Re: [Libreoffice] [Libreoffice-ux-advise] [PATCH] [PUSHED] fdo#45285 - Use more revealing error message for not allowed sheet names

[Ivan Timofeev]

30.01.2012 13:07, Stefan Knorr (Astron) пишет:

the idea sounds good, ... patch is simple, and the details seem to be
correct now. Thus, I've taken the liberty to replace "can" by "may"
and then push to master.


just added ':' to the sequence of characters
http://cgit.freedesktop.org/libreoffice/core/commit/?id=0a9d8c07ddf054930e1b000cc3c972a3d217d71c

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

[Libreoffice] [PATCH] SIMIAN: Reduced duplicate code

[Chr. Rossmanith]

Hi,

please have a look at the comments marked with "crossmanith". I can push 
on my own but a short review (and feedback) would be nice.


Christina
>From ac8d464245852f701478039f89ae876d6584fb88 Mon Sep 17 00:00:00 2001
From: Christina Rossmanith 
Date: Tue, 31 Jan 2012 10:07:19 +0100
Subject: [PATCH] SIMIAN: Reduced duplicate code

---
 vcl/source/gdi/outdev4.cxx |  118 +--
 1 files changed, 47 insertions(+), 71 deletions(-)

diff --git a/vcl/source/gdi/outdev4.cxx b/vcl/source/gdi/outdev4.cxx
index 735a85e..63c53e4 100644
--- a/vcl/source/gdi/outdev4.cxx
+++ b/vcl/source/gdi/outdev4.cxx
@@ -168,6 +168,45 @@ inline sal_uInt8 ImplGetGradientColorValue( long nValue )
 }
 
 // ---
+long ImplCalcGradientSteps( const Gradient& rGradient, const sal_Bool bMtf, const long nSize, const OutDevType outdevtype, const long nMaxSteps )
+{
+longnStepCount;
+
+// calculate number of steps if no value was passed in rGradient
+if( !rGradient.GetSteps() )
+{
+long nInc;
+
+if ( outdevtype != OUTDEV_PRINTER && !bMtf )
+{
+nInc = ( nSize < 50 ) ? 2 : 4;
+}
+else
+{
+// #105998# Use display-equivalent step size calculation
+nInc = ( nSize < 800 ) ? 10 : 20;
+}
+
+if( !nInc )
+nInc = 1;   // (crossmanith: code reachable???)
+
+nStepCount = nSize / nInc;
+}
+else {
+nStepCount = rGradient.GetSteps();
+}
+
+// minimum number of steps: 3 (crossmanith: or is it 2?)
+// maximum number of steps: nMaxSteps
+long nSteps = Max( nStepCount, 2L );
+
+if ( nSteps > nMaxSteps )
+nSteps = nMaxSteps;
+if ( !nSteps )
+nSteps = 1;
+
+return nSteps;
+}
 
 void OutputDevice::ImplDrawLinearGradient( const Rectangle& rRect,
const Gradient& rGradient,
@@ -226,50 +265,19 @@ void OutputDevice::ImplDrawLinearGradient( const Rectangle& rRect,
 longnRedSteps   = nEndRed   - nStartRed;
 longnGreenSteps = nEndGreen - nStartGreen;
 longnBlueSteps  = nEndBlue  - nStartBlue;
-longnStepCount = rGradient.GetSteps();
 
-// Bei nicht linearen Farbverlaeufen haben wir nur die halben Steps
+// Bei axialen Farbverlaeufen haben wir nur die halben Steps
 // pro Farbe
-if ( !bLinear )
+if ( rGradient.GetStyle() == GRADIENT_AXIAL )
 {
 nRedSteps   <<= 1;
 nGreenSteps <<= 1;
 nBlueSteps  <<= 1;
 }
 
-// Anzahl der Schritte berechnen, falls nichts uebergeben wurde
-if ( !nStepCount )
-{
-long nInc;
-
-if ( meOutDevType != OUTDEV_PRINTER && !bMtf )
-{
-nInc = (nMinRect < 50) ? 2 : 4;
-}
-else
-{
-// #105998# Use display-equivalent step size calculation
-nInc = (nMinRect < 800) ? 10 : 20;
-}
-
-if ( !nInc )
-nInc = 1;
-
-nStepCount = nMinRect / nInc;
-}
-// minimal drei Schritte und maximal die Anzahl der Farbunterschiede
-long nSteps = Max( nStepCount, 2L );
-long nCalcSteps  = Abs( nRedSteps );
-long nTempSteps = Abs( nGreenSteps );
-if ( nTempSteps > nCalcSteps )
-nCalcSteps = nTempSteps;
-nTempSteps = Abs( nBlueSteps );
-if ( nTempSteps > nCalcSteps )
-nCalcSteps = nTempSteps;
-if ( nCalcSteps < nSteps )
-nSteps = nCalcSteps;
-if ( !nSteps )
-nSteps = 1;
+// max(red,green,blue)
+long nMaxGradientSteps = Max( Max( Abs( nRedSteps ), Abs( nGreenSteps ) ), Abs( nBlueSteps ));
+long nSteps = ImplCalcGradientSteps( rGradient, bMtf, nMinRect, meOutDevType, nMaxGradientSteps );
 
 // Falls axialer Farbverlauf, muss die Schrittanzahl ungerade sein
 if ( !bLinear && !(nSteps & 1) )
@@ -437,7 +445,6 @@ void OutputDevice::ImplDrawComplexGradient( const Rectangle& rRect,
 longnRedSteps = nEndRed - nStartRed;
 longnGreenSteps = nEndGreen - nStartGreen;
 longnBlueSteps = nEndBlue   - nStartBlue;
-longnStepCount = rGradient.GetSteps();
 sal_uInt16  nAngle = rGradient.GetAngle() % 3600;
 
 rGradient.GetBoundRect( rRect, aRect, aCenter );
@@ -449,40 +456,9 @@ void OutputDevice::ImplDrawComplexGradient( const Rectangle& rRect,
 
 long nMinRect = Min( aRect.GetWidth(), aRect.GetHeight() );
 
-// Anzahl der Schritte berechnen, falls nichts uebergeben wurde
-if( !nStepCount )
-{
-long nInc;
-
-if ( meOutDevType != OUTDEV_PRINTER && !bMtf )
-{
-nInc = ( nMinRect < 50 ) ? 2 : 4;
-}
-else
-{
-// #105998# Use display-equivalent step size calculation
-nInc = (nMinRect < 800) ? 10 : 20;
-}
-
-if( !nInc )

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

[bugzilla-daemon]

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

--- Comment #157 from Rainer Bielefeld  
2012-01-31 01:16:45 PST ---
+ "Bug 45409 - FILEOPEN particular .doc: Table of contents broken, internal
reference information visible". Makes 3.5 useless for editing WORD documents,
dataloss, saved with 3.5 also unsable with more early versions.

-- 
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

Petr Mladek  changed:

   What|Removed |Added

 Depends on||45409

-- 
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] [REVIEW] [3.5.0] Some simple polish for named ranges dialogue

[Noel Power]

On 30/01/12 19:21, Petr Mladek wrote:

Stefan Knorr (Astron) píše v Po 30. 01. 2012 v 09:45 +0100:

Hi,

I know this is extremely late for a simple polish fix, but please consider
http://cgit.freedesktop.org/libreoffice/core/commit/?id=f1cb0a4ab4f11dc015be1696c7c7751802171915
as well as
http://cgit.freedesktop.org/libreoffice/core/commit/?id=0b6486772c52d7112e76555fca6525ad2eec2bae
for inclusion in 3.5.(0).
Both changes concern almost exclusively src files, so they should be
pretty safe.
(I accidentally pushed early, that's the only reason why it's split up
into two commits.)

There was still not enough space for German strings in the "Define name"
dialog, see the attached screenshots. So, I have increased it a bit
more, see
http://cgit.freedesktop.org/libreoffice/core/commit/?id=5be59017ca21301e70eca987e8323c92215f0ed4


I merged all 3 commits together and pushed to 3-5 branch, see
http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5&id=b15d3b50ba468defd3109e5ec0aef331e222cb5e


Two more appovals are needed for the 3-5-0 branch.


+1 from me
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PATCH] Remove unused code

[Noel Power]

Hi Elton
First thanks for submitting your patch, it's appreciated however could I 
ask if you could please repost your patch as an attachment [1] ? patch 
content inline in a mail message just does't work as the mail client 
breaks the lines in such a way as the patch program no longer 
understands the diff. A patch as an attachmentwill help the reviewers 
integrate your work easily, thanks again


Noel

p.s. could you also confirm you contribution is under the MPL 1.1 / 
GPLv3+ / LGPLv3+ license(s)


[1] 
http://wiki.documentfoundation.org/Development/Patch_Handling_Guidelines#How_to_submit_a_patch


On 31/01/12 07:58, Elton Chung wrote:

---
  sc/inc/scdll.hxx  |1 -
  sc/source/core/inc/ddelink.hxx|1 -
  sc/source/core/tool/ddelink.cxx   |   11 -
  sc/source/ui/app/scdll.cxx|   10 
  sc/source/ui/dbgui/fieldwnd.cxx   |5 
  sc/source/ui/inc/areasave.hxx |1 -
  sc/source/ui/inc/cellmergeoption.hxx  |1 -
  sc/source/ui/inc/fieldwnd.hxx |1 -
  sc/source/ui/inc/scuitphfedit.hxx |4 ---
  sc/source/ui/pagedlg/scuitphfedit.cxx |   39 -
  sc/source/ui/undo/areasave.cxx|5 
  sc/source/ui/view/cellmergeoption.cxx |9 ---
  unusedcode.easy   |9 ---
  13 files changed, 0 insertions(+), 97 deletions(-)

diff --git a/sc/inc/scdll.hxx b/sc/inc/scdll.hxx
index 93cff7e..43a9c9f 100644
--- a/sc/inc/scdll.hxx
+++ b/sc/inc/scdll.hxx
@@ -61,7 +61,6 @@ public:

  // DLL-init/exit-code must be linked to the DLL only
  static void Init(); // called directly after loading the DLL
-static void Exit(); // called directly befor unloading the DLL

  static sal_uLongDetectFilter( SfxMedium&  rMedium, const
SfxFilter** ppFilter,
  SfxFilterFlags nMust,
SfxFilterFlags nDont );
diff --git a/sc/source/core/inc/ddelink.hxx b/sc/source/core/inc/ddelink.hxx
index e3794e4..216d6d6 100644
--- a/sc/source/core/inc/ddelink.hxx
+++ b/sc/source/core/inc/ddelink.hxx
@@ -87,7 +87,6 @@ public:
  const String&GetItem() const { return aItem; }
  sal_uInt8   GetMode() const { return nMode; }

-voidResetValue();   // Wert zuruecksetzen
  voidTryUpdate();

  sal_BoolNeedsUpdate() const { return bNeedUpdate; }
diff --git a/sc/source/core/tool/ddelink.cxx b/sc/source/core/tool/ddelink.cxx
index 3bd75ae..5d36e52 100644
--- a/sc/source/core/tool/ddelink.cxx
+++ b/sc/source/core/tool/ddelink.cxx
@@ -228,17 +228,6 @@ sfx2::SvBaseLink::UpdateResult ScDdeLink::DataChanged(
  return SUCCESS;
  }

-void ScDdeLink::ResetValue()
-{
-pResult.reset();
-
-//  Es hat sich was getan...
-//  Tracking, FID_DATACHANGED etc. passiert von aussen
-
-if (HasListeners())
-Broadcast( ScHint( SC_HINT_DATACHANGED, ScAddress(), NULL ) );
-}
-
  void ScDdeLink::ListenersGone()
  {
  sal_Bool bWas = bIsInUpdate;
diff --git a/sc/source/ui/app/scdll.cxx b/sc/source/ui/app/scdll.cxx
index 9cf9555..0a57bf6 100644
--- a/sc/source/ui/app/scdll.cxx
+++ b/sc/source/ui/app/scdll.cxx
@@ -312,16 +312,6 @@ void ScDLL::Init()
  //  StarOne Services are now handled in the registry
  }

-void ScDLL::Exit()
-{
-// the SxxModule must be destroyed
-ScModule **ppShlPtr = (ScModule**) GetAppData(SHL_CALC);
-delete (*ppShlPtr);
-(*ppShlPtr) = NULL;
-
-//  ScGlobal::Clear ist schon im Module-dtor
-}
-
  //--
  //  Statusbar
  //--
diff --git a/sc/source/ui/dbgui/fieldwnd.cxx b/sc/source/ui/dbgui/fieldwnd.cxx
index 40eb873..9139be4 100644
--- a/sc/source/ui/dbgui/fieldwnd.cxx
+++ b/sc/source/ui/dbgui/fieldwnd.cxx
@@ -254,11 +254,6 @@ size_t ScDPFieldControlBase::GetSelectedField() const
  return mnFieldSelected;
  }

-void ScDPFieldControlBase::SetSelectedField(size_t nSelected)
-{
-mnFieldSelected = nSelected;
-}
-
  vector&  
ScDPFieldControlBase::GetFieldNames()
  {
  return maFieldNames;
diff --git a/sc/source/ui/inc/areasave.hxx b/sc/source/ui/inc/areasave.hxx
index c9ff348..1a0b1cd 100644
--- a/sc/source/ui/inc/areasave.hxx
+++ b/sc/source/ui/inc/areasave.hxx
@@ -78,7 +78,6 @@ public:

  const ScAreaLinkSaver* operator[](size_t nIndex) const;
  size_t size() const;
-void clear();
  void push_back(ScAreaLinkSaver* p);
  };

diff --git a/sc/source/ui/inc/cellmergeoption.hxx
b/sc/source/ui/inc/cellmergeoption.hxx
index ad5fe34..62e4279 100644
--- a/sc/source/ui/inc/cellmergeoption.hxx
+++ b/sc/source/ui/inc/cellmergeoption.hxx
@@ -44,7 +44,6 @@ struct ScCellMergeOption
  SCROW mnEndRow;
  bool mbCenter;

-explicit ScCellMergeOption();
  explicit ScCellMergeOption(const ScRange&  rRange);
  explicit ScCellMergeOpt

Re: [Libreoffice] [PUSHED^2] Re: [REVIEWED] Fix reading .doc comments (fdo#45255)

[Petr Mladek]

Michael Stahl píše v Po 30. 01. 2012 v 22:42 +0100:
> On 30/01/12 21:19, Caolán McNamara wrote:
> > On Mon, 2012-01-30 at 19:52 +, Michael Meeks wrote:
> >>Looks lovely to me; pushed to -3-5, Caolan are you happy for 3.5.0 ? we
> >> had busted comment interop with .docx in 3.4 (so people used .doc) :-)
> > 
> > Yeah, makes sense to me. +1
> > 
> > C.
> 
> doesn't look obviously wrong to me, so pushed to libreoffice-3-5-0

Cool, it fixed the new blocker
https://bugs.freedesktop.org/show_bug.cgi?id=45409

Best Regards,
Petr

___
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 45409, which changed state.

Bug 45409 Summary: FILEOPEN particular .doc: Table of contents broken, internal 
reference information visible
https://bugs.freedesktop.org/show_bug.cgi?id=45409

   What|Old Value   |New Value

 Resolution||FIXED
 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

[Libreoffice] Simplified C++ configuration access

[Stephan Bergmann]

Hi all,

configmgr is a dreaded beast.  Its UNO API is notorious for re-using 
each and every generic UNO interface type that exists, in a way that 
makes it hard to see the forest for the trees.  To add insult to injury, 
people always thought it a good idea to wrap the API in C++ "helper 
classes" that are themselves baroque and obscure 
(utl::ConfigItem::{EnableNotification,DisableNotification,IsInValueChange} 
-- wtf?).


So I finally set out to simplify this (on the master branch towards LO 3.6):

At the lowest level, there are two new UNO entities, a singleton 
css.configuration.ReadOnlyAccess (of type 
css.conainer.XHierarchicalNameAccess) and a service 
css.configuration.ReadWriteAccess (of type 
css.configuration.XReadWriteAccess, subsuming 
css.container.XHierarchicalNameReplace and css.util.XChangesBatch).


These simplify access to a given configuration item (node, set, or 
property) by name.  Modification still operates in batch mode.  Each 
instance of the ReadWriteAccess service creates its own batch context, 
and modifications done through that instance's XHierarchicalNameReplace 
need to be committed via its XChangesBatch.commitChanges method.


At the next level, the new unotools/configuration.hxx header provides 
type-safe C++ wrappers for the various configuration entities on top of 
the lower level UNO services.  For example, the 
utl::ConfigurationProperty class template can be used to wrap a given 
configuration property of a specific type in a C++ class whose get and 
set member functions internally handle the conversion between the UNO 
API's generic Any and the concrete type of the given property.  There is 
also a utl::ConfigurationChanges class that wraps an instance of 
ReadWriteAccess for (batch) modification.


At the top level, new headers automatically generated in module 
officecfg provide such C++ wrappers for all of the static configuration 
structure as described by the officecfg/registry/schema/**.xcs files.


For each configuration property (with path 
/org.openoffice.Foo.Bar/Baz/Whatever) there is a C++ class 
officecfg::Foo::Bar::Baz::Whatever (declared in header 
officecfg/Foo/Bar.hxx) based on the utl::ConfigurationProperty template. 
 It is type safe (see above) and also prevents mistypings of 
configuration paths (as they are represented by qualified C++ names 
here, not string literals).


To get a property's value, all that is needed is to call the wrapper 
class's static officecfg::Foo::Bar::Baz::Whatever::get member function, 
which requires the XComponentContext as argument (use 
comphelper::getProcessComponentContext in code that does not thread the 
component context more locally).


Setting a property's value is three-step.  First obtain a 
ConfigurationChanges instance (via static member function 
utl::ConfigurationChanges::create), then call 
officecfg::Foo::Bar::Baz::Whatever::set (with the component context and 
the given ConfigurationChanges instance in addition to the new value), 
then call commit on the given ConfigurationChanges instance.


Localized configuration properties are currently treated the same as 
non-localized ones, the wrappers only give access to the value for the 
current locale.


Nillable configuration properties complicate things, in that the wrapper 
needs to use boost::optional instead of plain T (e.g., 
boost::optional vs. rtl::OUString for a string-typed 
property).  This would have been quite a nuisance as most properties 
were implicitly declared as nillable in the .xcs files (a missing 
oor:nillable attribute means oor:nillable="true"), even though in 
practice they never were used with nil values.  I therefore changed all 
those properties that already define a default  in 
officecfg/registry/schema/**.xcs to an explicit oor:nillable="false". 
Even though, strictly speaking, this change is incompatible, I think it 
is warranted, as it more cleanly reflects the current use of those 
properties, anyway, and should not make a difference in practice.


(One caveat though is the rare case of a non-nillable property that does 
not have a default value defined in .xcs.  While this is technically 
legal, it will cause the C++ wrapper's get function to throw a 
RuntimeException upon retrieving an empty (i.e., void) Any from 
configmgr and trying to convert it to the given type.  To avoid this, 
either specify a default value in .xcs or make sure there's always a 
value defined in .xcu.)


There are also C++ wrapper classes for all the configuration's groups 
and sets, but they are probably needed less often (and only make 
available the underlying UNO interfaces, at least for now).  One case 
where they are necessary is to add listeners to (part of) the 
configuration.  For example, to listen for changes to certain properties 
within a given configuration group, one can obtain the group's 
XMultiPropertySet via the corresponding C++ wrapper class's static get 
member function and directly add an appropriate 
XPropertiesChangeLi

Re: [Libreoffice] [PATCH] SIMIAN: Reduced duplicate code

[Noel Power]

On 31/01/12 09:11, Chr. Rossmanith wrote:

Hi,

please have a look at the comments marked with "crossmanith". I can 
push on my own but a short review (and feedback) would be nice.


isn't it that  nInc can only be either 2, 4, 10 or 20 so that line 
indicated can't afaics be reached
and nSteps would seem to have a minimum of 2 imposed on it, looks like a 
typo ( or the limit was changed and not the comment or vice-versa )


patch looks fine to me, I wonder though why you changed

-if ( !bLinear )
+if ( rGradient.GetStyle() == GRADIENT_AXIAL )



Noel

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

[Libreoffice] [PATCH] Removed unused ImplSalLogFontToFontA()

[Chr. Rossmanith]

Hi,

ImplSalLogFontToFontA() seems to be unused. I don't have a Windows build 
environment I couldn't test whether it really builds after removing the 
method. But I don't see why it shouldn't...


Christina
>From 053a691750697916fd5cb176bec886b6da13b630 Mon Sep 17 00:00:00 2001
From: Christina Rossmanith 
Date: Tue, 31 Jan 2012 11:40:22 +0100
Subject: [PATCH] Removed unused ImplSalLogFontToFontA()

---
 vcl/inc/win/saldata.hxx|1 -
 vcl/win/source/gdi/salgdi3.cxx |   39 ---
 2 files changed, 0 insertions(+), 40 deletions(-)

diff --git a/vcl/inc/win/saldata.hxx b/vcl/inc/win/saldata.hxx
index 2d6bbf1..cc3a9e4 100644
--- a/vcl/inc/win/saldata.hxx
+++ b/vcl/inc/win/saldata.hxx
@@ -239,7 +239,6 @@ sal_Bool ImplSalPreDispatchMsg( MSG* pMsg );
 void ImplSalPostDispatchMsg( MSG* pMsg, LRESULT nDispatchResult );
 
 // \WIN\SOURCE\GDI\SALGDI3.CXX
-void ImplSalLogFontToFontA( HDC hDC, const LOGFONTA& rLogFont, Font& rFont );
 void ImplSalLogFontToFontW( HDC hDC, const LOGFONTW& rLogFont, Font& rFont );
 bool ImplIsFontAvailable( HDC hDC, const UniString& rName );
 
diff --git a/vcl/win/source/gdi/salgdi3.cxx b/vcl/win/source/gdi/salgdi3.cxx
index 6b6b917..692005d 100644
--- a/vcl/win/source/gdi/salgdi3.cxx
+++ b/vcl/win/source/gdi/salgdi3.cxx
@@ -1011,45 +1011,6 @@ static ImplWinFontData* ImplLogMetricToDevFontDataW( const ENUMLOGFONTEXW* pLogF
 
 // ---
 
-void ImplSalLogFontToFontA( HDC hDC, const LOGFONTA& rLogFont, Font& rFont )
-{
-String aFontName( ImplSalGetUniString( rLogFont.lfFaceName ) );
-if ( aFontName.Len() )
-{
-rFont.SetName( aFontName );
-rFont.SetCharSet( ImplCharSetToSal( rLogFont.lfCharSet ) );
-rFont.SetFamily( ImplFamilyToSal( rLogFont.lfPitchAndFamily ) );
-rFont.SetPitch( ImplLogPitchToSal( rLogFont.lfPitchAndFamily ) );
-rFont.SetWeight( ImplWeightToSal( rLogFont.lfWeight ) );
-
-long nFontHeight = rLogFont.lfHeight;
-if ( nFontHeight < 0 )
-nFontHeight = -nFontHeight;
-long nDPIY = GetDeviceCaps( hDC, LOGPIXELSY );
-if( !nDPIY )
-nDPIY = 600;
-nFontHeight *= 72;
-nFontHeight += nDPIY/2;
-nFontHeight /= nDPIY;
-rFont.SetSize( Size( 0, nFontHeight ) );
-rFont.SetOrientation( (short)rLogFont.lfEscapement );
-if ( rLogFont.lfItalic )
-rFont.SetItalic( ITALIC_NORMAL );
-else
-rFont.SetItalic( ITALIC_NONE );
-if ( rLogFont.lfUnderline )
-rFont.SetUnderline( UNDERLINE_SINGLE );
-else
-rFont.SetUnderline( UNDERLINE_NONE );
-if ( rLogFont.lfStrikeOut )
-rFont.SetStrikeout( STRIKEOUT_SINGLE );
-else
-rFont.SetStrikeout( STRIKEOUT_NONE );
-}
-}
-
-// ---
-
 void ImplSalLogFontToFontW( HDC hDC, const LOGFONTW& rLogFont, Font& rFont )
 {
 XubString aFontName( reinterpret_cast(rLogFont.lfFaceName) );
-- 
1.7.4.1

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

Re: [Libreoffice] [PATCH] SIMIAN: Reduced duplicate code

[Chr. Rossmanith]

Am 31.01.2012 11:43, schrieb Noel Power:

On 31/01/12 09:11, Chr. Rossmanith wrote:
please have a look at the comments marked with "crossmanith". I can 
push on my own but a short review (and feedback) would be nice.


isn't it that  nInc can only be either 2, 4, 10 or 20 so that line 
indicated can't afaics be reached

I agree. I'll remove the unreachable lines.
and nSteps would seem to have a minimum of 2 imposed on it, looks like 
a typo ( or the limit was changed and not the comment or vice-versa )
Then I'll modify the comment in a follow-up commit along with some 
translations.


patch looks fine to me, I wonder though why you changed

-if ( !bLinear )
+if ( rGradient.GetStyle() == GRADIENT_AXIAL )
I struggled because in ImplDrawLinearGradient() there is an if statement 
"if (!bLinear)" (why should bLinear be false if drawing a linear 
gradient?) then I had a look where this method is called and found that 
it is used if rGradient.GetStyle() is AXIAL or LINEAR. I think testing 
for AXIAL is more intuitive than !bLinear.


Christina

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

Re: [Libreoffice] [PATCH] Remove unused code

[Elton Chung]

Hi Noel, *,
I have attached the patch as an attachment. I also confirm that this
contribution, as well as future ones are licensed under MPL 1.1 / GPL
v3+ / LGPL v3+.

Best Regards,
Elton

--
Elton Chung
Administrator, sponsor of mirror.layerjet.com
Language Maintainer of ReactOS Open Source operating system project |
http://reactos.org

Email/ IM: el...@elton.tk


0001-Remove-unused-code.patch
Description: Binary data
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PATCH] Removed unused ImplSalLogFontToFontA()

[Noel Power]

Hi Christina
On 31/01/12 10:43, Chr. Rossmanith wrote:

Hi,

ImplSalLogFontToFontA() seems to be unused. I don't have a Windows 
build environment I couldn't test whether it really builds after 
removing the method. But I don't see why it shouldn't...


Christina
I'd say commit it, can't see this is used, anyway cc'ing Caolán as he 
might be interested because it doesn't turn up in the unused.easy



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

Re: [Libreoffice] [PUSHED][PATCH] Removed unused ImplSalLogFontToFontA()

[Ivan Timofeev]

31.01.2012 15:15, Noel Power пишет:

Hi Christina
On 31/01/12 10:43, Chr. Rossmanith wrote:

Hi,

ImplSalLogFontToFontA() seems to be unused. I don't have a Windows
build environment I couldn't test whether it really builds after
removing the method. But I don't see why it shouldn't...

Christina

I'd say commit it, can't see this is used, anyway cc'ing Caolán as he
might be interested because it doesn't turn up in the unused.easy


I've just pushed it.

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

Re: [Libreoffice] [PUSHED] partial patch for fdo#44551 libvisio: arrow head shapes not imported properly

[Fridrich Strba]

Hello, Bruno

On 31/01/12 00:35, Bruno Girin wrote:

Please find attached a partial patch to import line marker shapes from
Visio files. I would appreciate a review to confirm whether I am on the
right path or whether I should do something different.


Yes, currently this is the best approach to fill in for each id the 
right viewbox and path. I will have to find a way to parse it for the 
svg generation, but that is another question. That was an issue with the 
initial implementation too, but nothing unsolvable either.


Thanks for your great work.

Fridrich


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


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

Re: [Libreoffice] [REVIEW]Fix for annoying bug fdo#35669

[Noel Power]

On 30/01/12 17:16, Cedric Bosdonnat wrote:

Hi Michael,

On Thu, 2012-01-26 at 19:35 +0100, Michael Stahl wrote:

On 26/01/12 19:30, Michael Stahl wrote:

On 26/01/12 16:21, Cedric Bosdonnat wrote:

Hi all,

Could someone review and cherry-pick this commit to -3-5, -3-5-0 and
-3-4?
http://cgit.freedesktop.org/libreoffice/core/commit/?id=5e51960dede5015b862df05b7b16f02884647889

your fix looks good to me, except for the line:

   sal_uInt16 nMaxDstId = *aIds.end();

oops, now it occurs to me that dereferencing end() is a bad idea in any
case, you probably want to use rbegin() instead :)


if i'm reading this right, then aIds could be empty here, if the source
document has seq field references, and the target document has no seq
field references but note references.

suggest to wrap an "if (!aIds.empty())" around the whole block?
no need to adjust anything in that case, right?

Fixed in
http://cgit.freedesktop.org/libreoffice/core/commit/?id=458d56d9b35f2ecb2eea2589737079390b10ee26

Would be nice to cherry-pick these two commits to -3-5, -3-4 and 3-5-0
(if it's not too late).


the change looks good to me now too

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

Re: [Libreoffice] Simplified C++ configuration access

[Michael Meeks]

Hi Stephan,

On Tue, 2012-01-31 at 11:31 +0100, Stephan Bergmann wrote:
> To add insult to injury, people always thought it a good idea to
> wrap the API in C++ "helper classes" that are themselves baroque
> and obscure 

:-)

> So I finally set out to simplify this (on the master branch towards LO 3.6):

It's certainly a really beautiful result (modulo a lingering concern
about the context parameter of inscrutable purpose ;-)

> intend to continue with that -- until finally things like utl::ConfigItem
> can be removed.  Let me know if you would like to help with that.

Sounds like a great cleanup, I vaguely advertised it in my blog. Would
it be possible to have a simple "convert just one-key" patch to some old
code so newbies can get their cookie cutters correctly calibrated :-)

Thanks again !

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] [Libreoffice-ux-advise] [PATCH] [PUSHED] fdo#45285 - Use more revealing error message for not allowed sheet names

[Michael Meeks]

On Mon, 2012-01-30 at 20:40 +0100, Albert Thuswaldner wrote:
> Just a thought: should we push this to the 3.5 branch? This is a small
> polish of the user experience and it might be good to have it in a
> release soon

Well - hopefully six months away is not -so- far - but I appreciate
your pain. Of course - in theory with triple review features can get
into the -3-5 branch, but string changes are less welcome I suspect.

Perhaps the pain could be softened by adding the feature to the 3.6
features page: :-)

http://wiki.documentfoundation.org/ReleaseNotes/3.6

Interestingly, IIRC I asked someone to do this for some other feature
already, but it didn't make it there yet: could you be the first ?

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] contour handling for anchored draw objects in writer (fix for bugs 34611 and 45376)

[Noel Power]

On 30/01/12 22:53, Andreas Schierl wrote:

Hi, I found out that bugs
   https://bugs.freedesktop.org/show_bug.cgi?id=34611
   https://bugs.freedesktop.org/show_bug.cgi?id=45376
are caused by a missing call to
ClrContourCache( pObj );
in
/sw/source/core/layout/pagechg.cxx:
void lcl_MoveAllLowerObjs( SwFrm* pFrm, const Point&  rOffset )
in the
else if ( pAnchoredObj->ISA(SwAnchoredDrawObject) )
case (in the if case, pFlyFrm->NotifyDrawObj() calls ClrContourCache).

The attached patch fixes that problem.
seems like pretty impressive piece of detective work to find that, great 
work!! I've committed the patch to master ( bug doc works fine for me ) 
however since I know precious little about writer ( and in this case 
some layout stuff too ) I cc some writer experts who may suggest some 
alternative solution(s) or change ( or maybe say it is perfect )
Can you confirm your contribution is under the MPL 1.1 / GPLv3+ / 
LGPLv3+ license(s)


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

Re: [Libreoffice] Storage of user's configuration

[Michael Meeks]

Hi Stephan,

On Tue, 2012-01-31 at 08:28 +0100, Stephan Bergmann wrote:
> Martin, for this to keep working please add a nillable="false" attribute 
> to your new property in Common.xcs, see 
...
> "Fixed cppheader.xsl nillable treatment."  (In addition to adding an 
> empty default  so that the get() above does not throw an 
> exception upon a void Any, as discussed on IRC the other day.)

Incidentally - do we expect that all properties will end up with a
nillable="false" attribute on them in the end ? :-)

I ask mostly out of interest, since I'd love to shrink the size of the
XML we're parsing there - every byte counts (unfortunately) when you
have >16k oox:prop statements. Then again, if we end up with it being
false everywhere - we can perhaps flip the default for 4.0 ? :-)

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] [REVIEW][3-5] Prevent excessvie references to formula result tokens

[Michael Meeks]

On Mon, 2012-01-30 at 16:35 -0500, Kohei Yoshida wrote:
> increases its reference counter by one.  The problem is, this counter is
> unsigned 16-bit integer, and as soon as it goes above 65535 it rolls
> back to zero, and eventually the token instance gets deleted
> prematurely.

Wow - that is annoying :-)

> As an aside, although it's not necessary for this fix, on master we
> should probably use unsigned 32-bit integer to store reference counter
> for this just to future-proof ourselves.  16-bit integer seems a bit too
> small for this purpose.

Right - although, of course what we really want is the sexy region
based re-design, so that we don't hit the next limit: 16k cols x 1m rows
is > 4G items too ;-) so ... clearly we need to share a lot more in such
cases.

Interesting bug though :-)

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] [PUSHED] [3.5.0] Some simple polish for named ranges dialogue

[Andras Timar]

2012/1/31 Noel Power :
> On 30/01/12 19:21, Petr Mladek wrote:
>>
>> Stefan Knorr (Astron) píše v Po 30. 01. 2012 v 09:45 +0100:
>>>
>>> Hi,
>>>
>>> I know this is extremely late for a simple polish fix, but please
>>> consider
>>>
>>> http://cgit.freedesktop.org/libreoffice/core/commit/?id=f1cb0a4ab4f11dc015be1696c7c7751802171915
>>> as well as
>>>
>>> http://cgit.freedesktop.org/libreoffice/core/commit/?id=0b6486772c52d7112e76555fca6525ad2eec2bae
>>> for inclusion in 3.5.(0).
>>> Both changes concern almost exclusively src files, so they should be
>>> pretty safe.
>>> (I accidentally pushed early, that's the only reason why it's split up
>>> into two commits.)
>>
>> There was still not enough space for German strings in the "Define name"
>> dialog, see the attached screenshots. So, I have increased it a bit
>> more, see
>>
>> http://cgit.freedesktop.org/libreoffice/core/commit/?id=5be59017ca21301e70eca987e8323c92215f0ed4
>>
>>
>> I merged all 3 commits together and pushed to 3-5 branch, see
>>
>> http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5&id=b15d3b50ba468defd3109e5ec0aef331e222cb5e
>>
>>
>> Two more appovals are needed for the 3-5-0 branch.
>>
> +1 from me
>
+1 from me and pushed to libreoffice-3-5-0.

Best regards,
Andras
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PATCH] Remove unused code

[Noel Power]

On 31/01/12 11:05, Elton Chung wrote:

Hi Noel, *,
I have attached the patch as an attachment. I also confirm that this
contribution, as well as future ones are licensed under MPL 1.1 / GPL
v3+ / LGPL v3+.

ok, pushed, thank, ps. next time don't chop off the header of the 
formatted patch 'cause 'git am' had problems processing the patch, 
anyway not to worry as 'patch' still accepted it and I added the author 
and description manually


thanks again

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

Re: [Libreoffice] Storage of user's configuration

[Stephan Bergmann]

On 01/31/2012 12:37 PM, Michael Meeks wrote:

Incidentally - do we expect that all properties will end up with a
nillable="false" attribute on them in the end ? :-)


No.

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

[Libreoffice] [PATCH] unusedcode.easy: Removed unused code (basegfx)

[Alexander Bergmann]

---
 basegfx/inc/basegfx/matrix/b2dhommatrixtools.hxx |3 -
 basegfx/inc/basegfx/polygon/b2dpolygon.hxx   |   20 
 basegfx/inc/basegfx/polygon/b3dpolygon.hxx   |3 -
 basegfx/inc/basegfx/polygon/b3dpolypolygon.hxx   |3 -
 basegfx/inc/basegfx/range/b1drange.hxx   |8 ---
 basegfx/inc/basegfx/range/b2drange.hxx   |6 --
 basegfx/inc/basegfx/range/b3drange.hxx   |8 ---
 basegfx/inc/basegfx/tools/canvastools.hxx|   10 
 basegfx/inc/basegfx/tools/unopolypolygon.hxx |3 -
 basegfx/inc/basegfx/tuple/b2i64tuple.hxx |2 -
 basegfx/inc/basegfx/tuple/b2ituple.hxx   |2 -
 basegfx/inc/basegfx/tuple/b3dtuple.hxx   |7 ---
 basegfx/inc/basegfx/tuple/b3i64tuple.hxx |2 -
 basegfx/inc/basegfx/tuple/b3ituple.hxx   |2 -
 basegfx/inc/basegfx/vector/b2dvector.hxx |8 ---
 basegfx/inc/basegfx/vector/b2ivector.hxx |   14 -
 basegfx/source/matrix/b2dhommatrixtools.cxx  |   26 --
 basegfx/source/polygon/b2dpolygon.cxx|   24 -
 basegfx/source/polygon/b3dpolygon.cxx|5 --
 basegfx/source/polygon/b3dpolypolygon.cxx|   17 ---
 basegfx/source/range/b1drange.cxx|   18 ---
 basegfx/source/range/b2drange.cxx|7 ---
 basegfx/source/range/b3drange.cxx|   29 ---
 basegfx/source/tools/canvastools.cxx |   56 --
 basegfx/source/tools/unopolypolygon.cxx  |5 --
 basegfx/source/tuple/b2i64tuple.cxx  |5 --
 basegfx/source/tuple/b2ituple.cxx|5 --
 basegfx/source/tuple/b3dtuple.cxx|6 --
 basegfx/source/tuple/b3i64tuple.cxx  |4 --
 basegfx/source/tuple/b3ituple.cxx|4 --
 basegfx/source/vector/b2dvector.cxx  |8 ---
 basegfx/source/vector/b2ivector.cxx  |6 --
 unusedcode.easy  |   25 --
 33 files changed, 0 insertions(+), 351 deletions(-)

diff --git a/basegfx/inc/basegfx/matrix/b2dhommatrixtools.hxx 
b/basegfx/inc/basegfx/matrix/b2dhommatrixtools.hxx
index 3d0d22d..e53a598 100644
--- a/basegfx/inc/basegfx/matrix/b2dhommatrixtools.hxx
+++ b/basegfx/inc/basegfx/matrix/b2dhommatrixtools.hxx
@@ -226,9 +226,6 @@ namespace basegfx
 };
 } // end of namespace tools
 
-/// Returns a string with svg's "matrix(m00,m10,m01,m11,m02,m12)" 
representation
-::rtl::OUString exportToSvg( const B2DHomMatrix& rMatrix );
-
 } // end of namespace basegfx
 
 ///
diff --git a/basegfx/inc/basegfx/polygon/b2dpolygon.hxx 
b/basegfx/inc/basegfx/polygon/b2dpolygon.hxx
index 707590d..3443d72 100644
--- a/basegfx/inc/basegfx/polygon/b2dpolygon.hxx
+++ b/basegfx/inc/basegfx/polygon/b2dpolygon.hxx
@@ -187,26 +187,6 @@ namespace basegfx
 */
 B2DRange getB2DRange() const;
 
-/** insert other 2D polygons
-
-The default (with nIndex2 == 0 && nCount == 0) inserts the whole
-rPoly at position nIndex
-
-@param nIndex
-Target index for points to be inserted
-
-@param rPoly
-The source for new points
-
-@param nIndex2
-The index to the first source point into rPoly
-
-@param nCount
-How many points to add from rPoly to this polygon. Null
-means to copy all (starting from nIndex2)
-*/
-void insert(sal_uInt32 nIndex, const B2DPolygon& rPoly, sal_uInt32 
nIndex2 = 0, sal_uInt32 nCount = 0);
-
 /** append other 2D polygons
 
 The default (nIndex ==0 && nCount == 0) will append
diff --git a/basegfx/inc/basegfx/polygon/b3dpolygon.hxx 
b/basegfx/inc/basegfx/polygon/b3dpolygon.hxx
index 359adf2..2aae89b 100644
--- a/basegfx/inc/basegfx/polygon/b3dpolygon.hxx
+++ b/basegfx/inc/basegfx/polygon/b3dpolygon.hxx
@@ -69,9 +69,6 @@ namespace basegfx
 // assignment operator
 B3DPolygon& operator=(const B3DPolygon& rPolygon);
 
-/// unshare this polygon with all internally shared instances
-void makeUnique();
-
 // compare operators
 bool operator==(const B3DPolygon& rPolygon) const;
 bool operator!=(const B3DPolygon& rPolygon) const;
diff --git a/basegfx/inc/basegfx/polygon/b3dpolypolygon.hxx 
b/basegfx/inc/basegfx/polygon/b3dpolypolygon.hxx
index 98ba7cc..fab75a1 100644
--- a/basegfx/inc/basegfx/polygon/b3dpolypolygon.hxx
+++ b/basegfx/inc/basegfx/polygon/b3dpolypolygon.hxx
@@ -101,9 +101,6 @@ namespace basegfx
 // reset to empty state
 void clear();
 
-// closed state
-bool isClosed() const;
-
 // flip polygon direction
 void flip();
 
diff --git a/basegfx/inc/basegfx/range/b1drange.hxx 
b/basegfx/inc/basegfx/range/b1drange.hxx
index cac1efb..9

Re: [Libreoffice] [PATCH] contour handling for anchored draw objects in writer (fix for bugs 34611 and 45376)

[openoffice]

Noel Power wrote:
> I cc some writer experts who may suggest some
> alternative solution(s) or change ( or maybe say it is perfect )

The new patch in this e-mail should provide even better performance
(as it only clears the contour cache for contour wrapped drawing objects).

> Can you confirm your contribution is under the MPL 1.1 / GPLv3+ /
> LGPLv3+ license(s)

Yes, the contribution is licensed under  MPL 1.1 / GPLv3+ / LGPLv3+

Andreas

0001-ClrContourCache-for-SwAnchoredDrawObject.patch
Description: Binary data
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] Fatal error building Windows release (Cygwin)

[walter]

I followed all the instructions in
wiki.documentfoundation.org/Development/Native_Build and in
wiki.documentfoundation.org/Development/Windows_Build_Dependencies

but I’m unable to build the release.

 

Full logs (including getting the source) at pastebin.com/u/walterLO

 

Release libreoffice 3-5

 

$ git show

commit 5de902f2b66e64bc4b4755356db3b259c01ddcdd

Author: Cédric Bosdonnat 

Date:   Mon Jan 30 11:35:42 2012 +0100

 

 

 

Regards

 

walter

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

Re: [Libreoffice] Simplified C++ configuration access

[Stephan Bergmann]

On 01/31/2012 12:23 PM, Michael Meeks wrote:

Sounds like a great cleanup, I vaguely advertised it in my blog. Would
it be possible to have a simple "convert just one-key" patch to some old
code so newbies can get their cookie cutters correctly calibrated :-)


 
is an example of reading multiple properties, among them a nillable one.


 
is a corresponding example of setting those properties.


And 
 
is an example of a changes listener for those properties.


All three got in with 
 
"Replace SourceViewConfig with (simplified) direct configuration access" 
(plus 
 
"Fixed cppheader.xsl nillable treatment").


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

Re: [Libreoffice] Fatal error building Windows release (Cygwin)

[Noel Grandin]

looks like it's failing in the xmlreader module (part6 of your pastebin 
logs), but I don't know why.


try doing this:

source ./Env.Host.sh
cd xmlreader
build


On 2012-01-31 14:59, walter wrote:


I followed all the instructions in  
wiki.documentfoundation.org/Development/Native_Build and in 
wiki.documentfoundation.org/Development/Windows_Build_Dependencies


but I'm unable to build the release.

Full logs (including getting the source) at pastebin.com/u/walterLO

Release libreoffice 3-5

$ git show

commit 5de902f2b66e64bc4b4755356db3b259c01ddcdd

Author: Cédric Bosdonnat 

Date:   Mon Jan 30 11:35:42 2012 +0100

Regards

walter



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


Disclaimer: http://www.peralex.com/disclaimer.html


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

Re: [Libreoffice] [PATCH] Removed unused ImplSalLogFontToFontA()

[Caolán McNamara]

On Tue, 2012-01-31 at 11:15 +, Noel Power wrote:
> cc'ing Caolán as he might be interested because it doesn't turn up in
> the unused.easy

Its windows only code, and callcatcher can only see what gets compiled
as its just a compiler wrapper and I run it under Linux.

C.

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

[Libreoffice] tracking ODF compliance issues

[Michael Stahl]

hi all,

i'd like to have some easier way to track bugs related to ODF compliance
in the bugzilla.

what would be the best way to do this:
- a new keyword, say "odf" or "odf_validation"
- a tracking bug, that depends on all indidual ODF related bugs

the OpenOffice.org issuezilla used a "odf_validation" keyword for issues
where OOo wrote invalid ODF, but there was no way to track other kinds
of ODF issues.

so it would also be possible to have say 2 keywords "odf_validation" for
those issues and "odf" for everything else, or 2 tracking issues...

any opinions?

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

Re: [Libreoffice] [REVIEW][3-5] Prevent excessvie references to formula result tokens

[Eike Rathke]

Hi Kohei,

On Monday, 2012-01-30 16:35:46 -0500, Kohei Yoshida wrote:

> I'd like
> 
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=e2b11f4fd79dce4116badb0ecf6477546ca5d0d4
> 
> cherry-picked to the 3-5 branch.  It is probably too late for change
> like this to be in the 3-5-0 branch so I won't even try.

I'd like to ask for an improvement ;)  see below.


> To reproduce the problem, open the following file in Calc
> 
> http://people.freedesktop.org/~kohei/test-formula-fill-crash.ods
> 
> then
> 
> 1. select B1:B65536 (or just hit ctrl-shift-up)
> 2. fill down (or ctrl-D if you use the default key binding).
> 
> That will currently either crash, or do a totally wrong thing.  If it
> doesn't crash, try undo and redo and it will eventually crash.
> 
> The reason is that, during the fill, the formula token instance inside
> ScFormulaResult gets "copied" i.e. it re-uses the existing instance and
> increases its reference counter by one.  The problem is, this counter is
> unsigned 16-bit integer, and as soon as it goes above 65535 it rolls
> back to zero, and eventually the token instance gets deleted
> prematurely.

Argh, yes, right.

> The above change ensures that the formula result is cleared after each
> formula cell instance gets copied.  We don't need to copy the formula
> result during fill because they get re-interpreted once the copying is
> complete.

True. Though the patch fixes this only for the Fill case, the same crash
happens when copying B1 to clipboard, selecting B2:B65536 and paste.
That needs to be handled similar. Probably best would be to revert this
commit and change the ScFormulaResult copy-ctor to create a clone of the
token instead so one doesn't need to bother with it at various places.

> As an aside, although it's not necessary for this fix, on master we
> should probably use unsigned 32-bit integer to store reference counter
> for this just to future-proof ourselves.  16-bit integer seems a bit too
> small for this purpose.

I'd not recommend using 32-bit for the token reference count, that would
significantly increase memory footprint of every formula. Single tokens
weren't meant to be shared across formula cells.

  Eike

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


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

Re: [Libreoffice] [PATCH] CSC, SEC, CSCH & SECH Help missing, How to add one? ref: bug fdo#44792

[Regina Henschel]

Olivier Hallot schrieb:

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi Regina

It has been decades since I handled groups, rings, fields, and functions
in my french lycée. Now I am confused because it looks like the two
terms are swapped between french and english...

Regards

Olivier

In french:
http://fr.wikipedia.org/wiki/Cos%C3%A9cante#Fonctions_r.C3.A9ciproques

in English
http://en.wikipedia.org/wiki/Inverse_%28mathematics%29


A better reference for Engish is 
http://en.wikipedia.org/wiki/Sine#Reciprocal


The topic here are functions, not numbers.
For numbers you have the term "additive inverse of x is -x" or 
"multiplicative inverse of x is 1/x". But in context of functions 
"inverse" always means the inverse in respect to composition.


I don't know French. Astonishing, http://en.wikipedia.org/wiki/Sine has 
no direct translation to French.


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

[Libreoffice] R: Fatal error building Windows release (Cygwin)

[walter]

The problem is 

 

Cannot open include file: 'sal/config.h'

 

but this file exists in my 'git/libo/sal/inc/sal' directory !!

 

Regards

Walter

 




 

=

(1/1) Building module xmlreader

=

Entering /home/walter/git/libo/xmlreader/prj

 

[ build CXX ] xmlreader/source/pad

[ build CXX ] xmlreader/source/span

[ build CXX ] xmlreader/source/xmlreader

c:/cygwin/home/walter/git/libo/xmlreader/source/xmlreader.cxx(29) : fatal
error C1083: Cannot open include file: 'sal/config.h': No such file or
directory

c:/cygwin/home/walter/git/libo/xmlreader/source/pad.cxx(29) : fatal error
C1083: Cannot open include file: 'sal/config.h': No such file or directory

c:/cygwin/home/walter/git/libo/xmlreader/source/span.cxx(29) : fatal error
C1083: Cannot open include file: 'sal/config.h': No such file or directory

[ build RES ] xmlreader/default

[ build RES ]

make: *** No rule to make target
`/home/walter/git/libo/workdir/wntmsci12.pro/CxxObject/xmlreader/source/pad.
o', needed by
`/home/walter/git/libo/workdir/wntmsci12.pro/LinkTarget/Library/ixmlreader.l
ib'.  Stop.

make: *** Waiting for unfinished jobs

awk: fatal: can't open source file
`C:/cygwinc:/cygwin/home/walter/git/libo/solenv/gbuild/processdeps.awk' for
reading (No such file or directory)

make: ***
[/home/walter/git/libo/workdir/wntmsci12.pro/WinResTarget/xmlreader/default.
res] Error 2

dmake:  Error code 2, while making 'all'

Retrying /home/walter/git/libo/xmlreader/prj

 

[ build CXX ] xmlreader/source/pad

[ build CXX ] xmlreader/source/span

[ build CXX ] xmlreader/source/xmlreader

c:/cygwin/home/walter/git/libo/xmlreader/source/pad.cxx(29) : fatal error
C1083: Cannot open include file: 'sal/config.h': No such file or directory

c:/cygwin/home/walter/git/libo/xmlreader/source/xmlreader.cxx(29) : fatal
error C1083: Cannot open include file: 'sal/config.h': No such file or
directory

[ build RES ] xmlreader/default

[ build RES ]

c:/cygwin/home/walter/git/libo/xmlreader/source/span.cxx(29) : fatal error
C1083: Cannot open include file: 'sal/config.h': No such file or directory

make: *** No rule to make target
`/home/walter/git/libo/workdir/wntmsci12.pro/CxxObject/xmlreader/source/pad.
o', needed by
`/home/walter/git/libo/workdir/wntmsci12.pro/LinkTarget/Library/ixmlreader.l
ib'.  Stop.

make: *** Waiting for unfinished jobs

awk: fatal: can't open source file
`C:/cygwinc:/cygwin/home/walter/git/libo/solenv/gbuild/processdeps.awk' for
reading (No such file or directory)

make: ***
[/home/walter/git/libo/workdir/wntmsci12.pro/WinResTarget/xmlreader/default.
res] Error 2

dmake:  Error code 2, while making 'all'

 

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

Re: [Libreoffice] [Libreoffice-ux-advise] [PATCH] [PUSHED] fdo#45285 - Use more revealing error message for not allowed sheet names

[Eike Rathke]

Hi Ivan,

On Tuesday, 2012-01-31 12:55:50 +0400, Ivan Timofeev wrote:

> just added ':' to the sequence of characters
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=0a9d8c07ddf054930e1b000cc3c972a3d217d71c

Yup, good catch. Technically there's also the limitation that a ' single
quote character is not allowed as the very first or last character. The
message would get quite lengthy though if we added "and not the
character ' at the beginning or end".

  Eike

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


pgpxSqG7tUnBH.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] master vs. 3.5 regression ... (vcl/python) ...

[Michael Meeks]

Hi guys,

I just noticed this printing:

https://bugs.freedesktop.org/show_bug.cgi?id=45303

from master; looks like a particularly exciting memory corruption
regression between python and vcl somewhere. Prolly of my own creation,
but I wonder if it rings any bells. Supposedly it is not in 3.5 - which
is a mercy.

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] R: Fatal error building Windows release (Cygwin)

[Noel Grandin]

yeah, the gbuild code that figures out the INCLUDE path is obviously a 
little wonky.


As this point all I can suggest is doing:

cd xmlreader
rm -Rf wntmsci12.pro
build

Sometimes that sorts it out.

On 2012-01-31 15:31, walter wrote:


The problem is

Cannot open include file: 'sal/config.h'

but this file exists in my 'git/libo/sal/inc/sal' directory !!

Regards

Walter



=

(1/1) Building module xmlreader

=

Entering /home/walter/git/libo/xmlreader/prj

[ build CXX ] xmlreader/source/pad

[ build CXX ] xmlreader/source/span

[ build CXX ] xmlreader/source/xmlreader

c:/cygwin/home/walter/git/libo/xmlreader/source/xmlreader.cxx(29) : 
fatal error C1083: Cannot open include file: 'sal/config.h': No such 
file or directory


c:/cygwin/home/walter/git/libo/xmlreader/source/pad.cxx(29) : fatal 
error C1083: Cannot open include file: 'sal/config.h': No such file or 
directory


c:/cygwin/home/walter/git/libo/xmlreader/source/span.cxx(29) : fatal 
error C1083: Cannot open include file: 'sal/config.h': No such file or 
directory


[ build RES ] xmlreader/default

[ build RES ]

make: *** No rule to make target 
`/home/walter/git/libo/workdir/wntmsci12.pro/CxxObject/xmlreader/source/pad.o', 
needed by 
`/home/walter/git/libo/workdir/wntmsci12.pro/LinkTarget/Library/ixmlreader.lib'.  
Stop.


make: *** Waiting for unfinished jobs

awk: fatal: can't open source file 
`C:/cygwinc:/cygwin/home/walter/git/libo/solenv/gbuild/processdeps.awk' for 
reading (No such file or directory)


make: *** 
[/home/walter/git/libo/workdir/wntmsci12.pro/WinResTarget/xmlreader/default.res] 
Error 2


dmake:  Error code 2, while making 'all'

Retrying /home/walter/git/libo/xmlreader/prj

[ build CXX ] xmlreader/source/pad

[ build CXX ] xmlreader/source/span

[ build CXX ] xmlreader/source/xmlreader

c:/cygwin/home/walter/git/libo/xmlreader/source/pad.cxx(29) : fatal 
error C1083: Cannot open include file: 'sal/config.h': No such file or 
directory


c:/cygwin/home/walter/git/libo/xmlreader/source/xmlreader.cxx(29) : 
fatal error C1083: Cannot open include file: 'sal/config.h': No such 
file or directory


[ build RES ] xmlreader/default

[ build RES ]

c:/cygwin/home/walter/git/libo/xmlreader/source/span.cxx(29) : fatal 
error C1083: Cannot open include file: 'sal/config.h': No such file or 
directory


make: *** No rule to make target 
`/home/walter/git/libo/workdir/wntmsci12.pro/CxxObject/xmlreader/source/pad.o', 
needed by 
`/home/walter/git/libo/workdir/wntmsci12.pro/LinkTarget/Library/ixmlreader.lib'.  
Stop.


make: *** Waiting for unfinished jobs

awk: fatal: can't open source file 
`C:/cygwinc:/cygwin/home/walter/git/libo/solenv/gbuild/processdeps.awk' for 
reading (No such file or directory)


make: *** 
[/home/walter/git/libo/workdir/wntmsci12.pro/WinResTarget/xmlreader/default.res] 
Error 2


dmake:  Error code 2, while making 'all'



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


Disclaimer: http://www.peralex.com/disclaimer.html


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

Re: [Libreoffice] Simplified C++ configuration access

[Lubos Lunak]

On Tuesday 31 of January 2012, Stephan Bergmann wrote:
> To get a property's value, all that is needed is to call the wrapper
> class's static officecfg::Foo::Bar::Baz::Whatever::get member function,
> which requires the XComponentContext as argument (use
> comphelper::getProcessComponentContext in code that does not thread the
> component context more locally).

 How usual is the case that one does not want to pass 
comphelper::getProcessComponentContext() to all these functions? Or in other 
words, why not simply make it the default argument value?

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

[Libreoffice] R: R: Fatal error building Windows release (Cygwin)

[walter]

A simple question.

How can you compile The MASTER Win-x86_7-MinGW without any error ?

 

Which environment is used ?

 

Regards

Walter

 

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Power]

On 30/01/12 13:54, Noel Grandin wrote:

Hi

Attached patch converts SdCustomShow to use std::vector.

Code is contributed under MPL+/LGPL+/GPL+

Regards, Noel Grandin

Disclaimer: http://www.peralex.com/disclaimer.html

I hate these remove list etc. type patches as they tend to be tricky and 
time consuming to look at ;-) .  The patch looked ok for me, hopefully 
if I missed something ( 'cause these really are tricky ) then someone 
else will spot it. Anyway thanks again for the nice patch. So... pushed 
to master



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

[Libreoffice] [PATCH] convert SdPresLayoutDlg from tools/list.hxx to std::map

[Noel Grandin]

Hi

Attached patch converts SdPresLayoutDlg to use std::map.

Code is contributed under MPL+/LGPL+/GPL+

Regards, Noel Grandin

Disclaimer: http://www.peralex.com/disclaimer.html


diff --git a/sd/source/ui/dlg/sdpreslt.cxx b/sd/source/ui/dlg/sdpreslt.cxx
index 98f2df5..c8b1495 100644
--- a/sd/source/ui/dlg/sdpreslt.cxx
+++ b/sd/source/ui/dlg/sdpreslt.cxx
@@ -76,8 +76,6 @@ SdPresLayoutDlg::SdPresLayoutDlg(
 {
 FreeResource();
 
-mpLayoutNames = new List;
-
 maVS.SetDoubleClickHdl(LINK(this, SdPresLayoutDlg, ClickLayoutHdl));
 maBtnLoad.SetClickHdl(LINK(this, SdPresLayoutDlg, ClickLoadHdl));
 
@@ -92,14 +90,7 @@ SdPresLayoutDlg::SdPresLayoutDlg(
 
 SdPresLayoutDlg::~SdPresLayoutDlg()
 {
-String* pName = (String*)mpLayoutNames->First();
-while (pName)
-{
-delete pName;
-pName = (String*)mpLayoutNames->Next();
-}
-
-delete mpLayoutNames;
+maLayoutNames.clear();
 }
 
 /*
@@ -131,10 +122,10 @@ void SdPresLayoutDlg::Reset()
 
 FillValueSet();
 
-mnLayoutCount = mpLayoutNames->Count();
+mnLayoutCount = maLayoutNames.size();
 for( nName = 0; nName < mnLayoutCount; nName++ )
 {
-if (*((String*)mpLayoutNames->GetObject(nName)) == maName)
+if (maLayoutNames[nName] == maName)
 break;
 }
 DBG_ASSERT(nName < mnLayoutCount, "Layout nicht gefunden");
@@ -161,11 +152,11 @@ void SdPresLayoutDlg::GetAttr(SfxItemSet& rOutAttrs)
 {
 aLayoutName = maName;
 aLayoutName.Append( DOCUMENT_TOKEN );
-aLayoutName.Append( *(String*)mpLayoutNames->GetObject( nId - 1 ) );
+aLayoutName.Append( maLayoutNames[ nId - 1 ] );
 }
 else
 {
-aLayoutName = *(String*)mpLayoutNames->GetObject( nId - 1 );
+aLayoutName = maLayoutNames[ nId - 1 ];
 if( aLayoutName == maStrNone )
 aLayoutName.Erase(); //  so wird "- keine -" codiert (s.u.)
 }
@@ -202,10 +193,10 @@ void SdPresLayoutDlg::FillValueSet()
 {
 String aLayoutName(pMaster->GetLayoutName());
 aLayoutName.Erase( aLayoutName.SearchAscii( SD_LT_SEPARATOR ) );
-mpLayoutNames->Insert(new String(aLayoutName), LIST_APPEND);
+maLayoutNames.push_back(new String(aLayoutName));
 
 Bitmap aBitmap(mpDocSh->GetPagePreviewBitmap(pMaster, 90));
-maVS.InsertItem((sal_uInt16)mpLayoutNames->Count(), aBitmap, 
aLayoutName);
+maVS.InsertItem((sal_uInt16)maLayoutNames.size(), aBitmap, 
aLayoutName);
 }
 }
 
@@ -274,21 +265,22 @@ IMPL_LINK(SdPresLayoutDlg, ClickLoadHdl, void *, EMPTYARG)
 {
 // Pruefen, ob Vorlage schon vorhanden
 sal_Bool bExists = sal_False;
-String* pName = (String*)mpLayoutNames->First();
 String aCompareStr( maName );
 if( maName.Len() == 0 )
 aCompareStr = maStrNone;
 
-while( pName && !bExists )
+int aPos = 0;
+for (boost::ptr_vector::iterator it = 
maLayoutNames.begin();
+ it != maLayoutNames.end() && !bExists; ++it, ++aPos)
 {
-if( aCompareStr == *pName )
+String pName = *it;
+if( aCompareStr == pName )
 {
 bExists = sal_True;
 // Vorlage selektieren
-sal_uInt16 nId = (sal_uInt16) mpLayoutNames->GetCurPos() + 1;
+sal_uInt16 nId = (sal_uInt16) aPos + 1;
 maVS.SelectItem( nId );
 }
-pName = (String*)mpLayoutNames->Next();
 }
 
 if( !bExists )
@@ -313,10 +305,10 @@ IMPL_LINK(SdPresLayoutDlg, ClickLoadHdl, void *, EMPTYARG)
 {
 String aLayoutName(pMaster->GetLayoutName());
 aLayoutName.Erase( aLayoutName.SearchAscii( 
SD_LT_SEPARATOR ) );
-mpLayoutNames->Insert(new String(aLayoutName), 
LIST_APPEND);
+maLayoutNames.push_back(new String(aLayoutName));
 
 Bitmap 
aBitmap(pTemplDocSh->GetPagePreviewBitmap(pMaster, 90));
-
maVS.InsertItem((sal_uInt16)mpLayoutNames->Count(), aBitmap, aLayoutName);
+maVS.InsertItem((sal_uInt16)maLayoutNames.size(), 
aBitmap, aLayoutName);
 }
 }
 }
@@ -330,15 +322,15 @@ IMPL_LINK(SdPresLayoutDlg, ClickLoadHdl, void *, EMPTYARG)
 else
 {
 // leeres Layout
-mpLayoutNames->Insert( new String( maStrNone ), LIST_APPEND );
-maVS.InsertItem( (sal_uInt16) mpLayoutNames->Count(),
+maLayoutNames.push_back( new String( maStrNone ) );
+maVS.InsertItem( (sal_uInt16) maLayoutNames.size(),
 Bitmap( SdResId( BMP_FOIL_NONE ) ), maStrNone 

Re: [Libreoffice] [PATCH] convert SdPresLayoutDlg from tools/list.hxx to std::map

[Noel Grandin]

sorry, that should be "converted to std::vector"

On 2012-01-31 16:22, Noel Grandin wrote:

Hi

Attached patch converts SdPresLayoutDlg to use std::map.

Code is contributed under MPL+/LGPL+/GPL+

Regards, Noel Grandin

Disclaimer: http://www.peralex.com/disclaimer.html




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


Disclaimer: http://www.peralex.com/disclaimer.html


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

Re: [Libreoffice] [REVIEW]Fix for annoying bug fdo#35669

[Petr Mladek]

Noel Power píše v Út 31. 01. 2012 v 11:24 +:
> On 30/01/12 17:16, Cedric Bosdonnat wrote:
> > Hi Michael,
> >
> > On Thu, 2012-01-26 at 19:35 +0100, Michael Stahl wrote:
> >> On 26/01/12 19:30, Michael Stahl wrote:
> >>> On 26/01/12 16:21, Cedric Bosdonnat wrote:
>  Hi all,
> 
>  Could someone review and cherry-pick this commit to -3-5, -3-5-0 and
>  -3-4?
>  http://cgit.freedesktop.org/libreoffice/core/commit/?id=5e51960dede5015b862df05b7b16f02884647889
> >>> your fix looks good to me, except for the line:
> >>>
> >>>sal_uInt16 nMaxDstId = *aIds.end();
> >> oops, now it occurs to me that dereferencing end() is a bad idea in any
> >> case, you probably want to use rbegin() instead :)
> >>
> >>> if i'm reading this right, then aIds could be empty here, if the source
> >>> document has seq field references, and the target document has no seq
> >>> field references but note references.
> >>>
> >>> suggest to wrap an "if (!aIds.empty())" around the whole block?
> >>> no need to adjust anything in that case, right?
> > Fixed in
> > http://cgit.freedesktop.org/libreoffice/core/commit/?id=458d56d9b35f2ecb2eea2589737079390b10ee26
> >
> > Would be nice to cherry-pick these two commits to -3-5, -3-4 and 3-5-0
> > (if it's not too late).
> >
> the change looks good to me now too

Looked fine. Worked reasonable => pushed into 3-5-0, see
http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4632a18dae5e1770aa1da4ac55719a09c1ae461b
http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4925acf5648438589bca74f3dbfc2fc5fc9f61ab


Best Regards,
Petr

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

Re: [Libreoffice] [PATCH] contour handling for anchored draw objects in writer (fix for bugs 34611 and 45376)

[Noel Power]

On 31/01/12 12:46, openoff...@schierla.de wrote:

Noel Power wrote:

I cc some writer experts who may suggest some
alternative solution(s) or change ( or maybe say it is perfect )

The new patch in this e-mail should provide even better performance
(as it only clears the contour cache for contour wrapped drawing objects).


ah that is nicer, I committed the additional check too

Can you confirm your contribution is under the MPL 1.1 / GPLv3+ /
LGPLv3+ license(s)

Yes, the contribution is licensed under  MPL 1.1 / GPLv3+ / LGPLv3+


thanks again!!

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

Re: [Libreoffice] [PUSHED:3-5-0]]Fix for annoying bug fdo#35669

[Petr Mladek]

Updated subject.

Petr Mladek píše v Út 31. 01. 2012 v 15:26 +0100:
> Looked fine. Worked reasonable => pushed into 3-5-0, see
> http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4632a18dae5e1770aa1da4ac55719a09c1ae461b
> http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4925acf5648438589bca74f3dbfc2fc5fc9f61ab
> 
> 
> Best Regards,
> Petr


___
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 35669, which changed state.

Bug 35669 Summary: Incorrect cross-references to illustrations and equations in 
master documents
https://bugs.freedesktop.org/show_bug.cgi?id=35669

   What|Old Value   |New Value

 Resolution|FIXED   |
 Status|RESOLVED|REOPENED

-- 
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] R: R: Fatal error building Windows release (Cygwin)

[walter]

The suggestion doesn't solve the problem.

Regards

Walter

 

>>> 

cd xmlreader
rm -Rf wntmsci12.pro
build







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

[Libreoffice] [REVIEW][3.5][3.5.0?] fix for fdo#44836 (leaking page breaks)

[Cedric Bosdonnat]

Hi all,

I just fixed some remaining leaking page breaks. Could someone review
and cherry-pick it for 3.5 (and hopefully it's not too late for 3.5.0)?

http://cgit.freedesktop.org/libreoffice/core/commit/?id=9babbf346c4d6f76e8b3a243a00e73a173c16ba6

Thanks,

--
Cedric

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

Re: [Libreoffice] [REVIEW]Fix for annoying bug fdo#35669

[Jan Holesovsky]

Hi Petr, Cedric, all,

On 2012-01-31 at 15:26 +0100, Petr Mladek wrote:

> Looked fine. Worked reasonable => pushed into 3-5-0, see
> http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4632a18dae5e1770aa1da4ac55719a09c1ae461b
> http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5-0&id=4925acf5648438589bca74f3dbfc2fc5fc9f61ab

The patch improved the situation for sure, so great that you've pushed
that, thank you!  Unfortunately it hasn't fixed all the corner cases, so
I had to reopen the issue + described that; but still, worth having in
libreoffice-3-5-0 :-)

Thank you all,
Kendy

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

Re: [Libreoffice] [REVIEW][3-5] Prevent excessvie references to formula result tokens

[Kohei Yoshida]

Hi Eike,

On Tue, 2012-01-31 at 14:27 +0100, Eike Rathke wrote:
> Hi Kohei,
> 
> On Monday, 2012-01-30 16:35:46 -0500, Kohei Yoshida wrote:
> 
> > I'd like
> > 
> > http://cgit.freedesktop.org/libreoffice/core/commit/?id=e2b11f4fd79dce4116badb0ecf6477546ca5d0d4
> > 
> > cherry-picked to the 3-5 branch.  It is probably too late for change
> > like this to be in the 3-5-0 branch so I won't even try.
> 
> I'd like to ask for an improvement ;)  see below.

Well, for expediting the process (since the release of 3.5.0 is
nearing), could you take this over?  Sounds like you know exactly what
needs to get done, so it's probably better than me trying to figure this
out on my own.

Thanks a lot,

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

Re: [Libreoffice] Simplified C++ configuration access

[Stephan Bergmann]

On 01/31/2012 11:31 AM, Stephan Bergmann wrote:

At the next level, the new unotools/configuration.hxx header provides
type-safe C++ wrappers for the various configuration entities on top of
the lower level UNO services. For example, the
utl::ConfigurationProperty class template can be used to wrap a given
configuration property of a specific type in a C++ class whose get and
set member functions internally handle the conversion between the UNO
API's generic Any and the concrete type of the given property. There is
also a utl::ConfigurationChanges class that wraps an instance of
ReadWriteAccess for (batch) modification.


Had to move the header from unotools/configuration.hxx to 
comphelper/configuration.hxx now, so that some other code in comphelper 
can use it, too.  Only noticeable change to client code should be that 
ConfigurationChanges is now in namespace comphelper.


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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Ivan Timofeev]

31.01.2012 18:22, Noel Power пишет:

On 30/01/12 13:54, Noel Grandin wrote:

Attached patch converts SdCustomShow to use std::vector.

I hate these remove list etc. type patches as they tend to be tricky and
time consuming to look at ;-) . The patch looked ok for me, hopefully if
I missed something ( 'cause these really are tricky ) then someone else
will spot it.


std::remove doesn't remove elements from the vector. It just moves the 
elements to the front. So we have to 'vector::erase' garbage when 
replacing RemovePage by std::remove. Am I right?


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

Re: [Libreoffice] Simplified C++ configuration access

[Michael Stahl]

On 31/01/12 15:06, Lubos Lunak wrote:
> On Tuesday 31 of January 2012, Stephan Bergmann wrote:
>> To get a property's value, all that is needed is to call the wrapper
>> class's static officecfg::Foo::Bar::Baz::Whatever::get member function,
>> which requires the XComponentContext as argument (use
>> comphelper::getProcessComponentContext in code that does not thread the
>> component context more locally).
> 
>  How usual is the case that one does not want to pass 
> comphelper::getProcessComponentContext() to all these functions? Or in other 
> words, why not simply make it the default argument value?

default argument would make officecfg depend on comphelper, so you
couldn't use the new headers from comphelper implementation.


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

Re: [Libreoffice] [PATCH] CSC, SEC, CSCH & SECH Help missing, How to add one? ref: bug fdo#44792

[Michael Meeks]

On Tue, 2012-01-31 at 00:00 +0100, Regina Henschel wrote:
> The term "inverse" is not correct in this context. It should be 
> "reciprocal" instead.
> Example: The inverse of SIN is ASIN. The reciprocal of SIN is CSC.

I agree Regina, you have unusually good English :-)

Any chance of a patch to fix the docs ?

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] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Grandin]

So I should be calling
   maPages.erase( std::find(maPages.begin(), maPages.end(), pOldPage) );
??

On 2012-01-31 17:24, Ivan Timofeev wrote:

31.01.2012 18:22, Noel Power пишет:

On 30/01/12 13:54, Noel Grandin wrote:

Attached patch converts SdCustomShow to use std::vector.

I hate these remove list etc. type patches as they tend to be tricky and
time consuming to look at ;-) . The patch looked ok for me, hopefully if
I missed something ( 'cause these really are tricky ) then someone else
will spot it.


std::remove doesn't remove elements from the vector. It just moves the 
elements to the front. So we have to 'vector::erase' garbage when 
replacing RemovePage by std::remove. Am I right?


Ivan



Disclaimer: http://www.peralex.com/disclaimer.html


___
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 34432, which changed state.

Bug 34432 Summary: firebird ODBC: LibO crashes on 64 bits
https://bugs.freedesktop.org/show_bug.cgi?id=34432

   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] [Libreoffice-qa] [ANN] LibreOffice 3.5.0 RC2 available

[LL]

Rainer Bielefeld  bielefeldundbuss.de> writes:

> 
> Hi,
> 
> Concerning the Virus Kazy.7720 Problem: Can you all please contact 
> support of your Antivirus Program supplier so that they can check and 
> rework their filters? So we can save much trouble with false alarms 
> normal users would see when they install 3.5.0 before anti virus filters 
> are updated.
> 
> Best regards
> 
> Rainer



New definitions files from Avira today and Kazy.7720 is no longer a problem.

/LL



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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Ivan Timofeev]

31.01.2012 19:42, Noel Grandin пишет:

So I should be calling
maPages.erase( std::find(maPages.begin(), maPages.end(), pOldPage) );
??


maPages.erase( std::remove(maPages.begin(), maPages.end(), pOldPage), 
maPages.end() );


I'll fix that.

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

Re: [Libreoffice] [REVIEW][3.5][3.5.0?] fix for fdo#44836 (leaking page breaks)

[Michael Meeks]

On Tue, 2012-01-31 at 15:48 +0100, Cedric Bosdonnat wrote:
> I just fixed some remaining leaking page breaks. Could someone review
> and cherry-pick it for 3.5 (and hopefully it's not too late for 3.5.0)?
> 
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=9babbf346c4d6f76e8b3a243a00e73a173c16ba6

Well; it looks conceptually reasonable to me. I hope that VCL will not
do slowish X stuff for the hide if it is already hidden, and vv. for the
Show. Did you test with a few hundred pages ?

If so, I'm happy I guess ;-) thanks for fixing my pet hate.

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] Removed unused ImplSalLogFontToFontA()

[Michael Meeks]

On Tue, 2012-01-31 at 13:21 +, Caolán McNamara wrote:
> On Tue, 2012-01-31 at 11:15 +, Noel Power wrote:
> > cc'ing Caolán as he might be interested because it doesn't turn up in
> > the unused.easy
> 
> Its windows only code, and callcatcher can only see what gets compiled
> as its just a compiler wrapper and I run it under Linux.

Interesting (in a world with infinite time); would the mingw
cross-compile be call-catcher-able such that we could get a more precise
and comprehensive list ?

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] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Grandin]

thanks, still working out how this STL stuff hangs together.

On 2012-01-31 17:55, Ivan Timofeev wrote:

31.01.2012 19:42, Noel Grandin пишет:

So I should be calling
maPages.erase( std::find(maPages.begin(), maPages.end(), pOldPage) );
??


maPages.erase( std::remove(maPages.begin(), maPages.end(), pOldPage), 
maPages.end() );


I'll fix that.

Ivan



Disclaimer: http://www.peralex.com/disclaimer.html


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

Re: [Libreoffice] [PATCH] Removed unused ImplSalLogFontToFontA()

[Caolán McNamara]

On Tue, 2012-01-31 at 16:10 +, Michael Meeks wrote:
> On Tue, 2012-01-31 at 13:21 +, Caolán McNamara wrote:
> > On Tue, 2012-01-31 at 11:15 +, Noel Power wrote:
> > > cc'ing Caolán as he might be interested because it doesn't turn up in
> > > the unused.easy
> > 
> > Its windows only code, and callcatcher can only see what gets compiled
> > as its just a compiler wrapper and I run it under Linux.
> 
>   Interesting (in a world with infinite time); would the mingw
> cross-compile be call-catcher-able such that we could get a more precise
> and comprehensive list ?

Yes, in theory anyway, IIRC fridrich added mingw support to callcatcher
ages ago.

C.

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

Re: [Libreoffice] [PATCH] Removed unused ImplSalLogFontToFontA()

[Fridrich Strba]

Hello,

On 31/01/12 17:10, Michael Meeks wrote:
>   Interesting (in a world with infinite time); would the mingw
> cross-compile be call-catcher-able such that we could get a more precise
> and comprehensive list ?

Yes, at the level of callcatcher 1.1.3, I added the support for gnu
assembler for windows, so it could work.

Cheers

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Stephan Bergmann]

On 01/31/2012 03:22 PM, Noel Power wrote:

I hate these remove list etc. type patches as they tend to be tricky and
time consuming to look at ;-) . The patch looked ok for me, hopefully if
I missed something ( 'cause these really are tricky ) then someone else
will spot it. Anyway thanks again for the nice patch. So... pushed to
master


Looks like you missed to run make check:


LOG> 1. removeByIndex("1") ...OK
LOG> removing a non existent object to get an exception
LOG> 2. removeByIndex(): Expected exception - OK
Method removeByIndex() finished with state FAILED
LOG> removeByIndex(): PASSED.FAILED
* State for 
sd.SdXCustomPresentation::com::sun::star::container::XIndexContainer **
[sd.SdXCustomPresentation::com::sun::star::container::XIndexContainer::removeByIndex()]
 is testcode: [removeByIndex()] - PASSED.FAILED


Guess Ivan's fix will rectify that.

Stephan
___
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 42543, which changed state.

Bug 42543 Summary: Report Designer Wizard missing Label display and editing 
capability
https://bugs.freedesktop.org/show_bug.cgi?id=42543

   What|Old Value   |New Value

 Resolution||FIXED
 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] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Grandin]

I ran make check at the global level - does that not invoke the 
individual modules check targets?


On 2012-01-31 18:17, Stephan Bergmann wrote:

Looks like you missed to run make check:


LOG> 1. removeByIndex("1") ...OK
LOG> removing a non existent object to get an exception
LOG> 2. removeByIndex(): Expected exception - OK
Method removeByIndex() finished with state FAILED
LOG> removeByIndex(): PASSED.FAILED
* State for 
sd.SdXCustomPresentation::com::sun::star::container::XIndexContainer 
**
[sd.SdXCustomPresentation::com::sun::star::container::XIndexContainer::removeByIndex()] 
is testcode: [removeByIndex()] - PASSED.FAILED


Guess Ivan's fix will rectify that.

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



Disclaimer: http://www.peralex.com/disclaimer.html


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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Ivan Timofeev]

31.01.2012 19:55, Ivan Timofeev пишет:

I'll fix that.


Hum; guys, what about
- mpSdCustomShow->Remove(pPage->GetSdrPage());
+ ::std::remove(mpSdCustomShow->PagesVector().begin(), 
mpSdCustomShow->PagesVector().end(),

+ pPage->GetSdrPage());

mpSdCustomShow->Remove(pPage->GetSdrPage()); <- using the pointer as the 
index??? Or it removes the first occurence of the page?


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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Stephan Bergmann]

On 01/31/2012 05:19 PM, Noel Grandin wrote:

I ran make check at the global level - does that not invoke the
individual modules check targets?


It does.  Strange anyway.  This failure is consistently reproducible for 
me and looks suspiciously like the remove problem discussed in this 
thread...


But thanks for running make check!
Stephan
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] [REVIEW 3-5, 3-5-0] fdo#42543: Empty wizard page in reportdesign

[Jan Holesovsky]

Hi,

https://bugs.freedesktop.org/show_bug.cgi?id=42543

When you use the report designer, at some stage you get to a page (step
2) which is blank; but when you click there, you can edit the data.

The problem is that commit

http://cgit.freedesktop.org/libreoffice/core/commit/?id=efc6b840a0423099d6c3e0d33c14d34637fbba4c

changed the order of painting of the overlapped controls; in this case,
a background image is painted over the Edit fields; but the clicks go
through, so in the end they are shown when you click there.

As a fix, I propose this:

http://cgit.freedesktop.org/libreoffice/core/commit/?id=03e17a141fbb4e1242de9d9979b5b699e6840454

Just get rid of the background image; in my opinion, it even looks
better without that, and is the least intrusive change.

Can you please review for -3-5 and -3-5-0?

Thank you,
Kendy

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

Re: [Libreoffice] Simplified C++ configuration access

[Stephan Bergmann]

On 01/31/2012 03:06 PM, Lubos Lunak wrote:

  How usual is the case that one does not want to pass
comphelper::getProcessComponentContext() to all these functions? Or in other
words, why not simply make it the default argument value?


Default value indeed looks like a smart move.  (In principle, I object 
against comphelper::getProcessComponentContext and consider it a hack. 
In practice, we are probably stuck with it, anyway.)


Only drawback is that in multi-parameter functions (like the set() 
functions) it is currently the first parameter -- I generally like to 
sort parameters from broadest context to most specific.  But anyway, 
changed those parameter lists around and pushed now.


Thanks for the input,
Stephan
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [REVIEW 3-5, 3-5-0] fdo#42543: Empty wizard page in reportdesign

[Lubos Lunak]

On Tuesday 31 of January 2012, Jan Holesovsky wrote:
> As a fix, I propose this:
>
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=03e17a141fbb4e1242d
>e9d9979b5b699e6840454
>
> Just get rid of the background image; in my opinion, it even looks
> better without that, and is the least intrusive change.
>
> Can you please review for -3-5 and -3-5-0?

 Pushed to 3-5.

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

Re: [Libreoffice] Simplified C++ configuration access

[Stephan Bergmann]

On 01/31/2012 04:25 PM, Michael Stahl wrote:

default argument would make officecfg depend on comphelper, so you
couldn't use the new headers from comphelper implementation.


I just moved configuration.hxx from unotools to comphelper today (to 
then be able to use the officecfg headers in comphelper), without 
realizing that this -- of course -- will lead to a ring reference, anyway.


Anyway, the only time officecfg requires comphelper/configuration.hxx 
(and now also comphelper/processfactory.hxx) is during 
CppunitTest_officecfg_cppheader_test.  The workaround, I'd say, is to 
temporarily move that test to subsequentcheck.  Once everything is 
dissolved in tail_build this should no longer be a problem, right?


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

Re: [Libreoffice] [PATCH] CSC, SEC, CSCH & SECH Help missing, How to add one? ref: bug fdo#44792

[Olivier Hallot]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

Hi

I fixed the english terminology, according to the justifications by Regina.

regards

Olivier

Em 31-01-2012 11:28, Regina Henschel escreveu:
> Olivier Hallot schrieb:
>> -BEGIN PGP SIGNED MESSAGE-
>> Hash: SHA1
>>
>> Hi Regina
>>
>> It has been decades since I handled groups, rings, fields, and functions
>> in my french lycée. Now I am confused because it looks like the two
>> terms are swapped between french and english...
>>
>> Regards
>>
>> Olivier
>>
>> In french:
>> http://fr.wikipedia.org/wiki/Cos%C3%A9cante#Fonctions_r.C3.A9ciproques
>>
>> in English
>> http://en.wikipedia.org/wiki/Inverse_%28mathematics%29
> 
> A better reference for Engish is
> http://en.wikipedia.org/wiki/Sine#Reciprocal
> 
> The topic here are functions, not numbers.
> For numbers you have the term "additive inverse of x is -x" or
> "multiplicative inverse of x is 1/x". But in context of functions
> "inverse" always means the inverse in respect to composition.
> 
> I don't know French. Astonishing, http://en.wikipedia.org/wiki/Sine has
> no direct translation to French.
> 
> Kind regards
> Regina
> ___
> LibreOffice mailing list
> LibreOffice@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/libreoffice

- -- 
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/

iQEcBAEBAgAGBQJPKBklAAoJEJp3R7nH3vLxC7sIAIgLiitUH2qEH8ghN3HyatbZ
FHLJRXgDVmF002fOeG1CUyjETCdr2i4ZC0zD7hr+/6PK2u16tR1ZRIY0Mq0VNe2f
KWso+ZWFJRvas7hnwAqjJJjBNpWZ2wvHfg+9I8kMxo6FKoVBgB4vjjonD+NTQ5hr
g5trt6Y0jov5aCFvcAXPE4JphNkVh/W1uWq0/apbkwjKppoaz/0L+R6XDZ2YYvbS
xc/hlFEkhlaSzCMY6ULMSHO7y93pWu1Fk2INgcVm44wpLAenBRzJsCeH+bPvRbSZ
y+r+QKPrYRhII/xxqKfWz4vgm5RThK2iQ0UMqloyr4u7ZAKXRMuY1DRe1Q5d8es=
=mSq+
-END PGP SIGNATURE-
>From 182098d2b8a625e8da97193ec627a1cd5a580e6c Mon Sep 17 00:00:00 2001
From: Olivier Hallot 
Date: Tue, 31 Jan 2012 14:35:04 -0200
Subject: [PATCH] fix for fdo#44972, LOCALHELP for CSC,CSCH,SEC,SECH

fix function terminology.
---
 helpcontent2/source/text/scalc/01/04060106.xhp |   15 +++
 1 files changed, 7 insertions(+), 8 deletions(-)

diff --git a/helpcontent2/source/text/scalc/01/04060106.xhp b/helpcontent2/source/text/scalc/01/04060106.xhp
index a272e1d..e80b11f 100644
--- a/helpcontent2/source/text/scalc/01/04060106.xhp
+++ b/helpcontent2/source/text/scalc/01/04060106.xhp
@@ -341,7 +341,7 @@ ACOSH, ASINH, ATANH, ACOTH,   
 
 CSC
-Returns the cosecant of the given angle (in radians). The cosecant is the inverse of the sine (= 1/sin(x))
+Returns the cosecant of the given angle (in radians). The cosecant of an angle is equivalent to 1 divided by the sine of that angle
 Syntax
 CSC(Number)
@@ -365,12 +365,11 @@ oldref="159">CSCH
 Syntax
 CSCH(Number)
-Returns the hyperbolic cosecant of Number. The hyperbolic cosecant is the inverse of the hyperbolic sine (= 1/sinh(x)) 
+Returns the hyperbolic cosecant of Number.
 Example
 
-=CSCH(1) returns approximately 0.8509181282, the hyperbolic cosecant of 1.see also SINH, TANH, COTH, SECH, 
-ACOSH, ASINH, ATANH, ACOTH,   
+=CSCH(1) returns approximately 0.8509181282, the hyperbolic cosecant of 1.see also SINH, TANH, COTH, SECH,ACOSH, ASINH, ATANH, ACOTH,   
 
 
 DEGREES function
@@ -977,7 +976,7 @@ CEILING, FLOOR, EVEN, ODD, MROUND
 
 SEC
-Returns the secant of the given angle (in radians). The secant is the inverse of the cosine (= 1/cos(x))
+Returns the secant of the given angle (in radians). The secant of an angle is equivalent to 1 divided by the cosine of that angle
 Syntax
 SEC(Number)
@@ -989,7 +988,7 @@ oldref="154">Examples
 =SEC(PI()/4) returns approximately 1.4142135624, the inverse of the cosine of PI/4 radians.
 
 =SEC(RADIANS(60)) returns 2, the secant of 60 degrees.see also SIN, TAN, COT, CSC,
-ACOS, ASIN, ATAN, ATAN2, ACOT  
+ACOS, ASIN, ATAN, ATAN2, ACOT
 
 
 SECH function
@@ -1001,12 +1000,12 @@ oldref="159">SECH
 Syntax
 SECH(Number)
-Returns the hyperbolic secant of Number. The hyperbolic secant is the inverse of the hyperbolic cosine (= 1/cosh(x)) 
+Returns the hyperbolic secant of Number.
 Example
 
 =SECH(0) returns 1, the hyperbolic secant of 0.see also SINH, TANH, COTH, CSCH, 
-ACOSH, ASINH, ATANH, ACOTH,   
+ACOSH, ASINH, ATANH, ACOTH
 
 
 SIN function
-- 
1.7.5.4

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

[Libreoffice] hoping that someone will be able to solve this problem - unable to build windows version using cygwin

[walter]

I followed all the instructions in
wiki.documentfoundation.org/Development/Native_Build and in
wiki.documentfoundation.org/Development/Windows_Build_Dependencies

but I’m unable to build the release.

 

Full logs (including getting the source) at pastebin.com/u/walterLO

 

Release libreoffice 3-5

 

$ git show

commit 5de902f2b66e64bc4b4755356db3b259c01ddcdd

Author: Cédric Bosdonnat < 
cedric.bosdonnat@free.fr>

Date:   Mon Jan 30 11:35:42 2012 +0100

 

Regards

walter

 

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

[Libreoffice] [PUSHED] Re: [REVIEW] fdo#43965 fix RTF import of page break symbol

[Michael Stahl]

On 30/01/12 23:41, Miklos Vajna wrote:
> Hi Michael,
> 
> On Mon, Jan 30, 2012 at 10:07:50PM +0100, Michael Stahl  
> wrote:
>>> See
>>> http://cgit.freedesktop.org/libreoffice/core/commit/?id=2aa3d43a3746c797391afb1a08c024d512976006
>>
>> seems that with this change not only the new LN_PFPageBreakBefore from
>> \page is reset when a \par occurs, but also the LN_PFPageBreakBefore
>> inserted from \pagebb.
>> i have no idea if that is good or bad, but your comment says "since it's
>> a symbol", which is only true for \page, not \pagebb; is it intentional?
> 
> Thanks for the catch, that's indeed not intentional and it is
> definitely bad. Here is a followup commit:
> 
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=d1f3018b0e6d0889febcf09a6c95a2ec354137c0
> 
> I'm attaching a patch combining the two, to be pushed to -3-5.

pushed to libreoffice-3-5

http://cgit.freedesktop.org/libreoffice/core/commit/?h=libreoffice-3-5&id=d219815d282c66a4aacdb0ed9640bbd14b4f9bde

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

Re: [Libreoffice] [REVIEW 3-5, 3-5-0] fdo#42543: Empty wizard page in reportdesign

[Noel Power]

On 31/01/12 16:31, Lubos Lunak wrote:

On Tuesday 31 of January 2012, Jan Holesovsky wrote:

As a fix, I propose this:

http://cgit.freedesktop.org/libreoffice/core/commit/?id=03e17a141fbb4e1242d
e9d9979b5b699e6840454

Just get rid of the background image; in my opinion, it even looks
better without that, and is the least intrusive change.

Can you please review for -3-5 and -3-5-0?

  Pushed to 3-5.


+1, one more needed I think for 3-5-0

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

Re: [Libreoffice] [REVIEW 3-5, 3-5-0] fdo#42543: Empty wizard page in reportdesign

[Michael Stahl]

On 31/01/12 17:31, Lubos Lunak wrote:
> On Tuesday 31 of January 2012, Jan Holesovsky wrote:
>> As a fix, I propose this:
>>
>> http://cgit.freedesktop.org/libreoffice/core/commit/?id=03e17a141fbb4e1242d
>> e9d9979b5b699e6840454
>>
>> Just get rid of the background image; in my opinion, it even looks
>> better without that, and is the least intrusive change.
>>
>> Can you please review for -3-5 and -3-5-0?
> 
>  Pushed to 3-5.

that background thing looks unnecessary, +1 from me

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Power]

On 31/01/12 15:55, Ivan Timofeev wrote:

31.01.2012 19:42, Noel Grandin пишет:

So I should be calling
maPages.erase( std::find(maPages.begin(), maPages.end(), pOldPage) );
??


maPages.erase( std::remove(maPages.begin(), maPages.end(), pOldPage), 
maPages.end() );


I'll fix that.

Ivan

nice catch, I admit not being familiar with this but I did read 
"http://en.cppreference.com/w/cpp/algorithm/remove"; and even read the 
second paragraph where it mentions the items are removed but fails ( imo 
) to clearly indicate the removed items are not removed. Sure if you 
read it *very carefully* you get it :-/ ( appears I am not careful enough )

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Power]

On 31/01/12 16:17, Stephan Bergmann wrote:

On 01/31/2012 03:22 PM, Noel Power wrote:

I hate these remove list etc. type patches as they tend to be tricky and
time consuming to look at ;-) . The patch looked ok for me, hopefully if
I missed something ( 'cause these really are tricky ) then someone else
will spot it. Anyway thanks again for the nice patch. So... pushed to
master


Looks like you missed to run make check:

yep, correct. didn't even occur to me :-(
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Power]

On 31/01/12 16:20, Ivan Timofeev wrote:

31.01.2012 19:55, Ivan Timofeev пишет:

I'll fix that.


Hum; guys, what about
- mpSdCustomShow->Remove(pPage->GetSdrPage());
+ ::std::remove(mpSdCustomShow->PagesVector().begin(), 
mpSdCustomShow->PagesVector().end(),

+ pPage->GetSdrPage());

mpSdCustomShow->Remove(pPage->GetSdrPage()); <- using the pointer as 
the index??? Or it removes the first occurence of the page?
god bless your eyes, I read that even now '->Remove(pPage" when scanning 
it as RemovePage and was wondering what you were on about. should remove 
the first occurrence, care to fix  ?


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

Re: [Libreoffice] [PUSHED 3-5] Same expression on both sides of '||' in svtools/source/filter/sgvtext.cxx

[Eike Rathke]

Hi julien2412,

On Monday, 2012-01-30 12:24:07 -0800, julien2412 wrote:

> Commited and pushed on master (commit
> 10bfc4df5335c6cd2ae700df7485668d896998b5)
> 
> PS : if it could be useful to push this on 3.5, just tell me :-)

Just did so with my sign-off for 3-5, before someone with an incapable
editor comes along and mangles it..
109a84546ab33d17e29a6e6a04ea7e460219a357

Thanks
  Eike

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


pgpeipw8j77BM.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PUSHED][3.5][3.5.0] fix for fdo#44836 (leaking page breaks)

[Petr Mladek]

Michael Meeks píše v Út 31. 01. 2012 v 15:56 +:
> On Tue, 2012-01-31 at 15:48 +0100, Cedric Bosdonnat wrote:
> > I just fixed some remaining leaking page breaks. Could someone review
> > and cherry-pick it for 3.5 (and hopefully it's not too late for 3.5.0)?
> > 
> > http://cgit.freedesktop.org/libreoffice/core/commit/?id=9babbf346c4d6f76e8b3a243a00e73a173c16ba6
> 
>   Well; it looks conceptually reasonable to me. I hope that VCL will not
> do slowish X stuff for the hide if it is already hidden, and vv. for the
> Show. Did you test with a few hundred pages ?
> 
>   If so, I'm happy I guess ;-) thanks for fixing my pet hate.

I played with a big document. The patch definitely improved the
behavior. I did not see any evident problem. The patch looked
reasonable. Michael Stahl confirmed that it should not cause any evident
crash => I pushed it into both 3-5 and 3-5-0 branches.


Best Regards,
Petr

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Ivan Timofeev]

31.01.2012 21:12, Noel Power пишет:

should remove the first occurrence, care to fix ?


so the exact equivalent of

mpSdCustomShow->Remove(pPage->GetSdrPage())

will be:

SdCustomShow::PageVec::iterator it = std::find(
mpSdCustomShow->PagesVector().begin(),
mpSdCustomShow->PagesVector().end(),
pPage->GetSdrPage());
if (it != mpSdCustomShow->PagesVector().end())
mpSdCustomShow->PagesVector().erase(it);

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

[Libreoffice] default build verbosity

[Michael Stahl]

hi all,

a recurring problem with remote debugging newbie build problems is that
the default build is silent.

i wonder if it would make sense to change that:
with default configure invocation, build system outputs the full command
to the terminal/log, and there is an explicit configure option to
silence the build.

so when newbies pastebin their broken build, it's much easier for
developers to spot wrong command lines without wasting time writing a
lot of pointless mails and such.

actually for tinderbox builds there could also be advantages (though it
would also make the logs a lot bigger).

on the other hand, experienced developers can just put the configure
option in their autogen.lastrun as a one-time cost and won't be bothered
much.

regards,
 michael

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Noel Power]

On 31/01/12 17:31, Ivan Timofeev wrote:

31.01.2012 21:12, Noel Power пишет:

should remove the first occurrence, care to fix ?


so the exact equivalent of

mpSdCustomShow->Remove(pPage->GetSdrPage())

will be:

SdCustomShow::PageVec::iterator it = std::find(
mpSdCustomShow->PagesVector().begin(),
mpSdCustomShow->PagesVector().end(),
pPage->GetSdrPage());
if (it != mpSdCustomShow->PagesVector().end())
mpSdCustomShow->PagesVector().erase(it);
hey I've already proved my (non) STL credentials :-) , but.. if we are 
to believe the old code then 
mpSdCustomShow->Remove(pPage->GetSdrPage()); would call 
Container::Remove( void* p ) which would call Container::Remove( 
Container::GetPos( p ) ) which afaics deletes the first match :-) and 
that's what your patch seems to do, don't know if there is a briefer way 
to specify that but to me what you have looks fine



Noel


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

Re: [Libreoffice] default build verbosity

[Norbert Thiebaud]

On Tue, Jan 31, 2012 at 11:43 AM, Michael Stahl  wrote:
> hi all,
>
> a recurring problem with remote debugging newbie build problems is that
> the default build is silent.
>
> i wonder if it would make sense to change that:
> with default configure invocation, build system outputs the full command
> to the terminal/log, and there is an explicit configure option to
> silence the build.
>
> so when newbies pastebin their broken build, it's much easier for
> developers to spot wrong command lines without wasting time writing a
> lot of pointless mails and such.
>
> actually for tinderbox builds there could also be advantages (though it
> would also make the logs a lot bigger).

One thing that would mitigate the size of the log in that case would
be to hide the 'untar' output, even in verbose mode
(put it is a separate file in misc/logs for instance)
un-tarring of boost, for instance, account for thousands of line of
pointless output in verbose mode...

even better, for gbuild, would be to store the output of each command
in a file (again logs/xxx) and only display it inn case of problem.
this has the added benefit of keeping the whole output together (*)

Norbert


(*) and the added 'benefit' of allowing fairly stable eye-candy like
coloring the output
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PUSHED][PATCH V2] convert SdCustomShow from tools/list to vector

[Ivan Timofeev]

31.01.2012 21:44, Noel Power пишет:

hey I've already proved my (non) STL credentials :-) , but.. if we are
to believe the old code then
mpSdCustomShow->Remove(pPage->GetSdrPage()); would call
Container::Remove( void* p ) which would call Container::Remove(
Container::GetPos( p ) ) which afaics deletes the first match :-) and
that's what your patch seems to do, don't know if there is a briefer way
to specify that but to me what you have looks fine


Ok. I'm not a STL expert, too.
http://cgit.freedesktop.org/libreoffice/core/commit/?id=f2e17f8f6b6724d7e56d6583d79434372f6221bd

Marking as pushed, thanks a lot Noel! :)

Someone with more shiny eyes & more strong STL/Container knowledge may 
find another gotchas...


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

Re: [Libreoffice] [Libreoffice-ux-advise] [PATCH] [PUSHED] fdo#45285 - Use more revealing error message for not allowed sheet names

[Albert Thuswaldner]

Hi,

>        Well - hopefully six months away is not -so- far - but I appreciate
> your pain. Of course - in theory with triple review features can get
> into the -3-5 branch, but string changes are less welcome I suspect.
>

Well, for me it is not that important. I just wanted to make sure
before I told the guy who reported the bug that it wouldn't make it
into 3.5.x.

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

Re: [Libreoffice] [PATCH] Making default tab prefix name configurable in Calc

[Albert Thuswaldner]

Hi,
> However something I noticed is that there are still some problems with
> entering the 'Prefix name" in the options, firstly the input box seems to be
> overwritten by the label ( part of the input/edit field for the prefix is
> missing until you type )

Hmm, strange, I haven't notice that myself.

> Additionally you still can get the prefix name to
> be blank ( or at least I managed to do that ) by attempting to change the
> name by delete, when all the name is deleted you get a message saying the
> sheet prefix is illegal ( actually I find that behaviour annoying e.g. I
> would prefer/expect the error after I press ok ) And if fact with the blank
> ( from the error scenario above ) you can press ok and now the default
> prefix is blank :-)

For me allowing a blank prefix is a feature and not a bug! If someone
wants to have his/her default worksheets named with only numbers (1,
2, 3...) who are we to deny him/her of that? Unless there is some
technical reason against that I have overlooked of course.

> Anyway since this is master and 3.6 is some time away
> there is time to fix it. Anyway looking forward to the next iteration of
> this, keep up the great work

Yes, I will have another look at this.

Thanks Noel for pushing it.

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

[Libreoffice] [PATCH] Remove unused code

[Santiago Martinez]

This patch removes unused code as listed in unusedcode.easy
From 4eaa8adb17c0334f08c83509644b45553254376e Mon Sep 17 00:00:00 2001
From: Santiago Martinez 
Date: Tue, 31 Jan 2012 19:59:31 +0100
Subject: [PATCH] Remove unused code.

---
 chart2/source/view/axes/Tickmarks_Equidistant.cxx |   27 -
 chart2/source/view/axes/Tickmarks_Equidistant.hxx |5 
 unusedcode.easy   |3 --
 3 files changed, 0 insertions(+), 35 deletions(-)

diff --git a/chart2/source/view/axes/Tickmarks_Equidistant.cxx b/chart2/source/view/axes/Tickmarks_Equidistant.cxx
index 7e15dda..73379b0 100644
--- a/chart2/source/view/axes/Tickmarks_Equidistant.cxx
+++ b/chart2/source/view/axes/Tickmarks_Equidistant.cxx
@@ -626,33 +626,6 @@ bool EquidistantTickIter::gotoNext()
 return true;
 }
 
-bool EquidistantTickIter::gotoIndex( sal_Int32 nTickIndex )
-{
-if( nTickIndex < 0 )
-return false;
-if( nTickIndex >= m_nTickCount )
-return false;
-
-if( nTickIndex < m_nCurrentPos )
-if( !gotoFirst() )
-return false;
-
-while( nTickIndex > m_nCurrentPos )
-if( !gotoNext() )
-return false;
-
-return true;
-}
-
-sal_Int32 EquidistantTickIter::getCurrentIndex() const
-{
-return m_nCurrentPos;
-}
-sal_Int32 EquidistantTickIter::getMaxIndex() const
-{
-return m_nTickCount-1;
-}
-
 double* EquidistantTickIter::nextValue()
 {
 if( gotoNext() )
diff --git a/chart2/source/view/axes/Tickmarks_Equidistant.hxx b/chart2/source/view/axes/Tickmarks_Equidistant.hxx
index c312b2f..c9f2dc5 100644
--- a/chart2/source/view/axes/Tickmarks_Equidistant.hxx
+++ b/chart2/source/view/axes/Tickmarks_Equidistant.hxx
@@ -60,11 +60,6 @@ public:
 
 sal_Int32   getCurrentDepth() const { return m_nCurrentDepth; }
 
-protected:
-boolgotoIndex( sal_Int32 nTickIndex );
-sal_Int32   getCurrentIndex() const;
-sal_Int32   getMaxIndex() const;
-
 private: //methods
 sal_Int32   getIntervalCount( sal_Int32 nDepth );
 boolisAtLastPartTick();
diff --git a/unusedcode.easy b/unusedcode.easy
index a54e369..32f3dac 100644
--- a/unusedcode.easy
+++ b/unusedcode.easy
@@ -1228,9 +1228,6 @@ canvas::tools::mergeViewAndRenderState(com::sun::star::rendering::ViewState&, co
 canvas::tools::prependToViewState(com::sun::star::rendering::ViewState&, basegfx::B2DHomMatrix const&)
 canvas::tools::verifyInput(com::sun::star::geometry::RealSize2D const&, char const*, com::sun::star::uno::Reference const&, short)
 canvas::tools::verifyInput(com::sun::star::rendering::FloatingPointBitmapLayout const&, char const*, com::sun::star::uno::Reference const&, short)
-chart::EquidistantTickIter::getCurrentIndex() const
-chart::EquidistantTickIter::getMaxIndex() const
-chart::EquidistantTickIter::gotoIndex(int)
 cmis::Content::exchangeIdentity(com::sun::star::uno::Reference const&)
 cmis::Content::queryChildren(std::__debug::list, std::allocator > >&)
 comphelper::AccessibleEventNotifier::getEventListeners(unsigned int)
-- 
1.7.7.6

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

Re: [Libreoffice] [PATCH V2] convert SdCustomShow from tools/list to vector

[Eike Rathke]

Hi Ivan,

On Tuesday, 2012-01-31 21:31:11 +0400, Ivan Timofeev wrote:

> >should remove the first occurrence, care to fix ?
> 
> so the exact equivalent of
> 
> mpSdCustomShow->Remove(pPage->GetSdrPage())
> 
> will be:
> 
> SdCustomShow::PageVec::iterator it = std::find(
> mpSdCustomShow->PagesVector().begin(),
> mpSdCustomShow->PagesVector().end(),
> pPage->GetSdrPage());
> if (it != mpSdCustomShow->PagesVector().end())
> mpSdCustomShow->PagesVector().erase(it);
> 
> right?

Looks corect. Seeing this it would make sense to reintroduce the
Remove() method that in the patch was merged with the Replace() method,
which otherwise I would had preferred got a comment in the header file
that it actually removed if the replacement was NULL ...

  Eike

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


pgpkK8pmJWTZW.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] tracking ODF compliance issues

[Petr Mladek]

Michael Stahl píše v Út 31. 01. 2012 v 14:26 +0100:
> hi all,
> 
> i'd like to have some easier way to track bugs related to ODF compliance
> in the bugzilla.
> 
> what would be the best way to do this:
> - a new keyword, say "odf" or "odf_validation"
> - a tracking bug, that depends on all indidual ODF related bugs
> 
> the OpenOffice.org issuezilla used a "odf_validation" keyword for issues
> where OOo wrote invalid ODF, but there was no way to track other kinds
> of ODF issues.
> 
> so it would also be possible to have say 2 keywords "odf_validation" for
> those issues and "odf" for everything else, or 2 tracking issue.

I think that keyword might be better in this case. We will always have
such bugs, so the meta bug would stay forever, it would have too many
comments, ...

My only concern is that we should document is somewhere. For example at
http://wiki.documentfoundation.org/BugReport or
http://wiki.documentfoundation.org/BugTriage

Rainer what do you think?
Do we already have list of supported keywords somewhere in the wiki?


Best Regards,
Petr

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

Re: [Libreoffice] [REVIEW][3-5][3-5-0][3-4] Prevent excessvie references to formula result tokens

[Eike Rathke]

Hi Kohei,

On Tuesday, 2012-01-31 09:54:03 -0500, Kohei Yoshida wrote:

> Well, for expediting the process (since the release of 3.5.0 is
> nearing), could you take this over?  Sounds like you know exactly what
> needs to get done, so it's probably better than me trying to figure this
> out on my own.

Done, please review and cherry-pick to 3-5 and 3-5-0 and 3-4
http://cgit.freedesktop.org/libreoffice/core/commit/?id=742b36ed8bd3880d4dcf6022fbc3165a213d80bb

This replaces the previous commit on master which I reverted. The result
token is now shared until exceeding 60k references, leaving 4k for
temporarily passing around the token, undo/redo et all, from that point
on a copy is created for each new result. Later on during recalculation
the result usually will be replaced anyway unless the formula expression
yields a constant, so we don't gain in memory consumption compared to
a new-always approach, but up to 60k cells the copy process is faster
than already new'ing tokens at that stage.

  Eike

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


pgpEOIY75jWdS.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: [Libreoffice] [PUSHED] partial patch for fdo#44551 libvisio: arrow head shapes not imported properly

[Bruno Girin]

On 31/01/12 11:22, Fridrich Strba wrote:

Hello, Bruno

On 31/01/12 00:35, Bruno Girin wrote:

Please find attached a partial patch to import line marker shapes from
Visio files. I would appreciate a review to confirm whether I am on the
right path or whether I should do something different.


Yes, currently this is the best approach to fill in for each id the 
right viewbox and path. I will have to find a way to parse it for the 
svg generation, but that is another question. That was an issue with 
the initial implementation too, but nothing unsolvable either.


Fine, I'll keep working on the remaining 40-odd IDs and may pick your 
brain at FOSDEM if I get stuck on any of them.


Cheers,

Bruno

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

Re: [Libreoffice] [PATCH] Making default tab prefix name configurable in Calc

[Noel Power]

On 31/01/12 18:35, Albert Thuswaldner wrote:

Additionally you still can get the prefix name to
>  be blank ( or at least I managed to do that ) by attempting to change the
>  name by delete, when all the name is deleted you get a message saying the
>  sheet prefix is illegal ( actually I find that behaviour annoying e.g. I
>  would prefer/expect the error after I press ok ) And if fact with the blank
>  ( from the error scenario above ) you can press ok and now the default
>  prefix is blank:-)

For me allowing a blank prefix is a feature and not a bug!
:-) well, I guess one of the calc experts might have an opinion on that. 
My gripe is that as I delete the prefix when all letters are gone ( e.g. 
now we have a blank prefix ) I get a dialog telling me that the sheet 
name is illegal, so.. naturally I assumed that's what you intended. 
Additionally I can enter any other so-called illegal character e.g. '*' 
or '/' etc. get the same error dialog and  *still* can save a prefix 
containing those characters.
Personally I would not check every keypress, an alternative would be to 
check the prefix when you press 'ok' and if it is illegal then don't 
close the option dialog but instead restore the previous 'legal' entry ( 
selecting the entry probably is an idea also ) and raise the error 
dialog ( but there are other workflows I suppose )

  If someone
wants to have his/her default worksheets named with only numbers (1,
2, 3...) who are we to deny him/her of that? Unless there is some
technical reason against that I have overlooked of course.

just what I mentioned above

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

Re: [Libreoffice] [REVIEW][3-5][3-5-0][3-4] Prevent excessvie references to formula result tokens

[Kohei Yoshida]

On Tue, 2012-01-31 at 20:54 +0100, Eike Rathke wrote:
> Hi Kohei,
> 
> On Tuesday, 2012-01-31 09:54:03 -0500, Kohei Yoshida wrote:
> 
> > Well, for expediting the process (since the release of 3.5.0 is
> > nearing), could you take this over?  Sounds like you know exactly what
> > needs to get done, so it's probably better than me trying to figure this
> > out on my own.
> 
> Done, please review and cherry-pick to 3-5 and 3-5-0 and 3-4
> http://cgit.freedesktop.org/libreoffice/core/commit/?id=742b36ed8bd3880d4dcf6022fbc3165a213d80bb

Thanks Eike.  Pushed to 3-5 and 3-4 with my sign-off.

We still need 2 more sign-off's push this into 3-5-0.

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

Re: [Libreoffice] [PATCH] Making default tab prefix name configurable in Calc

[Kohei Yoshida]

On Tue, 2012-01-31 at 20:07 +, Noel Power wrote:
> On 31/01/12 18:35, Albert Thuswaldner wrote:
> >> Additionally you still can get the prefix name to
> >> >  be blank ( or at least I managed to do that ) by attempting to change 
> >> > the
> >> >  name by delete, when all the name is deleted you get a message saying 
> >> > the
> >> >  sheet prefix is illegal ( actually I find that behaviour annoying e.g. I
> >> >  would prefer/expect the error after I press ok ) And if fact with the 
> >> > blank
> >> >  ( from the error scenario above ) you can press ok and now the default
> >> >  prefix is blank:-)
> > For me allowing a blank prefix is a feature and not a bug!
> :-) well, I guess one of the calc experts might have an opinion on that. 

Honestly, no clue whether an empty sheet prefix is truly a problem or
not.  We can certainly experiment it and if that's indeed a problem we
can add that check back in.  I would think that, as long as the actual
sheet name is not empty, it *should* probably be okay.

> Personally I would not check every keypress, an alternative would be to 
> check the prefix when you press 'ok' and if it is illegal then don't 
> close the option dialog but instead restore the previous 'legal' entry ( 
> selecting the entry probably is an idea also ) and raise the error 
> dialog ( but there are other workflows I suppose )

Well, that's rather hard to do with the Options dialog, since you can
freely switch between different option pages and we then would have to
check the validity of all the option tab pages upon dismissal of the
whole Options dialog...  I've been on that boat before, and it's very
tricky to implement.  Especially when two or more option pages want to
throw an error dialog, you may end up with lots of error dialogs when
clicking OK.

Given that, I still think it's better (and easier) to check for every
key-press, but instead of launching an error dialog for an illegal name,
just put the old legal name back into the text box and sound an beep or
something.  That's less intrusive than a big dialog being thrown right
in your face.  I do exactly that for the argument separator value inputs
in the Formula option page.

But that's not really a big issue and I (or someone else) can fix that
later once the craze of 3.5 bug fixing becomes less intense.

Just my 2-cents.

Kohei

-- 
Kohei Yoshida, LibreOffice hacker, Calc

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

Re: [Libreoffice] [PATCH] Making default tab prefix name configurable in Calc

[Eike Rathke]

Hi Albert,

On Tuesday, 2012-01-31 19:35:54 +0100, Albert Thuswaldner wrote:

> > However something I noticed is that there are still some problems with
> > entering the 'Prefix name" in the options, firstly the input box seems to be
> > overwritten by the label ( part of the input/edit field for the prefix is
> > missing until you type )
> 
> Hmm, strange, I haven't notice that myself.

Just to say that I see the same here, seems the fixed text partly
overwrites the edit box, it's just a few pixels, probably an error in
calculating size of fixed text and position of edit box in the resource
file.

> > Additionally you still can get the prefix name to
> > be blank ( or at least I managed to do that ) by attempting to change the
> > name by delete, when all the name is deleted you get a message saying the
> > sheet prefix is illegal ( actually I find that behaviour annoying e.g. I
> > would prefer/expect the error after I press ok ) And if fact with the blank
> > ( from the error scenario above ) you can press ok and now the default
> > prefix is blank :-)
> 
> For me allowing a blank prefix is a feature and not a bug! If someone
> wants to have his/her default worksheets named with only numbers (1,
> 2, 3...) who are we to deny him/her of that? Unless there is some
> technical reason against that I have overlooked of course.

No, there isn't, a number only is fine and is often used with year
numbers. The user will have to bear with the consequences that the sheet
name will be enclosed in single quotes when formulas are displayed ;-)

  Eike

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


pgppbkfOiTr5X.pgp
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/libreoffice

[Libreoffice] [PUSHD][PATCH] Remove unused code

[Caolán McNamara]

On Tue, 2012-01-31 at 20:03 +0100, Santiago Martinez wrote:
> This patch removes unused code as listed in unusedcode.easy

Looks good to me, pushed, thanks for this.

C.

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