opened this up again, and will add it to the most annoying list for 3.5.

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

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Improved copying between different calc documents

[Kohei Yoshida]

On Wed, 2012-01-04 at 17:04 +0100, Markus Mohrhard wrote:
> 
> I propose this patch as a late feature. 

Just as a matter of record, I have already provided some feedback in his
proposed patch on IRC.

As far as the behavior goes, you have my approval.

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

--- Comment #72 from John Manko  2012-01-04 09:09:00 PST 
---
Nominate Bug 42073: Writer input fields are not populated using the Java UNO
API

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Stephan Bergmann]

On 01/04/2012 05:42 PM, Alexander Thurgood wrote:

I can't say when exactly this happened, but can confirm that those
connectors I have built from master only work with 3.5 and other current
master builds.

I was wondering whether this had anything to with some of the following
changes in the run up to master :
- version bump for the mysql library (Lionel's commits) ?
- component registration changes for the connectors ?

For a while, the connector I was building worked both on master and 3.4
builds (and also 3.3 builds). Now, at least on Linux, it only works on
3.5 builds. If I try and use it on a 3.3/3.4 installation, I get the uno
shared object error.


At least mysql-connector-ooo.oxt built in module myslqc apparently still 
uses active registration, so that should not be a reason to let it no 
longer work in old installations.  (Plus, passive component registration 
was already available in 3.4.)


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

Re: [Libreoffice] [PATCH] Fix for fdo43460 Part XXVIII getLength() to isEmpty()

[Tommy]

On Wed, 04 Jan 2012 10:45:26 +0100, Olivier Hallot  
 wrote:



just curious... how many parts are still missing?
you did already 28 of them


I think I am in the middle of the task, so perhaps there is about 60
parts in total. I do it by modules, to keep the mental sanity of the
reviewer and myself.

Feel free to join the team :-) .

Regards
- --
Olivier Hallot


first of all congrats for your effort
unfortunately I'm just an user without any coding knowlegde
the only thing I can do is to keep up to date the Italian OOo/LibO  
discussion group
(  
https://groups.google.com/group/it-alt.comp.software.openoffice/topics?hl=it&gvc=2  
)

with news about LibO development

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

[Libreoffice] Introducing Grande Latte

[Jesús Corrius]

Hello everybody,

Grande Latte is a small and standalone Windows application that allows
you to modify the LibreOffice installer to select what UI languages
are installed during a default installation.

https://github.com/downloads/jcorrius/grande_latte/grande_latte.zip

Please take into account that as it physically modifies the file it
will invalidate the signature if the MSI is signed.

Have a great 2012.

-- 
Jesús Corrius
Deputy of the Board of Directors
The Document Foundation
Mobile: +34 661 11 38 26
Skype: jcorrius | Twitter: @jcorrius
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

Bug 37361 depends on bug 44093, which changed state.

Bug 44093 Summary: Forms open in printer view. They should open in web-view.
https://bugs.freedesktop.org/show_bug.cgi?id=44093

   What|Old Value   |New Value

 Resolution||DUPLICATE
 Status|NEW |RESOLVED

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Alexander Thurgood]

Hi Michael,


Le 04/01/12 17:20, Michael Meeks a écrit :
> Oh - that is very odd. It should be a UNO component which "just works"
> and we should be backwards-compatible with it. So that sounds like a
> pretty serious bug; can you make a most-annoying out of it and provide
> all the relevant details: the old download, what versions it worked
> with etc. and comment on it for the "Most annoying" bug ? Thanks :-)
> Michael. 

This is actually also the case for the mysql connectors that I have
built from master and made available via the extensions website :
- Linux 32bit (Ubuntu Oneiric)
- Mac OSX

I can't say when exactly this happened, but can confirm that those
connectors I have built from master only work with 3.5 and other current
master builds.

I was wondering whether this had anything to with some of the following
changes in the run up to master :
- version bump for the mysql library (Lionel's commits) ?
- component registration changes for the connectors ?

For a while, the connector I was building worked both on master and 3.4
builds (and also 3.3 builds). Now, at least on Linux, it only works on
3.5 builds. If I try and use it on a 3.3/3.4 installation, I get the uno
shared object error.

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

Re: [Libreoffice] ooo_custom_images to icon-themes, was: Re: Do we support classic and industrial theme?

[Michael Meeks]

On Tue, 2011-12-27 at 13:54 +0100, Stefan Knorr (Astron) wrote:
> FYI, this just went live (well, the folder name is icon-themes). I
> hope this doesn't cause too much mayhem...

Oooh :-) nice - thanks.

> Everything else mentioned in the old thread is still on the todo list.

Great; so - moving default_images into there and making it 'galaxy'
would be a nice cleanup I guess; then again it may need quite some
tweaking of scripts that hard-code it's path I suppose; particularly
(IIRC) 'rsc' checks to make sure that the icons actually exist in there
and would need a path.

Thanks for the cleanup though :-)

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 05:11 PM, Lubos Lunak wrote:

  Ah, so the URL above is still generated the old way?


Yes.


  But doxygen can generate such an overview too - see
http://api.kde.org/4.x-api/kdelibs-apidocs/kdeui/html/globals.html (it is
the 'File Members' link on the left).


Ah, good to know.  (It's included in what I generated with doxygen, too, 
just apparently a little non-obvious to find.)


Stephan

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

Re: [Libreoffice] Cherry pick request: Fix always-false condition of FullCommentURL

[Cedric Bosdonnat]

Hi Harri,

On Wed, 2012-01-04 at 18:23 +0200, Harri Pitkänen wrote:
> Release notes for 3.5 mention a new feature for grammar checker
> developers, the FullCommentURL property. I tried to use it but could not
> get any visible results with 3.5beta2. Then I looked at git history and
> noticed that the following fix from László Németh seems to be present only
> on master:
> 
>   
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=e2812dd41916cb90c28782b855a100edb4b681b0
> 
> I think it would make sense to cherry pick that to libreoffice-3-5.

Sure it makes sense. I just cherry-picked it.

Thanks for spotting that one ;)

Regards,

--
Cedric

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

[Libreoffice] Bug 42073 - Writer input fields are not populated using the Java UNO API

[Grover Blue]

Does anyone have insight on the following bug ( "Writer input fields are
not populated using the Java UNO API" )?

https://bugs.freedesktop.org/show_bug.cgi?id=42073

I can't process any documents with LO 3.4.3 or greater.  Here is the
summary:

I've been using OpenOffice and LibreOffice to batch process documents using the
Java UNO interface.  Upon updating to LibreOffice 3.4.3, LibreOffice is
ignoring the Java UNO calls to populate the input fields with data.  Running
the same code with LibreOffice 3.3 works flawlessly.  Below is sample code,
which I've stepped through in debug mode, and everything is running correctly.
The fields are found, but calling xPropertySet.setPropertyValue("Content",
newValue) leaves the document unchanged. Updating table cells with 3.4.3,
however, is working as expected.  Has something changed with the API that is
not documented?
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] Cherry pick request: Fix always-false condition of FullCommentURL

[Harri Pitkänen]

Release notes for 3.5 mention a new feature for grammar checker
developers, the FullCommentURL property. I tried to use it but could not
get any visible results with 3.5beta2. Then I looked at git history and
noticed that the following fix from László Németh seems to be present only
on master:

  
http://cgit.freedesktop.org/libreoffice/core/commit/?id=e2812dd41916cb90c28782b855a100edb4b681b0

I think it would make sense to cherry pick that to libreoffice-3-5.

Harri

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

Re: [Libreoffice] Old Oracle Mysql Connector no longer works with 3.5 on Windows - looking for a regular build solution ?

[Michael Meeks]

On Wed, 2011-12-21 at 19:15 +0100, Alexander Thurgood wrote:
> I've been informed by Fernand Vanrie (who has been involved in getting
> Base to do a whole lot of stuff that I never thought possible for a long
> time) that the old mysql connector extension supplied by Oracle no
> longer functions with 3.5 on Windows, whereas it did still continue to
> work on 3.3.x and 3.4.x.

Oh -  that is very odd. It should be a UNO component which "just works"
and we should be backwards-compatible with it.

So that sounds like a pretty serious bug; can you make a most-annoying
out of it and provide all the relevant details: the old download, what
versions it worked with etc. and comment on it for the "Most annoying"
bug ?

Thanks :-)

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Lubos Lunak]

On Wednesday 04 of January 2012, Stephan Bergmann wrote:
> On 01/04/2012 04:15 PM, Lubos Lunak wrote:
> > On Wednesday 04 of January 2012, Stephan Bergmann wrote:
> >> (One downside of doxygen appears to be that
> >> it does not generate an overview of all the non-class-member entities,
> >> like free functions; it only generates overviews of all the namespaces,
> >> classes, and files.)
> >
> >   I'm not sure what you mean here - the overviews do include also global
> > non-namespaced objects, search
> > http://api.libreoffice.org/docs/cpp/ref/names/index.html for
> > e.g. 'typereg_writer_destroy'.
>
> But doxygen does not produce such an overview.

 Ah, so the URL above is still generated the old way?

 But doxygen can generate such an overview too - see 
http://api.kde.org/4.x-api/kdelibs-apidocs/kdeui/html/globals.html (it is 
the 'File Members' link on the left).

-- 
 Lubos Lunak
 l.lu...@suse.cz
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PUSHED] Re: [PATCH] fix EasyHack FDO42454, remove code associated with unused icons

[Michael Meeks]

On Thu, 2011-12-29 at 20:28 +0100, Jan Holesovsky wrote:
> But I did not touch these (ie. did not use this part of your patch).
> There is something more rotten going on - as you can see, the images
> in .src use .bmp (!) suffix, not .png, so I suspect there is even more
> unused code there, probably.

Right - there is an automagic mangling from foo.bmp -> foo.png when we
compile the .src files; -ideally- we would have a nice EasyHack patch to
update all the .src files to switch the file-extensions over - at least
for the images that are not one of the 3 last remaining .bmp files in
default_images ;-)

All the best,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

Petr Mladek  changed:

   What|Removed |Added

 Depends on||44093

--- Comment #71 from Petr Mladek  2012-01-04 08:04:49 PST ---
Nominate 44093: Forms open in printer view. They should open in web-view

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 04:35 PM, Hanno Meyer-Thurow wrote:

On Wed, 04 Jan 2012 15:57:58 +0100
Stephan Bergmann  wrote:

On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:

[...]

Doxygen won't be shipped with LO source like all the other 3rd parties?! 
Interesting ...


God no!


And what do you mean with "generate the documentation anyway"?


If doxygen is there, you can use it anyway.


Which happens with the implicit --with-doxygen=yes (i.e., pick doxygen 
from path) option.



If the generation has to be configurable, chain it to --with-help or 
--with-helppack-integration or so?!



I'd say these are rather unrelated, so would prefer not to combine them.


One kind of help for a developer is what doxygen generates I would say.


But it would be odd if generation of HTML docu for SDK C++ stuff were 
controlled by --with-help while generation of HTML docu for SDK Java 
stuff were not.  And an SDK w/o docu makes little sense, so generally 
controlling generation of HTML docu for any SDK stuff via --with-help 
does not look right, either.


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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 04:17 PM, Bjoern Michaelsen wrote:

On Wed, Jan 04, 2012 at 03:30:03PM +0100, Stephan Bergmann wrote:

With autodoc no longer used to document C++, the next step would be
to either replace its use to generate IDL docu as well (and
completely remove autodoc), or at least remove the C++-specific
autodoc code.  The SDK includes the autodoc executable, but I guess
it would be OK to change that...


Would you consider that -- or parts of that -- easyhackifiable?


Given the style of the autodoc code, removing just the C++-specific code 
from it is probably not an easy hack (but rather a painful one). 
Anyway, the hopes are to remove autodoc wholesale.


What can become easy hacks, though, is to clean up the warnings doxygen 
still produces (about unknown @ commands, mismatches between parameter 
names in documentation and declaration, etc.), to clean up the generated 
output (e.g., consistency in using full stops or not at end of 
documentation phrases, add missing documentation, use proper markup), 
and to tweak the output to make it look more like the existing 
.


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

Re: [Libreoffice] [PATCH] Fix for bug fdo#44110, New Math symbols issues in Windows

[Michael Meeks]

On Tue, 2012-01-03 at 12:12 -0200, Olivier Hallot wrote:
> WRT to this patch, there are two files; the binary file opens.ttf
> and the text file OpenSymbol.sfd.

Right - the original / authoritative version should be the .sdf file
which (IIRC) as well as being 'diff -u' able - so we can see what
changed, is/was also more descriptive & complete :-)

> I used fontforge tool under ubuntu to draw the new characters.

Great - that's the right approach.

> The sfd file is generated based on the ttf binaries.

So - IMHO it should be the other way around ;-) at least, when I last
edited that font; if we've fluffed that up I suggest we load the
original sfd, copy in the new glyphs from the TTF, save it as sfd, and
re-export as ttf ;-)

Of course - it'd be nice to do that compilation to TTF during the
build, but ... it's not so easy without fontforge installed.

HTH,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] How to write an import filter?

[Michael Meeks]

Hi mete0r,

On Wed, 2012-01-04 at 01:29 +0900, mete0r wrote:
> I'm interested in learning how to write an import filter for some legacy
> document formats. I tried to figure out by myself but failed. Could you
> point some documents / wiki pages, or even some files in the source code
> repository?

Lovely ! :-) so, the best filters to base your code on are probably
Fridrich's - I imagine the libvisio code is (by now) the most recent and
prolly best place to look - at least for a drawing filter. Otherwise
'libwpd' for a document one. Sadly Fridrich's slides on "writing import
filters for LibreOffice" didn't make it to here:

http://wiki.documentfoundation.org/Marketing/SpecialEvents/LibreOffice_Conference_2011_Paris#Media_Used_For_Presentations_At_The_LibreOffice_Conference_2011

> Especially I need to know how to register an import filter to the
> libreoffice installation and how the core invokes a filter, i.e. the
> entry points of the filters, to start some hacking.

Sure - so that stuff is non-obvious at least initially; so I would tend
to just gut an existing filter - and leave it up to someone who wants to
merge your patch :-) and/or perhaps Fridrich (having been through it so
many times) has document it with a HOWTO by now ;-)

Anyhow - excited to have you involved, what file format are you
particularly interested in ?

All the best,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 04:11 PM, Thorsten Behrens wrote:

Stephan Bergmann wrote:

With autodoc no longer used to document C++, the next step would be
to either replace its use to generate IDL docu as well (and
completely remove autodoc), or at least remove the C++-specific
autodoc code.  The SDK includes the autodoc executable, but I guess
it would be OK to change that...


Quite. It seems the folks over at Mozilla are using doxygen for
their IDL stuff, so retiring autodoc for that seems desirable, too.
I had once looked into the feature gap with Frank Schönheit, let me
try&  remember - presumably it's only a bit of @since mapping, no?


Hopefully, yes (I did not look into it at all yet).

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 04:15 PM, Lubos Lunak wrote:

On Wednesday 04 of January 2012, Stephan Bergmann wrote:

(One downside of doxygen appears to be that
it does not generate an overview of all the non-class-member entities,
like free functions; it only generates overviews of all the namespaces,
classes, and files.)


  I'm not sure what you mean here - the overviews do include also global
non-namespaced objects, search
http://api.libreoffice.org/docs/cpp/ref/names/index.html for
e.g. 'typereg_writer_destroy'.


But doxygen does not produce such an overview.

Stephan

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

Re: [Libreoffice] What are the criteria to nominate bugs as most annoying bug?

[Michael Meeks]

Hi Winfried,

So - this best belongs on the QA list (which I've CC'd):

On Wed, 2012-01-04 at 11:44 +0100, Winfried Donkers wrote:
> There are some bugs in calc that damage formulas/conditional formats
> of cells. As this goes further than just being a nuisance I would like
> to nominate them for solving with high(er) priority.
> 
> But I don't want to make things worse and I don't want to annoy people
> who are already working hard on improving LibreOffice. (The bugs are
> too complicated for me to hack)

So - the best way to increase their visibility is to suggest them on
the 'most annoying' tracker bug for 3.5:

https://bugs.freedesktop.org/show_bug.cgi?id=37361

With a description of why they're most annoying :-) if the QA / release
guys like them they'll get added as dependencies & at least more people
will see them.

Thanks !

Michael.

> These are the bugs: 

> https://bugs.freedesktop.org/show_bug.cgi?id=37987
> https://bugs.freedesktop.org/show_bug.cgi?id=44383


-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

[Libreoffice] [Bug 35673] LibreOffice 3.4 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=35673

Bug 35673 depends on bug 39659, which changed state.

Bug 39659 Summary: Java 1.7.0 not recognised
https://bugs.freedesktop.org/show_bug.cgi?id=39659

   What|Old Value   |New Value

 Resolution||FIXED
 Status|ASSIGNED|RESOLVED

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Hanno Meyer-Thurow]

On Wed, 04 Jan 2012 15:57:58 +0100
Stephan Bergmann  wrote:

> On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:
> > I would not like to see a new configure flag for doxygen, the odk flag 
> > suffices;
> > generate the documentation anyway. Otherwise it is always good to use 
> > standards.
> 
> Do you mean, those who do not have doxygen and don't want to or can't 
> install it should use --disable-odk (i.e., do not build SDK at all) 
> instead of --without-doxygen (i.e., build SDK but without C++ 
> documentation)?

Doxygen won't be shipped with LO source like all the other 3rd parties?! 
Interesting ...

> And what do you mean with "generate the documentation anyway"?

If doxygen is there, you can use it anyway.

> > If the generation has to be configurable, chain it to --with-help or 
> > --with-helppack-integration or so?!

> I'd say these are rather unrelated, so would prefer not to combine them.

One kind of help for a developer is what doxygen generates I would say.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Importing smartArt from

[Jan Holesovsky]

Hi Michael,

On 2012-01-04 at 13:54 +, Michael Meeks wrote:

> > So you have my go-ahead, the code looks nice, and self-contained
> 
>   Mine too - it looks extremely short & sweet; and without it we just
> have a huge white hole so ... ;-)
> 
>   One more review needed, I'd love to get Thorsten's input of course.

For the late feature - the ideal would be cross-affiliation, if
possible :-)  So if anybody non-SUSE can have a look, it would be great
- Caolan?  Miklos?  Ivan?  Anybody else?

As explained earlier, this makes many non-working pptx presentations
suddenly working; and we will see even more over the lifetime of 3.5.
The patch itself is really short & self-contained, so I think no real
Impress or PPTX knowledge is necessary for the approval.  And we will
still have enough time to revert it in case of trouble if it gets into
Beta3.

Thank you,
Kendy

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Bjoern Michaelsen]

On Wed, Jan 04, 2012 at 03:30:03PM +0100, Stephan Bergmann wrote:
> So, I tried to switch to doxygen when generating the C/C++ header
> documentation in odk/pack/gendocu. 
> And it worked quite well. 
> If there are no objections, I will commit this to master later this week.

Yes, yes, yes! great!

> The build then has an additional dependency on the doxygen
> executable as a prerequisite.  This can be controlled via
> --with-doxygen, where an explicit --without-doxygen disables
> generation of the C/C++ documentation in odk (and --disable-odk
> skips the doxygen check completely in configure).  Tinderboxes that
> have no doxygen installed would need an explicit --without-doxygen
> to avoid configure failures. (Reportedly, doxygen works well on Mac
> OS X and Windows, but I did not check that.)
> 
> With autodoc no longer used to document C++, the next step would be
> to either replace its use to generate IDL docu as well (and
> completely remove autodoc), or at least remove the C++-specific
> autodoc code.  The SDK includes the autodoc executable, but I guess
> it would be OK to change that...

Would you consider that -- or parts of that -- easyhackifiable?

Best,

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

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Thorsten Behrens]

Stephan Bergmann wrote:
> And it worked quite well.
>
Yay! \o/

> The build then has an additional dependency on the doxygen
> executable as a prerequisite.  This can be controlled via
> --with-doxygen, where an explicit --without-doxygen disables
> generation of the C/C++ documentation in odk (and --disable-odk
> skips the doxygen check completely in configure).
>
Sounds fine to me.

> With autodoc no longer used to document C++, the next step would be
> to either replace its use to generate IDL docu as well (and
> completely remove autodoc), or at least remove the C++-specific
> autodoc code.  The SDK includes the autodoc executable, but I guess
> it would be OK to change that...
> 
Quite. It seems the folks over at Mozilla are using doxygen for
their IDL stuff, so retiring autodoc for that seems desirable, too.
I had once looked into the feature gap with Frank Schönheit, let me
try & remember - presumably it's only a bit of @since mapping, no?

Cheers,

-- Thorsten


pgpknpERiO5PZ.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Lubos Lunak]

On Wednesday 04 of January 2012, Stephan Bergmann wrote:
> (One downside of doxygen appears to be that
> it does not generate an overview of all the non-class-member entities,
> like free functions; it only generates overviews of all the namespaces,
> classes, and files.)

 I'm not sure what you mean here - the overviews do include also global 
non-namespaced objects, search 
http://api.libreoffice.org/docs/cpp/ref/names/index.html for 
e.g. 'typereg_writer_destroy'.

-- 
 Lubos Lunak
 l.lu...@suse.cz
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PUSHED] [PATCH] tinderbox status: email on success, email authors

[Thorsten Behrens]

> Will you push that to the buildbot repo, please?
> 
Done (the 0001- one).

Cheers,

-- Thorsten


pgpH2RI5FhVQf.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Hanno Meyer-Thurow]

On Wed, 04 Jan 2012 15:30:03 +0100
Stephan Bergmann  wrote:

> Hi all,

Hi Stephan,

> 
>  
> "Temporary hack to work around autodoc bug" made it even more evident 
> that our home-brewed autodoc is not up to its task and should be replaced.
> 
> So, I tried to switch to doxygen when generating the C/C++ header 
> documentation in odk/pack/gendocu.  (There are apparently various 
> alternatives to doxygen out there; I did not bother to look at any of 
> them, given that doxygen is something of an "industry standard."  Speak 
> up if you know of a superior alternative.)

> The build then has an additional dependency on the doxygen executable as 
> a prerequisite.  This can be controlled via --with-doxygen, where an 
> explicit --without-doxygen disables generation of the C/C++ 
> documentation in odk (and --disable-odk skips the doxygen check 
> completely in configure).  Tinderboxes that have no doxygen installed 
> would need an explicit --without-doxygen to avoid configure failures. 
> (Reportedly, doxygen works well on Mac OS X and Windows, but I did not 
> check that.)

I would not like to see a new configure flag for doxygen, the odk flag suffices;
generate the documentation anyway. Otherwise it is always good to use standards.

If the generation has to be configurable, chain it to --with-help or 
--with-helppack-integration or so?!


Regards,
Hanno
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

On 01/04/2012 03:58 PM, Hanno Meyer-Thurow wrote:

I would not like to see a new configure flag for doxygen, the odk flag suffices;
generate the documentation anyway. Otherwise it is always good to use standards.


Do you mean, those who do not have doxygen and don't want to or can't 
install it should use --disable-odk (i.e., do not build SDK at all) 
instead of --without-doxygen (i.e., build SDK but without C++ 
documentation)?


And what do you mean with "generate the documentation anyway"?


If the generation has to be configurable, chain it to --with-help or 
--with-helppack-integration or so?!


I'd say these are rather unrelated, so would prefer not to combine them.

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

Re: [Libreoffice] [PATCH][PUSHED] Fix for fdo43460 Part XXVIII getLength() to isEmpty()

[Michael Meeks]

On Wed, 2012-01-04 at 13:20 +0100, Lubos Lunak wrote:
>  After 28 patches and about so many still to come, would you consider getting 
> commit access? These getLength()->isEmpty() patches I've reviewed are 

I just asked Olivier if he wanted that; it seems the list wants that
too ;-) so ... please create a bug like this:

http://www.freedesktop.org/wiki/AccountRequests

And poke me with the number; account names with a clear association to
first/last-name much preferred ;-)

Thanks !

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

[Libreoffice] Doxygen for C/C++ URE Documentation

[Stephan Bergmann]

Hi all,

 
"Temporary hack to work around autodoc bug" made it even more evident 
that our home-brewed autodoc is not up to its task and should be replaced.


So, I tried to switch to doxygen when generating the C/C++ header 
documentation in odk/pack/gendocu.  (There are apparently various 
alternatives to doxygen out there; I did not bother to look at any of 
them, given that doxygen is something of an "industry standard."  Speak 
up if you know of a superior alternative.)


And it worked quite well.  The parsed headers need some fixing (e.g., 
there's no @HTML command in doxygen, @tplparam is just @tparam, and 
@internal does something rather different and unfortunately needs to be 
replaced with more verbose @cond INTERNAL ... @endcond constructs), but 
overall it looks like the generated HTML pages are comparable in content 
to the existing . 
Plus, the CPPUHELPER_DLLPUBLIC problem from the above commit is a 
non-issue with doxygen...  (One downside of doxygen appears to be that 
it does not generate an overview of all the non-class-member entities, 
like free functions; it only generates overviews of all the namespaces, 
classes, and files.)


If there are no objections, I will commit this to master later this week.

The build then has an additional dependency on the doxygen executable as 
a prerequisite.  This can be controlled via --with-doxygen, where an 
explicit --without-doxygen disables generation of the C/C++ 
documentation in odk (and --disable-odk skips the doxygen check 
completely in configure).  Tinderboxes that have no doxygen installed 
would need an explicit --without-doxygen to avoid configure failures. 
(Reportedly, doxygen works well on Mac OS X and Windows, but I did not 
check that.)


With autodoc no longer used to document C++, the next step would be to 
either replace its use to generate IDL docu as well (and completely 
remove autodoc), or at least remove the C++-specific autodoc code.  The 
SDK includes the autodoc executable, but I guess it would be OK to 
change that...


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

Re: [Libreoffice] merging libraries: problem with multiple definitions

[Michael Meeks]

Hi Matus,

On Wed, 2011-12-28 at 23:30 +0100, Matúš Kukan wrote:
> I was again working on merging libraries from tail_build:

Cool ! :-)

> avmedia canvastools cppcanvas deploymentmisc drawinglayer editeng
> filterconfig fwe fwi fwk lng package2 sb sfx sofficeapp spl svx
> svxcore uui xo

Wow - can we really merge all that lot together ? if so - we're really
starting to get somewhere I think :-)

> But there is problem with multiple definitions:
> 
> nCreateSlots and nConvertSlots are references in:
> svx/source/form/fmshimp.o
> framework/source/uielement/controlmenucontroller.o

So - I think here we just add a 'static' keyword in front of those
(clearly module local) variables.

> aSfxUInt32Item_Impl aSfxBoolItem_Impl aSfxVoidItem_Impl
> aSfxUInt16Item_Impl aSfxStringItem_Impl are referenced in:
> svx/source/form/typemap.o
> sfx2/source/appl/appbas.o

Gosh - that is more unpleasant :-) I guess we prolly need some magic
inside the svidl .sdi compiler (urgh) and svx/sdi/svxslots.hxx to ensure
that we don't emit anything except externs for:

extern SfxType3 aSvxEscapementItem_Impl;
#ifdef SFX_TYPEMAP
SfxType3 aSvxEscapementItem_Impl = 

with an #ifdef SVX_TYPEMAP or something there ? When we had a similar
problem in Bonobo land (years ago) we did something like:

#ifndef __Bonobo_COMPILATION
#ifdef __ORBIT_IDL__
%{
#pragma include_defs bonobo/Bonobo.h
%}
#pragma inhibit push
#endif
#endif
...
#ifndef __Bonobo_COMPILATION
#ifdef __ORBIT_IDL__
#pragma inhibit pop
#endif
#endif

Which then causes plenty of grief in various ways from different
pre-processors: the idea being that this would get through and we could
adapt the #include of that specific IDL file into either a #include<> or
a set of IDL decls depending who compiled it.

> My guess is they are from svxslots.hxx and sfxslots.hxx which are
> somehow created from s.xslots.sdi ?
> What can we do about this ?

Bit of an annoyance really :-) Quite possibly:

idl/source/objects/types.cxx-
idl/source/objects/types.cxx-// write the implementation part
idl/source/objects/types.cxx:rOutStm << "#ifdef SFX_TYPEMAP" << endl
idl/source/objects/types.cxx-<< aTypeName.GetBuffer() <<
aVarName.GetBuffer()
idl/source/objects/types.cxx-<< " = " << endl;

Could be adapted to use a name based on the mangled name of the SDI
that is included; so:

#ifdef XOITEMS_SDI_TYPEMAP
...
#ifdef SVXITEMS_SDI_TYPEMAP
...
#ifdef FMSLOTS_SDI_TYPEMAP
...

And then we could choose in either of the files you mention - which of
these we really want to generate.

That is (of course) assuming we can get the path name into the IDL
compiler's types.cxx ;-) prolly it is hanging around somewhere.

Does that make some sort of sense ? Really excited to have you working
on this - there should be a lot of scope for improvement from graunching
all of that together and LTO'ing it :-)

All the best,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

[Libreoffice] [PATCH] Fix for fdo43460 Part XXII getLength() to isEmpty()

[Gustavo Pacheco]

  Hi!

  Partial fix for Easy Hack FDO43460 (part XXII) for review.

  Part XXII
  modules
  jvmfwk, l10ntools, lingucomponent

  Best regards
  Gustavo Pacheco.
From abf405bd1ab4595f02dab4eb8d9c9df515748542 Mon Sep 17 00:00:00 2001
From: Gustavo Buzzatti Pacheco 
Date: Wed, 4 Jan 2012 11:38:41 -0200
Subject: [PATCH] Fix for fdo43460 Part XXII getLength() to isEmpty()

Part XXII
modules
jvmfwk, l10ntools, lingucomponent
---
 jvmfwk/plugins/sunmajor/pluginlib/gnujre.cxx   |8 ++--
 .../plugins/sunmajor/pluginlib/sunjavaplugin.cxx   |   30 +-
 jvmfwk/plugins/sunmajor/pluginlib/util.cxx |   18 +-
 jvmfwk/plugins/sunmajor/pluginlib/vendorbase.cxx   |8 ++--
 jvmfwk/source/elements.cxx |8 ++--
 jvmfwk/source/framework.cxx|4 +-
 jvmfwk/source/fwkbase.cxx  |   28 
 l10ntools/source/directory.cxx |2 +-
 l10ntools/source/export.cxx|   32 ++--
 l10ntools/source/help/HelpCompiler.hxx |2 +-
 l10ntools/source/helpmerge.cxx |2 +-
 l10ntools/source/merge.cxx |2 +-
 lingucomponent/source/lingutil/lingutil.cxx|2 +-
 .../source/spellcheck/spell/sspellimp.cxx  |4 +-
 .../source/thesaurus/libnth/nthesimp.cxx   |2 +-
 15 files changed, 76 insertions(+), 76 deletions(-)

diff --git a/jvmfwk/plugins/sunmajor/pluginlib/gnujre.cxx b/jvmfwk/plugins/sunmajor/pluginlib/gnujre.cxx
index 53ac693..628461c 100644
--- a/jvmfwk/plugins/sunmajor/pluginlib/gnujre.cxx
+++ b/jvmfwk/plugins/sunmajor/pluginlib/gnujre.cxx
@@ -164,7 +164,7 @@ bool GnuInfo::initialize(vector > props)
 }
 else if (!bAccess && sAccessProperty.equals(i->first))
 {
-if (i->second.getLength() > 0)
+if (!i->second.isEmpty())
 {
 m_bAccessibility = true;
 bAccess = true;
@@ -177,11 +177,11 @@ bool GnuInfo::initialize(vector > props)
 if (!bVersion || !bVendor || !bHome)
 return false;
 
-if (!m_sJavaHome.getLength())
+if (m_sJavaHome.isEmpty())
 m_sJavaHome = rtl::OUString(RTL_CONSTASCII_USTRINGPARAM("file:///usr/lib"));
 
 // init m_sRuntimeLibrary
-OSL_ASSERT(m_sHome.getLength());
+OSL_ASSERT(!m_sHome.isEmpty());
 //call virtual function to get the possible paths to the runtime library.
 
 int size = 0;
@@ -266,7 +266,7 @@ bool GnuInfo::initialize(vector > props)
 return false;
 
 // init m_sLD_LIBRARY_PATH
-OSL_ASSERT(m_sHome.getLength());
+OSL_ASSERT(!m_sHome.isEmpty());
 size = 0;
 char const * const * arLDPaths = getLibraryPaths( & size);
 vector ld_paths = getVectorFromCharArray(arLDPaths, size);
diff --git a/jvmfwk/plugins/sunmajor/pluginlib/sunjavaplugin.cxx b/jvmfwk/plugins/sunmajor/pluginlib/sunjavaplugin.cxx
index 892e88f..7e481e5 100644
--- a/jvmfwk/plugins/sunmajor/pluginlib/sunjavaplugin.cxx
+++ b/jvmfwk/plugins/sunmajor/pluginlib/sunjavaplugin.cxx
@@ -116,7 +116,7 @@ OString getPluginJarPath(
 //but this does not harm. 1.5.0-beta < 1.5.0
 sName = sName2;
 }
-if (sName.getLength())
+if (!sName.isEmpty())
 {
 sName = sLocation + OUSTR("/lib/") + sName;
 OSL_VERIFY(
@@ -140,7 +140,7 @@ OString getPluginJarPath(
 sPath = sPath1 + OUString::createFromAscii(sep) + sPath2;
 }
 }
-OSL_ASSERT(sPath.getLength());
+OSL_ASSERT(!sPath.isEmpty());
 }
 ret = rtl::OUStringToOString(sPath, osl_getThreadTextEncoding());
 
@@ -167,7 +167,7 @@ JavaInfo* createJavaInfo(const rtl::Reference & info)
 pInfo->nRequirements = info->needsRestart() ? JFW_REQUIRE_NEEDRESTART : 0;
 rtl::OUStringBuffer buf(1024);
 buf.append(info->getRuntimeLibrary());
-if (info->getLibraryPaths().getLength() > 0)
+if (!info->getLibraryPaths().isEmpty())
 {
 buf.appendAscii("\n");
 buf.append(info->getLibraryPaths());
@@ -239,8 +239,8 @@ javaPluginError jfw_plugin_getAllJavaInfos(
 OUString ouMinVer(sMinVersion);
 OUString ouMaxVer(sMaxVersion);
 
-OSL_ASSERT(ouVendor.getLength() > 0);
-if (ouVendor.getLength() == 0)
+OSL_ASSERT(!ouVendor.isEmpty());
+if (ouVendor.isEmpty())
 return JFW_PLUGIN_E_INVALID_ARG;
 
 JavaInfo** arInfo = NULL;
@@ -258,7 +258,7 @@ javaPluginError jfw_plugin_getAllJavaInfos(
 if (ouVendor.equals(cur->getVendor()) == sal_False)
 continue;
 
-if (ouMinVer.getLength() > 0)
+if (!ouMinVer.isEmpty())
 {
 try
 {
@@ -276,7 +276,7 @@ javaPluginError jfw_plugin_getAllJavaInfos(
 }
 }
 
-if (ouMaxVer.getLength() > 0)
+if (!ouMaxVer.isEmpty())
 {
 try
 {
@@ -358,8 +358,8 @@ j

Re: [Libreoffice] [Patch] [Review 3.5 - Late Feature] Importing smartArt from

[Michael Meeks]

On Fri, 2011-12-30 at 20:01 +0100, Jan Holesovsky wrote:
> So you have my go-ahead, the code looks nice, and self-contained

Mine too - it looks extremely short & sweet; and without it we just
have a huge white hole so ... ;-)

One more review needed, I'd love to get Thorsten's input of course.

Thanks,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

[Libreoffice] development summary: year 2011, week 52

[Petr Mladek]

Hi,

this time a brief summary of what happened during the 52nd week in 2011
on LibreOffice repositories and the living branches:

+ master:LO-3.5 development
+ libreoffice-3-5:   fixes for LO-3.5.0 release
+ libreoffice-3-4:   fixes for potential LO-3.4.6 bug fix release
+ libreoffice-3-4-5: final stabilization of the LO-3.4.5 release

There are two logs for each branch:

+ bugfixes--week--.txt lists all commits that 
reference a proper
  bug id from a variety of trackers, i.e. #i... referring to the 
OpenOffice
  issuezilla, fdo# to freedesktop, rhbz# to RedHat bugzilla

+ commit-log--week--.txt lists all relevant commits 
on the actual
  source repositories

Many thanks to all contributors - you make all the difference!


Best Regards,
Petr

none this week
+ base
+ missing icons in Index Design dialog (fdo#38435) [Andras Timar]
+ components
+ align dictionary address with DICT_REPO_URL in instsetoo_native/util/openoffice.lst (fdo#44178) [Lior Kaplan]
+ core
+ fix RTF export of date field (fdo#35099) [Miklos Vajna]
+ fix crash in DomainMapper_Impl::SetFieldFFData (fdo#43380) [Miklos Vajna]
+ missing icons in Index Design dialog (fdo#38435) [Andras Timar]
+ missing pointer check in new multiline input code, (fdo#43856) [Markus Mohrhard]
+ related: fix parsing of the \cf RTF token (fdo#43380) [Miklos Vajna]
+ remove Tahoma from the Hebrew font list (except the UI list) (fdo#32936) [Lior Kaplan]
+ shrink the first column in the open dialog, remove unused flags (fdo#44204) [Ivan Timofeev]
+ core
+ convert some SvStrings to std::vector (fdo#38831) [Brad Sowden]
+ convert some SvStrings to std::vector (part 2) (fdo#38831) [Brad Sowden]
+ easyHack remove code associated with unused icons (fdo#42454) [Gustavo Buzzatti Pacheco]
+ fix RTF export of date field (fdo#35099) [Miklos Vajna]
+ fix ScDatabaseRangeObj test, (fdo#43312) [Markus Mohrhard]
+ fix crash in DomainMapper_Impl::SetFieldFFData (fdo#43380) [Miklos Vajna]
+ importing extLst of drawings in diagrams. (bnc#655408) [Muthu Subramanian]
+ missing icons in Index Design dialog (fdo#38435) [Andras Timar]
+ missing pointer check in new multiline input code, (fdo#43856) [Markus Mohrhard]
+ related: fix parsing of the \cf RTF token (fdo#43380) [Miklos Vajna]
+ remove Tahoma from the Hebrew font list (except the UI list) (fdo#32936) [Lior Kaplan]
+ shrink the first column in the open dialog, remove unused flags (fdo#44204) [Ivan Timofeev]
+ bootstrap
+ fix build with bleeding edge external boost and gcc [Caolán McNamara]
+ libs-core
+ we need to include the localized file [David Tardon]
+ translations
+ translation update for LibreOffice 3.4.5 rc2 [Andras Timar]
+ base
+ missing icons in Index Design dialog (fdo#38435) [Andras Timar]
+ components
+ align dictionary address with DICT_REPO_URL in instsetoo_native/util/openoffice.lst (fdo#44178) [Lior Kaplan]
+ translations
+ translation update for LibreOffice 3.4.5 rc2 [Andras Timar]
+ core
+ build qadevOOo only if  OOO_JUNIT_JAR are set (add BUILD_TYPE) [Rene Engelhard]
+ build qadevOOo only if SOLAR_JAVA and OOO_JUNIT_JAR are set [Rene Engelhard]
+ fdo42286 better solution: extend and shrink end of row if needed [Pierre-André Jacquod]
+ fdo43033 do not export to PDF hidden slides [Pierre-André Jacquod]
+ fix KeyID generation for a corner case [Andras Timar]
+ fix RTF export of date field (fdo#35099) [Miklos Vajna]
+ fix assertion error in GtkFolderPicker during recovery dialog [Markus Mohrhard]
+ fix crash in DomainMapper_Impl::SetFieldFFData (fdo#43380) [Miklos Vajna]
+ fix drawing of the either horizontally, or vertically mirrored bitmaps. [Jan Holesovsky]
+ fix init order [Julien Nabet]
+ fix order init [Julien Nabet]
+ fix position of the 'evaluate up to level' text and numeric field. [Jan Holesovsky]
+ keep abstract dialog terminology aligned [Lior Kaplan]
+ missing icons in Index Design dialog (fdo#38435) [Andras Timar]
+ missing pointer check in new multiline input code, (fdo#43856) [Markus Mohrhard]
+ move OSL_FAIL to correct location [Markus Mohrhard]
+ now that QADEVOOO build_type exist we need to use it... [Norbert Thiebaud]
+ option to export hidden slides to pdf files [Pierre-André Jacquod]
+ related: fix parsing of the \cf RTF token (fdo#43380) [Miklos Vajna]
+ remove Tahoma from the Hebrew font list (except the UI list) (fdo#32936) [Lior Kaplan]
+ remove an unnecessary full stop [Lior Kaplan]
+ remove the dog [Stefan Knorr (astron)]
+ revert "allow to call ooinstall twice with $DESTDIR" [Tomas Chvatal]
+ shrink the first column in the open dialog, remove unused flags (fdo#44204) [Ivan Timofeev]
+ thanks to Ivan Timofeev, fix the 'Preview' checkbox in 'Indexes and Tables...' [Jan Holesovsky]
+ try to draw re

[Libreoffice] development summary: year 2011, week 51

[Petr Mladek]

Hi,

this time a brief summary of what happened during the 51st week in 2011
on LibreOffice repositories and the living branches:

+ master:LO-3.5 development
+ libreoffice-3-5:   fixes for LO-3.5.0 release
+ libreoffice-3-4:   fixes for potential LO-3.4.6 bug fix release
+ libreoffice-3-4-5: final stabilization of the LO-3.4.5 release

There are two logs for each branch:

+ bugfixes--week--.txt lists all commits that 
reference a proper
  bug id from a variety of trackers, i.e. #i... referring to the 
OpenOffice
  issuezilla, fdo# to freedesktop, rhbz# to RedHat bugzilla

+ commit-log--week--.txt lists all relevant commits 
on the actual
  source repositories

Many thanks to all contributors - you make all the difference!


Best Regards,
Petr
+ components
+ revert "Fix " (bnc#653688) [Thorsten Behrens]
+ components
+ revert "Fix " (bnc#653688) [Thorsten Behrens]
+ libs-core
+ fix hyphenation attributes: (fdo#44082) [Michael Stahl]
+ oDF export: fix layout grid invalid ODF (fdo#44073) [Michael Stahl]
+ writer
+ sw: temp selection print doc: (fdo#40482, fdo#39159) [Michael Stahl]
+ core
+ "*.*" is not "all files" (bnc#738021) [Luboš Luňák]
+ always load res of LibO UI locale, not system deflocale (fdo#43475) [Andras Timar]
+ can't claim its not a DXF just because didn't see SECTION (fdo#43082) [Caolán McNamara]
+ correctly handle empty paragraphs in slideshow animation (fdo#43684) [Thorsten Behrens]
+ don't hide images in menus (fdo#41264, fdo#31471) [Ivan Timofeev]
+ fix (bad hyphenation of French words with hyphen and apostrophe) (#43931) [László Németh]
+ fix RTF import of fields inside tables (fdo#42109) [Miklos Vajna]
+ fix RTF import of images with invalid crop data (fdo#40937) [Miklos Vajna]
+ fix for - prevent invalid shape text indices. (fdo#43837) [Thorsten Behrens]
+ fix hyphenation attributes: (fdo#44082) [Michael Stahl]
+ fix slideshow text animations (fdo#43837) [Thorsten Behrens]
+ header/Footer,Page Break: fix bad test for printing mode (fdo#43962) [Cédric Bosdonnat]
+ if a bracket pair has no left/right bracket, it needs to explicit (fdo#32636) [Luboš Luňák]
+ import m:eqArr (part of (fdo#32636) [Luboš Luňák]
+ oDF export: fix layout grid invalid ODF (fdo#44073) [Michael Stahl]
+ remove Tahoma from the Hebrew font list (except the UI list) (fdo#32936) [Lior Kaplan]
+ resolved - Date formatting in Spreadsheet is inconsistent (fdo#37978) [Eike Rathke]
+ revert "aw084: Added code to mimic old behaviour and call user layout link when empty paragraph is rendered" (fdo#43684, fdo#43837, i#108052) [Thorsten Behrens]
+ sw: Fix cursor accessibility API (fdo#43390) [Vincent Povirk]
+ sw: avoid creating cursor with non-text node text range (fdo#40195) [Noel Power]
+ sw: temp selection print doc: (fdo#40482, fdo#39159) [Michael Stahl]
+ binfilter
+ tweak for pre language-defect (#77) [Caolán McNamara]
+ core
+ "*.*" is not "all files" (bnc#738021) [Luboš Luňák]
+ allow the use of real access() calls (fdo#43095) [Moritz Bechler]
+ always load res of LibO UI locale, not system deflocale (fdo#43475) [Andras Timar]
+ can't claim its not a DXF just because didn't see SECTION (fdo#43082) [Caolán McNamara]
+ correctly handle empty paragraphs in slideshow animation (fdo#43684) [Thorsten Behrens]
+ don't hide images in menus (fdo#41264, fdo#31471) [Ivan Timofeev]
+ fix (bad hyphenation of French words with hyphen and apostrophe) (#43931) [László Németh]
+ fix RTF import of fields inside tables (fdo#42109) [Miklos Vajna]
+ fix RTF import of images with invalid crop data (fdo#40937) [Miklos Vajna]
+ fix for - prevent invalid shape text indices. (fdo#43837) [Thorsten Behrens]
+ fix hyphenation attributes: (fdo#44082) [Michael Stahl]
+ fix slideshow text animations (fdo#43837) [Thorsten Behrens]
+ header/Footer,Page Break: fix bad test for printing mode (fdo#43962) [Cédric Bosdonnat]
+ if a bracket pair has no left/right bracket, it needs to explicit (fdo#32636) [Luboš Luňák]
+ import m:eqArr (part of (fdo#32636) [Luboš Luňák]
+ oDF export: fix layout grid invalid ODF (fdo#44073) [Michael Stahl]
+ rTF/DOCX import: fixed the import of consecutive frames (bnc#703032) [Cédric Bosdonnat]
+ recover fix for (fdo#43931) [Caolán McNamara]
+ remove Tahoma from the Hebrew font list (except the UI list) (fdo#32936) [Lior Kaplan]
+ replace ScChangeActionTable with std::map (fdo#38832) [Marcel Metz]
+ resolved - Date formatting in Spreadsheet is inconsistent (fdo#37978) [Eike Rathke]
+ revert "aw084: Added code to mimic old behaviour and call user layout link when empty paragraph is rendered" (fdo#43684, fdo#43837, i#108052) [Thorsten Behrens]
+ sw: Fix cursor accessibility API (fdo#43390) [Vincent Povirk]
+ sw: avoid creating cursor with non-te

Re: [Libreoffice] [PUSHED] Re: [PATCH] Translate German comments to English and correct the English ones

[Philipp Weissenbacher]

On 4 January 2012 14:07, Michael Stahl  wrote:
> On 04/01/12 12:57, Philipp Weissenbacher wrote:
>> Hi all,
>>
>> Here's another patch that translates the German comments to English
>> and fixes some errors in the English ones.
>
> thanks, pushed (tweaked the translations a bit in a follow-up)
>
>> In htmltbl.cxx:
>> I interpreted 'tab' in some OSL_ENSUREs as 'table' and replaced 'table
>> in table' with 'nested table' (that seems somewhat clearer to me).
>
> why not :)
>
>> In number.cxx:
>> The code on lines 531/532 seems to be duplicated on lines 535/536. The
>> same is repeated on the next lines. Is that intended?
>
> surprisingly that is actually correct, because there are 2 arrays of
> size MAXLEVEL, i.e. it is an array [2][MAXLEVEL]; see the constructor of
> the class.  but of course writing it in this way is rather ugly, nested
> loop would be nicer.
>
> some misc. translation points:
>
> - "eventual" means something completely different from German
>  "eventuell"
> - "echt groesser" is a mathematical term and the "echt" is translated
>  as "strictly"
> - when it comes to numbering, "Gliederung" is "Outline"

Thanks a lot for these!
This "ambiguity of terms" is what makes translating somewhat involved
(next to monster/broken sentences).

Regards,
Philipp
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [Libreoffice-ux-advise] [REVIEW 3-5: Late feature] Improvements in the header / footer behavior

[Cor Nouws]

Michael Meeks wrote (04-01-12 13:04)


Is it easy / possible to trigger the header appear etc. only on a
mouse-move event ? :-)


+1 (as expressed earlier ;-) )

--
 - Cor
 - http://nl.libreoffice.org

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

[Libreoffice] [PUSHED] Re: [PATCH] Translate German comments to English and correct the English ones

[Michael Stahl]

On 04/01/12 12:57, Philipp Weissenbacher wrote:
> Hi all,
> 
> Here's another patch that translates the German comments to English
> and fixes some errors in the English ones.

thanks, pushed (tweaked the translations a bit in a follow-up)

> In htmltbl.cxx:
> I interpreted 'tab' in some OSL_ENSUREs as 'table' and replaced 'table
> in table' with 'nested table' (that seems somewhat clearer to me).

why not :)

> In number.cxx:
> The code on lines 531/532 seems to be duplicated on lines 535/536. The
> same is repeated on the next lines. Is that intended?

surprisingly that is actually correct, because there are 2 arrays of
size MAXLEVEL, i.e. it is an array [2][MAXLEVEL]; see the constructor of
the class.  but of course writing it in this way is rather ugly, nested
loop would be nicer.

some misc. translation points:

- "eventual" means something completely different from German
  "eventuell"
- "echt groesser" is a mathematical term and the "echt" is translated
  as "strictly"
- when it comes to numbering, "Gliederung" is "Outline"

regards,
 michael

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

Re: [Libreoffice] [PATCH][PUSHED] Easyhack fdo#38831 remove SvStrings

[Lubos Lunak]

On Thursday 29 of December 2011, Brad Sowden wrote:
> Hi,
>
> Following on from the 2 previous patches that I sent to the list, this
> series of patches removes all remaining instances of SvStrings.
>
> Patches contributed under LGPLv3+/MPL1.1.

 Pushed the rest of the patches, thanks (and all the notes were correct).

-- 
 Lubos Lunak
 l.lu...@suse.cz
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] [Bug 37361] LibreOffice 3.5 most annoying bugs

[bugzilla-daemon]

https://bugs.freedesktop.org/show_bug.cgi?id=37361

--- Comment #70 from Rainer Bielefeld  
2012-01-04 04:36:30 PST ---
Nominate "Bug 41436 - FORMATTING: User style not trans ferred with copy / paste
to other document", this bug makes effective work with DRAW impossible
(enchains me to OOo 3.1.1)

-- 
Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email
--- You are receiving this mail because: ---
You are on the CC list for the bug.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PATCH][PUSHED] Fix for fdo43460 Part XXVIII getLength() to isEmpty()

[Lubos Lunak]

On Wednesday 04 of January 2012, Olivier Hallot wrote:
> Please find attached a partial fix for Easy Hack FDO43460
>
> Part XXVIII
> Modules
> reportdesign

 Pushed.

 After 28 patches and about so many still to come, would you consider getting 
commit access? These getLength()->isEmpty() patches I've reviewed are 
correct, I expect the rest will be as well, and by now you are the expert in 
this field anyway :). You can still post for review other patches where you 
won't be sure.

-- 
 Lubos Lunak
 l.lu...@suse.cz
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [REVIEW] Patch for "Same expression on both sides of '&&'" in basegfx module

[Caolán McNamara]

On Tue, 2012-01-03 at 20:05 +0100, Eike Rathke wrote:
> Hi julien2412,
> 
> On Wednesday, 2011-12-28 13:13:43 -0800, julien2412 wrote:
> 
> > On cppcheck report, there were 2 "Same expression on both sides of '&&'" on 
> > b2dpolypolygoncutter.cxx
> > 
> > I don't know how this part works so I recognize it's just guessing.
> > 
> > diff --git a/basegfx/source/polygon/b2dpolypolygoncutter.cxx
> > b/basegfx/source/polygon/b2dpolypolygoncutter.cxx
> > index d32ee87..8745241 100644
> > --- a/basegfx/source/polygon/b2dpolypolygoncutter.cxx
> > +++ b/basegfx/source/polygon/b2dpolypolygoncutter.cxx
> > @@ -282,7 +282,7 @@ namespace basegfx
> >  bOnEdge = false;
> >  }
> >  }
> > -while(bOnEdge && pPNa2 != &rPNa && pPNa2 != &rPNa);
> > +while(bOnEdge && pPNa2 != &rPNa && pPNb2 != &rPNb);
> >  
> >  if(bOnEdge)
> >  {
> > @@ -408,7 +408,7 @@ namespace basegfx
> >  bOnEdge = false;
> >  }
> >  }
> > -while(bOnEdge && pPNa2 != &rPNa && pPNa2 != &rPNa);
> > +while(bOnEdge && pPNa2 != &rPNa && pPNb2 != &rPNb);
> >  
> >  if(bOnEdge)
> >  {
> > 
> > Does it seem ok ? (If yes, I can commit and push it on master branch)
> 
> I think so, but I'm not sure either. Anyone with a second opinion?

Change seems reasonable to me FWIW, presumably Thorsten knows best what
basegfx wants to do here.

C.

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

Re: [Libreoffice] gcc 4.6.2 + boost 1.48.0 and unordered_map assignment operator const issues [was gcc 4.6.2]

[Andreas Radke]

While doing the db 5.3.15 rebuilds I'm running into this build issue.
It might be caused by your fix but could also be our new glibc updated
toolchain breaking gcc itself. CC'ing our toolchain maintainer also.

It builds fine with old internal boost 1.44.

[andyrtr@workstation64 tmp]$ pacman -Q boost gcc glibc binutils
boost 1.48.0-2
gcc 4.6.2-5
glibc 2.15-3
binutils 2.22-4


libreoffice 3-4-5rc2


[ build CXX ] framework/source/uifactory/menubarfactory
In file included from 
/build/src/build/framework/inc/uielement/toolbarmerger.hxx:39:0,
 from 
/build/src/build/framework/source/uielement/toolbarmerger.cxx:32:
/build/src/build/framework/inc/uielement/commandinfo.hxx: In constructor 
'framework::CommandInfo::CommandInfo(framework::CommandInfo&&)':
/build/src/build/framework/inc/uielement/commandinfo.hxx:48:8:   instantiated 
from 'std::pair<_T1, _T2>::pair(std::pair<_U1, _U2>&&) [with _U1 = const 
rtl::OUString, _U2 = framework::CommandInfo, _T1 = const rtl:
:OUString, _T2 = framework::CommandInfo]'
/usr/include/boost/unordered/detail/emplace_args.hpp:268:9:   instantiated from 
'void boost::unordered::detail::construct_impl(T*, Args&& ...) [with T = 
std::pair, Arg
s = {std::pair}]'
/usr/include/boost/unordered/detail/buckets.hpp:73:13:   instantiated from 
'void 
boost::unordered::detail::node_constructor::construct_value(Args&& 
...) [with Args = {std::pair}, NodeAlloc = 
std::allocator > >]'
/usr/include/boost/unordered/detail/unique.hpp:399:13:   instantiated from 
'boost::unordered::detail::table_impl::emplace_return 
boost::unordered::detail::table_impl::emplace_impl(const key_type&, A
rgs&& ...) [with Args = {std::pair}, Types = 
boost::unordered::detail::map >, rtl::OUString, fram
ework::OUStringHashCode, std::equal_to >, 
boost::unordered::detail::table_impl::emplace_return = 
std::pair >*, std::pair >, bool>, typename 
boost::unordered::detail::table::iterator = 
boost::unordered::iterator_detai
l::iterator >*, std::pair >, boost::unordered::detail::table_impl::key
_type = rtl::OUString]'
/usr/include/boost/unordered/detail/unique.hpp:368:48:   instantiated from 
'boost::unordered::detail::table_impl::emplace_return 
boost::unordered::detail::table_impl::emplace(Args&& ...) [with Args 
= {std::pair}, Types = 
boost::unordered::detail::map >, rtl::OUString, framework::OUStringHashCod
e, std::equal_to >, 
boost::unordered::detail::table_impl::emplace_return = 
std::pair >*, std::pair >, bool>, typename 
boost::unordered::detail::table::iterator = 
boost::unordered::iterator_detail::iterator 
>*, std::pair >]'
/usr/include/boost/unordered/unordered_map.hpp:221:62:   instantiated from 
'std::pair >::type, K, H, 
P>::table::iterator, bool> boost::unordered::unordered_map::emplace(Args&& ...) [with Args = {std::pair}, K = rtl::OUString, T = framework::CommandInfo,
 H = framework::OUStringHashCode, P = std::equal_to, A = 
std::allocator >, 
typename boost::unordered::detail::map >::type, K, H, P>::table::iterator 
= 
boost::unordered::iterator_detail::iterator >*, std::pair >]'
/usr/include/boost/unordered/unordered_map.hpp:296:48:   instantiated from 
'std::pair >::type, K, H, 
P>::table::iterator, bool> boost::unordered::unordered_map::insert(boost::unordered::unordered_map::value_type&&) [with 
K = rtl::OUString, T = framework::CommandInfo, H = framework::
OUStringHashCode, P = std::equal_to, A = 
std::allocator >, 
typename boost::unordered::detail::map >::type, K, H, 
P>::table::iterator = 
boost::unordered::iterator_detail::iterator >*, std::pair >, boost::unordered::unordered_map::value_type = std::pair]'
/build/src/build/framework/source/uielement/toolbarmerger.cxx:521:100:   
instantiated from here
/build/src/build/framework/inc/uielement/commandinfo.hxx:48:8: internal 
compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See  for instructions.
make[1]: *** 
[/build/src/build/solver/340/unxlngx6.pro/workdir/CxxObject/framework/source/uielement/toolbarmerger.o]
 Error 1


-Andy
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [Libreoffice-ux-advise] [REVIEW 3-5: Late feature] Improvements in the header / footer behavior

[Michael Meeks]

On Fri, 2011-12-30 at 17:22 +0100, Jan Holesovsky wrote:
> So - what I'll try to do is that the appearing will start later (not 1
> sec, but maybe 200 - 500 ms), but the functionality will be available
> immediately when the mouse is there; so the impatient user who knows
> where to click will be able to perform immediately, while the extensive
> mouse traveller will not be annoyed.  Hopefully that will be the best
> from the both worlds :-)

Heh :-) one thing that is somewhat annoying when actually using writer
is that of the mouse not being moved - but ending up over some area that
subsequently has a header area moved into it (by moving the page down
etc.).

Is it easy / possible to trigger the header appear etc. only on a
mouse-move event ? :-)

ATB,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] .: sot/Library_sot.mk

[Michael Stahl]

On 04/01/12 12:06, Caolán McNamara wrote:
> On Tue, 2012-01-03 at 17:26 +0100, Michael Stahl wrote:
>> wonder if it would make sense to define NOMINMAX globally in windows.mk
> 
> probably, unless there is some other godawful chunk of windows specific
> code hidden somewhere which falls over and dies if it *isn't* a
> macro :-)

let's try it out, and fix what breaks :)

http://cgit.freedesktop.org/libreoffice/core/commit/?id=473a1a4a48003e926ebed16f3535f57bfcf371d9

grep found a NOMINMAX in mingw header windef.h as well, so i've just
moved the identical gb_OSDEFS to windows.mk.

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

Re: [Libreoffice] .: sot/Library_sot.mk

[Caolán McNamara]

On Tue, 2012-01-03 at 17:26 +0100, Michael Stahl wrote:
> wonder if it would make sense to define NOMINMAX globally in windows.mk

probably, unless there is some other godawful chunk of windows specific
code hidden somewhere which falls over and dies if it *isn't* a
macro :-)

C.

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

[Libreoffice] [ANNOUNCE] LibreOffice 3.6 most annoying bugs (fdo#44446)

[Petr Mladek]

Hi,

LibreOffice-3.6 is already developed in master, daily builds are
produced, so I created the usual meta bug for most annoying bugs,
see https://bugs.freedesktop.org/show_bug.cgi?id=6

Also note the page for 3.6 release notes, see
http://wiki.documentfoundation.org/ReleaseNotes/3.6. Do not forget to
mention your nice improvements there.


Best Regards,
Petr

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

[Libreoffice] What are the criteria to nominate bugs as most annoying bug?

[Winfried Donkers]

Hi,

There are some bugs in calc that damage formulas/conditional formats of cells. 
As this goes further than just being a nuisance I would like to nominate them 
for solving with high(er) priority.
But I don't want to make things worse and I don't want to annoy people who are 
already working hard on improving LibreOffice. (The bugs are too complicated 
for me to hack)
These are the bugs:

https://bugs.freedesktop.org/show_bug.cgi?id=37987


https://bugs.freedesktop.org/show_bug.cgi?id=44383




Winfried

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

Re: [Libreoffice] About enabling Icedove addressbook connection for Base

[Michael Meeks]

Hi Julien,

On Wed, 2011-12-28 at 03:08 -0800, julien2412 wrote:
> As most of you (everybody ?) must know, Icedove is the rebranded version of
> Thunderbird (like Iceweasel for Firefox).

:-) What a sexy TM policy that is ... thankfully ours is better in that
regard.

> I tried to use Base with Thunderbird connection to reproduce some bugs and
> saw that addressbook couldn't be found.

Oh - odd; I imagine that is the detection of icedove going wrong.
Poking in connectivity/source/drivers/mozab/ would prolly help - do we
load the mozilla module in MDriver.cxx ? (generated from SLO2FILES in
makefile.mk) ?

> I opengroked Thunderbird and Icedove and wonder now how to enable Icedove
> addressbook connection for Base with the minimum of duplication.
> Should there be an Icedove connection or should it be Thunderbird/Icedove in
> dialog for example ?

Oh - adding a Thunderbird/Icedove in the dialog sounds by far the best
from the perspective of reducing clutter / duplication. Also - (from my
cursory glance) it is unclear that we require any real code changes to
support icedove - I don't see (off hand) - is it possible that IceDove
tweaks the product type enumerations in some way eg.

return ::com::sun::star::mozilla::MozillaProductType_Thunderbird;

Anyhow - I hope that helps !

All the best,

Michael.

-- 
michael.me...@suse.com  <><, Pseudo Engineer, itinerant idiot

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

Re: [Libreoffice] [PATCH] Fix for fdo43460 Part XXVIII getLength() to isEmpty()

[Olivier Hallot]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi

Em 04-01-2012 07:17, Tommy escreveu:
> On Wed, 04 Jan 2012 00:17:13 +0100, Olivier Hallot
>  wrote:
> 
>> Please find attached a partial fix for Easy Hack FDO43460
>>
>> Part XXVIII
>> Modules
>> reportdesign
>> - -- 
>> Olivier Hallot
>> Founder, Board of Directors Member - The Document Foundation
>> LibreOffice translation leader for Brazilian Portuguese
>> +55-21-8822-8812
> 
> 
> 
> just curious... how many parts are still missing?
> you did already 28 of them

I think I am in the middle of the task, so perhaps there is about 60
parts in total. I do it by modules, to keep the mental sanity of the
reviewer and myself.

Feel free to join the team :-) .

Regards
- -- 
Olivier Hallot
Founder, Board of Directors Member - The Document Foundation
LibreOffice translation leader for Brazilian Portuguese
+55-21-8822-8812
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/

iQEcBAEBAgAGBQJPBB+2AAoJEJp3R7nH3vLxe+sH/RAxLhPmzOjn729XEgRXL4xg
TeBywFLevUVIlVcqpO3DUqrml6A/OoFDICE2/rkEddhD7BdiXKJimNqHI2pqJVg4
TJSz4el/6J8rGNkOV4qY5DOyzJjjWXeHOZyS8yQf4hZNBWs93gfuizYKAvPNC+Ci
DpZZ9qxaHEYQvIhhVUKadLl8uyZYyg6/Tj0WPFMKgSFKO6pNzrTTLz3ohSzRr04i
MnrDP15FgYGt7X/BtcwP3XnJpSji3i4pqlTPJfwQBIJTxjML36XHVv8F7ft/ctpd
tcKkk9Oyaw90y8kRGSynxgWbNKNIZbRFzC2dcBpxF8sw1HEJFVSEfC7fCI9R2gI=
=e5AA
-END PGP SIGNATURE-
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] make dev-install error, dict-xx.oxt files not found

[Winfried Donkers]

>> Still strange that git pull -r from core doesn't/didn't help. That's what I 
>> do to update my sources.

>Do a "./g pull -r" in core instead.  It calls "git pull -r" in all the 
>sub-repos underneath core/clone/ (and in core itself, of course).

Thank you. I will do ./g instead of git in the future and reread 
http://wiki.documentfoundation.org/Development#Using_Git_for_LibreOffice_development
 :)

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

Re: [Libreoffice] [PATCH] Fix for fdo43460 Part XXVIII getLength() to isEmpty()

[Tommy]

On Wed, 04 Jan 2012 00:17:13 +0100, Olivier Hallot  
 wrote:



Please find attached a partial fix for Easy Hack FDO43460

Part XXVIII
Modules
reportdesign
- --
Olivier Hallot
Founder, Board of Directors Member - The Document Foundation
LibreOffice translation leader for Brazilian Portuguese
+55-21-8822-8812




just curious... how many parts are still missing?
you did already 28 of them

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

Re: [Libreoffice] Updating Italian dictionaries

[Tommy]

On Tue, 03 Jan 2012 22:33:10 +0100, Andrea Pescetti  
 wrote:



(I follow this list, but please do CC me explicitly on replies)

The Italian dictionary in LibreOffice master is not the latest version,
that includes some bugfixes and a detailed history of contributions that
was missing from the version shipped in OpenOffice.org 3.3.0 and
imported into LibreOffice. This should be fixed.




+1 from an italian user

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

Re: [Libreoffice] make dev-install error, dict-xx.oxt files not found

[Stephan Bergmann]

On 01/04/2012 07:43 AM, Winfried Donkers wrote:

The error is that dict-xx.oxt (today they were dict-an.oxt, dict-be.oxt, 
dict-el.oxt, dict-gd.oxt, dict-si.oxt and dict-te.oxt) files cannot be found.

I think running "git pull -r" in core/dictionaries should fix this.


Thanks, it helped.
Still strange that git pull -r from core doesn't/didn't help. That's what I do 
to update my sources.


Do a "./g pull -r" in core instead.  It calls "git pull -r" in all the 
sub-repos underneath core/clone/ (and in core itself, of course).


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

Re: [Libreoffice] Python Bridge on MacOS x64 with python default

[Stephan Bergmann]

On 01/03/2012 09:58 PM, Ovnicraft wrote:

I want to know if there is any plan to get LO for x64, python comes in
LO was built in x32.


Nobody is working on such a port as far as I know.


I get the SDK for Libre office but has no python bridge.


The Python--UNO bridge is included in LO itself (see 
LibreOffice.app/Contents/MacOS/py*), the SDK traditionally has little 
(additional) support for PyUNO.


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