Re: Crash test update

[Caolán McNamara]

On Wed, 2023-11-15 at 05:04 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  0c0b8b8e92177da65/>.
> 
> 27 files have crashed during import.
> 
> 3 files have crashed during export.

Seeing as I keep a list of which crashes are duplicates and what
commit, if known, they started happening at, I've put that at
https://wiki.documentfoundation.org/Development/Crashtesting
and will try to keep it up to date there

Re: Crash test update

[Caolán McNamara]

On Mon, 2023-09-18 at 09:52 +0200, Miklos Vajna wrote:
> Hi,
> 
> On Sat, Sep 16, 2023 at 01:59:22PM +, crashtest
>  wrote:
> > New crashtest update available at
> >  > af647d82b9a51871a22/>.
> > 
> > 22 files have crashed during import.
> 
> My notes on the Writer ones:
> ...
> I haven't tried to bisect what causes the new
> SwTableBoxFormula::GetTableBox() and SwFrame::GetVirtPageNum()
> crashes yet.

I did that earlier, and both are bisected to
https://gerrit.libreoffice.org/c/core/+/156306
there's now a note there with a sample how to reproduce

Re: Crash test update

[Miklos Vajna]

Hi,

On Sat, Sep 16, 2023 at 01:59:22PM +, crashtest 
 wrote:
> New crashtest update available at 
> .
> 
> 22 files have crashed during import.

My notes on the Writer ones:

New:
forums/odt/forum-de3-8445.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
forums/odt/forum-de2-6191.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
forums/odt/forum-de2-3966.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
bugtrackers/odt/tdf92502-1.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
bugtrackers/odt/tdf92502-3.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
bugtrackers/odt/tdf109301-1.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
bugtrackers/sxw/ooo58307-1.sxw through disposed -> crash in 
SwTableBoxFormula::GetTableBox()
forums/odt/forum-de2-2840.odt through disposed -> crash in 
SwFrame::GetVirtPageNum(), called by SwTextFormatter::NewFieldPortion()
forums/odt/forum-en-3872.odt through disposed -> crash in 
SwTableBoxFormula::GetTableBox()

PDF/UA and SDT asserts:
forums/docx/forum-mso-de-118466.docx through disposed: "this is presumably dead 
code" assertion -> I submitted a fix for this
forums/docx/forum-mso-de-76224.docx through disposed: "this is presumably dead 
code" assertion
forums/docx/forum-mso-de-118517.docx through disposed: "this is presumably dead 
code" assertion
forums/docx/forum-mso-de-118440.docx through disposed: "this is presumably dead 
code" assertion
forums/docx/forum-mso-de-118507.docx through disposed: "this is presumably dead 
code" assertion
forums/docx/forum-mso-de-118414.docx through disposed: "this is presumably dead 
code" assertion

To fix, older:
forums/docx/forum-mso-en-4208.docx through disposed: crash in sw::CalcBreaks
forums/docx/forum-mso-en4-137999.docx through disposed: crash in 
SwSortedObjs::Insert
forums/docx/forum-mso-en-9381.docx through disposed: crash in 
SwSortedObjs::Insert
bugtrackers/docx/fdo45195-1.docx through disposed: crash in SwSortedObjs::Insert
bugtrackers/docx/fdo45193-1.docx through disposed: crash in SwSortedObjs::Insert
forums/odt/forum-en-12562.odt through disposed -> sw::mark::CrossRefBookmark 
ctor

12 old Writer ones + 9 new ones.

I haven't tried to bisect what causes the new
SwTableBoxFormula::GetTableBox() and SwFrame::GetVirtPageNum() crashes
yet.

Regards,

Miklos

Re: Crash test update

[Caolán McNamara]

On Fri, 2023-07-07 at 18:05 +, crashtest wrote:
> Hi,
> 
> New crashtest update available
> at:https://dev-builds.libreoffice.org/crashtest/650ab6e645c51aaffc8dc5e
> 638295d6da74d44ff
> 
> 26 files have crashed during import.

Seeing as we have returned to "normal" I'll summarize to the 6
outstanding issues:

==recent==

SwNodes::FindPrvNxtFrameNode
-> bisected to https://gerrit.libreoffice.org/c/core/+/127272
   (document provided)
# 60 forums/odt/forum-de3-11230.odt

==old==

ScFormulaCell::MaybeInterpret() <--- these are the PITA ones
# 891 forums/xls/forum-mso-de-48401.xls
# 2679 forums/xlsx/forum-mso-en4-237685.xlsx
# 2735 forums/xlsx/forum-mso-en4-271721.xlsx
# 3354 forums/xlsx/forum-mso-en4-237668.xlsx
# 3609 forums/xlsx/forum-mso-en4-604372.xlsx
# 3748 forums/xlsx/forum-mso-en4-314789.xslx
# 6837 forums/xlsx/forum-mso-en4-264966.xlsx
# 6480 forums/xlsx/forum-mso-en4-264977.xlsx
# 5577 forums/xlsx/forum-mso-en4-457928.xlsx
# 5217 forums/xlsx/forum-mso-en4-604336.xlsx
# 5053 forums/xlsx/forum-mso-en4-784502.xlsx
# 4935 forums/xlsx/forum-mso-en4-184573.xlsx
# 4813 forums/xlsx/forum-mso-en4-264966.xlsx

SwSubFont::GetTextSize_
# 629 forums/docx/forum-mso-de-118466.docx
# 641 forums/docx/forum-mso-de-76224.docx
# 649 forums/docx/forum-mso-de-118517.docx
# 656 forums/docx/forum-mso-de-118440.docx
# 664 forums/docx/forum-mso-de-118507.docx
# 679 forums/docx/forum-mso-de-118414.docx

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192
(noted there)
# 672 forums/docx/forum-mso-en-9381.docx
# 666 forums/docx/forum-mso-en4-137999.docx
# 7704 bugtrackers/docx/fdo45193-1.docx
# 7671 bugtrackers/docx/fdo45195-1.docx

sw::CalcBreaks
# 653 forums/docx/forum-mso-en-4208.docx
   -> https://gerrit.libreoffice.org/c/core/+/127015 
   -> https://gerrit.libreoffice.org/c/core/+/127127
I suspect the SwSubFont::GetTextSize_ are similar
   -> https://bugs.documentfoundation.org/show_bug.cgi?id=152038

sw::mark::CrossRefBookmark::CrossRefBookmark
# 92 forums/odt/forum-en-12562.odt

Re: Crash test update

[Caolán McNamara]

On Wed, 2023-05-10 at 00:04 +, crashtest wrote:
> Hi,
> 
> New crashtest update available
> athttps://dev-builds.libreoffice.org/crashtest/6c042848b688f64b3c56d65
> dd9dc5fe85412660
> 
> 43 files have crashed during import.
> 
> 17 files have crashed during export.

I'll add my categorization here. As far as I can tell there are no more
split fly crashes, and the mysterious calc crash I had blamed on
threaded import has also happily gone away.

==new==

SwPosition::SetContent
-> bisected to https://gerrit.libreoffice.org/c/core/+/150772 (noted
there)
# 36 forums/odt/forum-de3-17466.odt
# 51 forums/odt/forum-de3-17467.odt
# 7087 bugtrackers/odt/tdf99692-1.odt
# 7041 bugtrackers/odt/tdf116139-3.odt

==recent==

SwNodes::FindPrvNxtFrameNode
-> bisected to https://gerrit.libreoffice.org/c/core/+/127272 (document
provided)
# 60 forums/odt/forum-de3-11230.odt

==old==

ScFormulaCell::MaybeInterpret()
# 891 forums/xls/forum-mso-de-48401.xls
# 2679 forums/xlsx/forum-mso-en4-237685.xlsx
# 2735 forums/xlsx/forum-mso-en4-271721.xlsx
# 3354 forums/xlsx/forum-mso-en4-237668.xlsx
# 3609 forums/xlsx/forum-mso-en4-604372.xlsx
# 3748 forums/xlsx/forum-mso-en4-314789.xslx
# 6837 forums/xlsx/forum-mso-en4-264966.xlsx
# 6480 forums/xlsx/forum-mso-en4-264977.xlsx
# 5577 forums/xlsx/forum-mso-en4-457928.xlsx
# 5217 forums/xlsx/forum-mso-en4-604336.xlsx
# 5053 forums/xlsx/forum-mso-en4-784502.xlsx
# 4935 forums/xlsx/forum-mso-en4-184573.xlsx
# 4813 forums/xlsx/forum-mso-en4-264966.xlsx

SwSubFont::GetTextSize_
# 629 forums/docx/forum-mso-de-118466.docx
# 641 forums/docx/forum-mso-de-76224.docx
# 649 forums/docx/forum-mso-de-118517.docx
# 656 forums/docx/forum-mso-de-118440.docx
# 664 forums/docx/forum-mso-de-118507.docx
# 679 forums/docx/forum-mso-de-118414.docx

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192 (noted
there)
# 672 forums/docx/forum-mso-en-9381.docx
# 666 forums/docx/forum-mso-en4-137999.docx
# 7704 bugtrackers/docx/fdo45193-1.docx
# 7671 bugtrackers/docx/fdo45195-1.docx

sw::CalcBreaks
# 653 forums/docx/forum-mso-en-4208.docx
   -> https://gerrit.libreoffice.org/c/core/+/127015 
   -> https://gerrit.libreoffice.org/c/core/+/127127
I suspect the SwSubFont::GetTextSize_ are similar
   -> https://bugs.documentfoundation.org/show_bug.cgi?id=152038

sw::mark::CrossRefBookmark::CrossRefBookmark
# 92 forums/odt/forum-en-12562.odt

SwNode::GetFlyFormat
# 705 forums/docx/forum-mso-de-107658.docx
 -> bisected to https://gerrit.libreoffice.org/c/core/+/145162 (noted
there)

==fix submitted==

SwTextPainter::DrawTextLine
-> bisected to https://gerrit.libreoffice.org/c/core/+/148556 (noted
there)
# 7181 bugtrackers/odt/moz386569-2.odt

SwTextFrame::PaintSwFrame
-> bisected to https://gerrit.libreoffice.org/c/core/+/149058 (noted
there)
# 7192 bugtrackers/odt/fdo38226-1.odt
# 7063 bugtrackers/odt/ooo58028-1.odt
# 7042 bugtrackers/odt/ooo34309-3.odt
# 7035 bugtrackers/odt/ooo90841-1.odt
# 621 forums/docx/forum-mso-de-59967.odt
# 1922 forums/doc/forum-mso-de-69704.odt
# 1915 forums/doc/forum-mso-de-69714.odt
# 1909 forums/doc/forum-mso-de-70196.odt
# 1908 forums/doc/forum-mso-de-69695.odt
# 1886 forums/doc/forum-mso-de-70197.odt

Re: Crash test update

[Miklos Vajna]

Hi,

On Sun, Apr 30, 2023 at 03:43:27PM +, crashtest 
 wrote:
> 165 files have crashed during import.

Here are the import crashes I can reproduce:

PDF/UA and SDT asserts:
- forums/doc/forum-mso-de-70197.doc: oTaggedLabel assertion
- forums/doc/forum-mso-de-69695.doc: oTaggedLabel assertion
- forums/doc/forum-mso-de-70196.doc: oTaggedLabel assertion
- forums/doc/forum-mso-de-69714.doc: oTaggedLabel assertion
- forums/doc/forum-mso-de-69704.doc: oTaggedLabel assertion
- forums/docx/forum-mso-de-59967.docx: oTaggedLabel assertion
- forums/docx/forum-mso-de-118466.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-en-2718.docx: oTaggedParagraph assertion
- forums/docx/forum-mso-de-76224.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-de-118517.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-de-118440.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-de-118507.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-de-118414.docx: "this is presumably dead code" assertion
- forums/docx/forum-mso-en-2717.docx: oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-1.docx: oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-3.docx: oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-4.docx: oTaggedLabel assertion

This doesn't include the 2 fixes from earlier today by Michael S.

To fix, older:
- forums/docx/forum-mso-en-4208.docx: crash in sw::CalcBreaks
- forums/docx/forum-mso-en4-137999.docx: crash in SwSortedObjs::Insert
- forums/docx/forum-mso-en-9381.docx: crash in SwSortedObjs::Insert
- bugtrackers/docx/fdo45195-1.docx: crash in SwSortedObjs::Insert
- bugtrackers/docx/fdo45193-1.docx: crash in SwSortedObjs::Insert

To fix, newer:

- bugtrackers/doc/ooo9470-1.doc: split flys, I'm working on a fix.

Regards,

Miklos

Re: Crash test update

[Miklos Vajna]

Hi,

On Tue, May 02, 2023 at 11:18:48AM +0200, Miklos Vajna  
wrote:
> I see that 67 of these are DOCX/DOC/RTF files, I'll take a look at those
> to address split fly problems.

The subset I can still reproduce is:

Tagged PDF and SDT asserts:
- forums/doc/forum-mso-de-70197.doc -> oTaggedLabel assertion
- forums/doc/forum-mso-de-69695.doc -> oTaggedLabel assertion
- forums/doc/forum-mso-de-70196.doc -> oTaggedLabel assertion
- forums/doc/forum-mso-de-69714.doc -> oTaggedLabel assertion
- forums/doc/forum-mso-de-69704.doc -> oTaggedLabel assertion
- forums/docx/forum-mso-de-59967.docx -> oTaggedLabel assertion
- forums/docx/forum-mso-de-118466.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-en-2718.docx -> oTaggedParagraph assertion
- forums/docx/forum-mso-de-76224.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-de-118517.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-de-118440.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-de-118507.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-de-118414.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-de-98687.docx -> "this is presumably dead code" 
assertion
- forums/docx/forum-mso-en-2717.docx -> oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-1.docx -> oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-3.docx -> oTaggedParagraph assertion
- bugtrackers/docx/tdf124604-4.docx -> oTaggedLabel assertion

Other, older:

- forums/docx/forum-mso-en-4208.docx -> crash in sw::CalcBreaks
- forums/docx/forum-mso-en4-137999.docx -> crash in SwSortedObjs::Insert
- forums/docx/forum-mso-en-9381.docx -> crash in SwSortedObjs::Insert
- bugtrackers/docx/fdo45195-1.docx -> crash in SwSortedObjs::Insert
- bugtrackers/docx/fdo45193-1.docx -> crash in SwSortedObjs::Insert

Split flys:
- bugtrackers/doc/ooo9470-1.doc: assertion failure in 
SwObjectFormatterTextFrame::DoFormatObj() -> this is probably related to split 
flys, I'll investigate.

Regards,

Miklos

Re: Crash test update

[Miklos Vajna]

Hi,

On Mon, Apr 24, 2023 at 10:37:25PM +, crashtest 
 wrote:
> New crashtest update available at 
> .
> 
> 94 files have crashed during import.

I see that 67 of these are DOCX/DOC/RTF files, I'll take a look at those
to address split fly problems.

Regards,

Miklos

Re: Crash test update

[Miklos Vajna]

Hi Caolan,

On Tue, Apr 18, 2023 at 07:42:44PM +0100, Caolán McNamara  
wrote:
> SwLayoutFrame::Lower
> -> https://gerrit.libreoffice.org/c/core/+/150252 (noted there)
> # 7718 bugtrackers/docx/tdf128700-2.docx
> # 7667 bugtrackers/docx/tdf96724-1.docx
> # 730 forums/docx/forum-mso-de-117064.docx
> # 700 forums/docx/forum-mso-de-117070.docx
> # 692 forums/docx/forum-mso-de-77474.docx
> # 651 forums/docx/forum-mso-de-116268.docx
> # 628 forums/docx/forum-mso-de-133201.docx
> 
> SwRectFnSet::GetHeight
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7711 bugtrackers/docx/fdo80989-1.docx
> 
> SwFrame::PrepareMake
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7689 bugtrackers/docx/tdf136909-1.docx
> 
> SwFrame::ImplInvalidateNextPos
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7678 bugtrackers/docx/fdo72790-1.docx
> 
> SwFrame::Grow
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7676 bugtrackers/docx/fdo80989-2.docx
> # 691 forums/docx/forum-fr-21005.docx
> # 638 forums/docx/forum-mso-en-14827.docx
> 
> CalcClipRect
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7675 bugtrackers/docx/tdf97520-2.docx
> # 7673 bugtrackers/docx/tdf114111-3.docx
> # 698 forums/docx/forum-mso-de-133531.docx
> # 689 forums/docx/forum-mso-en3-26783.docx
> # 688 forums/docx/forum-mso-en4-600195.docx
> # 682 forums/docx/forum-mso-en-18580.docx
> # 650 forums/docx/forum-mso-de-113939.docx
> # 626 forums/docx/forum-mso-en4-548497.docx
> 
> SwFrame::IsInSplitTableRow
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 681 forums/docx/forum-mso-de-98687.docx
> 
> SwFrame::PrepareMake
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # bugtrackers/docx/tdf136931-1.docx
> 
> SwLayoutFrame::GrowFrame
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7670 bugtrackers/docx/moz1023726-1.docx
> 
> SwEnterLeave::SwEnterLeave
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7668 bugtrackers/docx/fdo72775-2.docx
> 
> SwFlowFrame::GetFollow
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 7665 bugtrackers/docx/fdo75943-1.docx
> 
> SwFrame::GetNextSctLeaf
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 728 forums/docx/forum-mso-de-127520.docx
> # 712 forums/docx/forum-mso-de-118811.docx
> # 695 forums/docx/forum-mso-en-13367.docx
> # 694 forums/docx/forum-mso-de-102228.docx
> # 665 forums/docx/forum-mso-de-127521.docx
> # 623 forums/docx/forum-mso-de-127524.docx
> 
> SwCellFrame::FindStartEndOfRowSpanCell
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 670 forums/docx/forum-mso-en-17219.docx
> 
> SwFrame::FindNext_
> -> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
> # 618 forums/docx/forum-mso-de-118743.docx

Thanks, will look into these.

Regards,

Miklos

Re: Crash test update

[Caolán McNamara]

On Tue, 2023-04-18 at 02:17 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  f5bce7a4ee30df38d/>.
> 
> 91 files have crashed during import.
> 
> 20 files have crashed during export.

Here's my breakdown of the categories, I've only confirmed the first
new category as being from the new floating tables as split fly work
and assumed the same for the new writer layout asserts/crashes.

==new==

SwLayoutFrame::Lower
-> https://gerrit.libreoffice.org/c/core/+/150252 (noted there)
# 7718 bugtrackers/docx/tdf128700-2.docx
# 7667 bugtrackers/docx/tdf96724-1.docx
# 730 forums/docx/forum-mso-de-117064.docx
# 700 forums/docx/forum-mso-de-117070.docx
# 692 forums/docx/forum-mso-de-77474.docx
# 651 forums/docx/forum-mso-de-116268.docx
# 628 forums/docx/forum-mso-de-133201.docx

SwRectFnSet::GetHeight
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7711 bugtrackers/docx/fdo80989-1.docx

SwFrame::PrepareMake
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7689 bugtrackers/docx/tdf136909-1.docx

SwFrame::ImplInvalidateNextPos
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7678 bugtrackers/docx/fdo72790-1.docx

SwFrame::Grow
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7676 bugtrackers/docx/fdo80989-2.docx
# 691 forums/docx/forum-fr-21005.docx
# 638 forums/docx/forum-mso-en-14827.docx

CalcClipRect
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7675 bugtrackers/docx/tdf97520-2.docx
# 7673 bugtrackers/docx/tdf114111-3.docx
# 698 forums/docx/forum-mso-de-133531.docx
# 689 forums/docx/forum-mso-en3-26783.docx
# 688 forums/docx/forum-mso-en4-600195.docx
# 682 forums/docx/forum-mso-en-18580.docx
# 650 forums/docx/forum-mso-de-113939.docx
# 626 forums/docx/forum-mso-en4-548497.docx

SwFrame::IsInSplitTableRow
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 681 forums/docx/forum-mso-de-98687.docx

SwFrame::PrepareMake
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# bugtrackers/docx/tdf136931-1.docx

SwLayoutFrame::GrowFrame
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7670 bugtrackers/docx/moz1023726-1.docx

SwEnterLeave::SwEnterLeave
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7668 bugtrackers/docx/fdo72775-2.docx

SwFlowFrame::GetFollow
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 7665 bugtrackers/docx/fdo75943-1.docx

SwFrame::GetNextSctLeaf
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 728 forums/docx/forum-mso-de-127520.docx
# 712 forums/docx/forum-mso-de-118811.docx
# 695 forums/docx/forum-mso-en-13367.docx
# 694 forums/docx/forum-mso-de-102228.docx
# 665 forums/docx/forum-mso-de-127521.docx
# 623 forums/docx/forum-mso-de-127524.docx

SwCellFrame::FindStartEndOfRowSpanCell
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 670 forums/docx/forum-mso-en-17219.docx

SwFrame::FindNext_
-> https://gerrit.libreoffice.org/c/core/+/150252 (assumed)
# 618 forums/docx/forum-mso-de-118743.docx

==recent==

SwTextPainter::DrawTextLine
-> https://gerrit.libreoffice.org/c/core/+/148556 (mentioned there)
# 7953 bugtrackers/doc/fdo44018-2.odt
# 7186 bugtrackers/odt/ooo95981-1.odt
# 7181 bugtrackers/odt/moz386569-2.odt

SwNodes::FindPrvNxtFrameNode
-> https://gerrit.libreoffice.org/c/core/+/127272 (document provided)
# 60 forums/odt/forum-de3-11230.odt

SwTextFrame::PaintSwFrame
-> https://gerrit.libreoffice.org/c/core/+/149058
# 7712 bugtrackers/docx/tdf124604-4.odt
# 7697 bugtrackers/docx/tdf124604-3.odt
# 7672 bugtrackers/docx/tdf124604-1.odt
# 7192 bugtrackers/odt/fdo38226-1.odt
# 711 docx/forum-mso-en-2717.odt
# 7063 bugtrackers/odt/ooo58028-1.odt
# 7042 bugtrackers/odt/ooo34309-3.odt
# 7035 bugtrackers/odt/ooo90841-1.odt
# 632 forums/docx/forum-mso-en-2718.odt
# 621 forums/docx/forum-mso-de-59967.odt
# 1922 forums/doc/forum-mso-de-69704.odt
# 1915 forums/doc/forum-mso-de-69714.odt
# 1909 forums/doc/forum-mso-de-70196.odt
# 1908 forums/doc/forum-mso-de-69695.odt
# 1886 forums/doc/forum-mso-de-70197.odt

==old==

ScFormulaCell::MaybeInterpret()
# 891 forums/xls/forum-mso-de-48401.xls
# 2679 forums/xlsx/forum-mso-en4-237685.xlsx
# 2735 forums/xlsx/forum-mso-en4-271721.xlsx
# 3354 forums/xlsx/forum-mso-en4-237668.xlsx
# 3609 forums/xlsx/forum-mso-en4-604372.xlsx
# 3748 forums/xlsx/forum-mso-en4-314789.xslx
# 6837 forums/xlsx/forum-mso-en4-264966.xlsx
# 6480 forums/xlsx/forum-mso-en4-264977.xlsx
# 5577 forums/xlsx/forum-mso-en4-457928.xlsx
# 5217 forums/xlsx/forum-mso-en4-604336.xlsx
# 5053 forums/xlsx/forum-mso-en4-784502.xlsx
# 4935 forums/xlsx/forum-mso-en4-184573.xlsx
# 4813 forums/xlsx/forum-mso-en4-264966.xlsx

SwSubFont::GetTextSize_
# 629 forums/docx/forum-mso-de-118466.docx
# 641 forums/docx/forum-mso-de-76224.docx
# 649 forums/docx/forum-mso-de-118517.docx
# 656 forums/docx/forum-mso-de-118440.docx
# 664 forums/docx/forum-mso-de-118507.docx
# 679 forums/docx/

Re: Crash test update

[Caolán McNamara]

On Wed, 2023-03-08 at 06:58 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  a0cba5390a7b935d4/>.
> 
> 58 files have crashed during import.
> 
> 4 files have crashed during export.

Things have returned to mostly normal, so I'll add in the groupings of
issues, there appear to be 10 or 11 issues.

Testing the "ScInterpreter::ScCountIfs" one I was able to reproduce it
and find that SC_NO_THREADED_CALCULATION=1 makes it pass so I presume a
threading issue. tsan is silent on the topic and debugging when
threaded vs unthreaded to me seems to be quite divergent on what
happens so it's unclear what exactly gets clobbered in the threaded
case.

new
---

DocumentFieldsManager::UpdateTableFields
 -> bisect to https://gerrit.libreoffice.org/c/core/+/146611 (which is
a weird outcome)
696 forums/docx/forum-mso-en-8069.docx
663 forums/docx/forum-mso-en-8072.docx

old
---

ScQueryCellIterator< FindEqualOrSortedLastInRange
 -> assert(!"FindEqualOrSortedLastInRange: rEntry.nField -= nColDiff <
0")
 -> proposed fix of: https://gerrit.libreoffice.org/c/core/+/148415
796 forums/xls/forum-mso-en4-144513.xls
901 forums/xls/forum-mso-en4-433310.xls
1034 forums/xls/forum-mso-en4-130543.xls
1132 forums/xls/forum-mso-en4-433308.xls
1461 forums/xls/forum-mso-en4-433031.xls
1601 forums/xls/forum-mso-en4-156881.xls
2074 forums/xlsx/forum-mso-en4-491002.xlsx
2504 forums/xlsx/forum-mso-en4-527778.xlsx
2564 forums/xlsx/forum-mso-en4-606273.xlsx
2945 forums/xlsx/forum-mso-en4-643938.xlsx
3016 forums/xlsx/forum-mso-en4-544867.xlsx
3111 forums/xlsx/forum-mso-en4-629391.xlsx
3928 forums/xlsx/forum-mso-en4-353197.xlsx
4501 forums/xlsx/forum-mso-en4-544866.xlsx
4801 forums/xlsx/forum-mso-en4-606285.xlsx
6552 forums/xlsx/forum-mso-en4-606274.xlsx
6119 forums/xlsx/forum-mso-en4-70163.xlsx
6101 forums/xlsx/forum-mso-en4-183459.xlsx
5500 forums/xlsx/forum-mso-en4-112469.xlsx
5253 forums/xlsx/forum-mso-en4-606289.xlsx
5176 forums/xlsx/forum-mso-de-62049.xlsx
5379 forums/xlsx/forum-mso-de-62049.xlsx

ScFormulaCell::MaybeInterpret()
891 forums/xls/forum-mso-de-48401.xls
2679 forums/xlsx/forum-mso-en4-237685.xlsx
2735 forums/xlsx/forum-mso-en4-271721.xlsx
3354 forums/xlsx/forum-mso-en4-237668.xlsx
3748 forums/xlsx/forum-mso-en4-314789.xslx
6837 forums/xlsx/forum-mso-en4-264966.xlsx
6480 forums/xlsx/forum-mso-en4-264977.xlsx
5577 forums/xlsx/forum-mso-en4-457928.xlsx
5217 forums/xlsx/forum-mso-en4-604336.xlsx
5053 forums/xlsx/forum-mso-en4-784502.xlsx
4935 forums/xlsx/forum-mso-en4-184573.xlsx
4813 forums/xlsx/forum-mso-en4-264966.xlsx

SwSubFont::GetTextSize_
629 forums/docx/forum-mso-de-118466.docx
641 forums/docx/forum-mso-de-76224.docx
649 forums/docx/forum-mso-de-118517.docx
656 forums/docx/forum-mso-de-118440.docx
664 forums/docx/forum-mso-de-118507.docx
679 forums/docx/forum-mso-de-118414.docx

ScInterpreter::ScCountIfs
 -> I note that SC_NO_THREADED_CALCULATION=1 doesn't crash
reproducible for me, but seems quite different threaded
vs unthreaded.
1468 forums/xls/forum-mso-en4-76565.xls
3097 forums/xlsx/forum-mso-en4-554783.xlsx
3344 forums/xlsx/forum-mso-de-103125.xlsx
3853 forums/xlsx/forum-mso-en4-623773.xslx
790 forums/xls/forum-mso-en4-75791.xls
835 forums/xls/forum-mso-en4-601987.xls

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192 (noted
there)
672 forums/docx/forum-mso-en-9381.docx
666 forums/docx/forum-mso-en4-137999.docx
7704 bugtrackers/docx/fdo45193-1.docx
7671 bugtrackers/docx/fdo45195-1.docx

sw::CalcBreaks
653 forums/docx/forum-mso-en-4208.docx
   -> https://gerrit.libreoffice.org/c/core/+/127015 
   -> https://gerrit.libreoffice.org/c/core/+/127127
I suspect the SwSubFont::GetTextSize_ are similar
   -> https://bugs.documentfoundation.org/show_bug.cgi?id=152038

sw::mark::CrossRefBookmark::CrossRefBookmark
92 forums/odt/forum-en-12562.odt

infinite export
704 forums/docx/forum-mso-de-102589.docx

SwNode::GetFlyFormat
705 forums/docx/forum-mso-de-107658.docx
 -> bisected to https://gerrit.libreoffice.org/c/core/+/145162 (noted
there)

ScInterpreter::ScSumIfs
 -> same as Scinterpreter::IterateParametersIfs ?
3609 forums/xlsx/forum-mso-en4-604372.xlsx

- fix submitted -

ImplCalcMapResolution
7445 files/bugtrackers/svg/moz1297206-1.odg

Re: Crash test update

[Caolán McNamara]

On Thu, 2023-02-16 at 07:57 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  fe3cc634c3573f778/>.
> 
> 141 files have crashed during import.
> 
> 4 files have crashed during export.

I've categorized these here again. The jump is from the "new" section,
mostly asserts to find documents that have a specific feature or
unexpected use of properties. Though the bisect from
DocumentFieldsManager::UpdateTableFields puzzle me, perhaps just
different font metrics triggers a bug we always had rather than a
true root cause.

new
---

writerfilter::dmapper::IsFieldNestingAllowed
 -> temp from https://gerrit.libreoffice.org/c/core/+/145931 assert to
find examples
# 8122 cloudon/File_258.docx
# 7708 bugtrackers/docx/tdf91384-1.docx
# 7685 bugtrackers/docx/tdf64372-3.docx
# 7666 bugtrackers/docx/tdf111341-6.docx
# 727 forums/docx/forum-mso-en-3703.docx
# 723 forums/docx/forum-mso-en-18492.docx
# 722 forums/docx/forum-mso-en-9978.docx
# 718 forums/docx/forum-mso-en-11813.docx
# 716 forums/docx/forum-mso-en-6863.docx"
# 715 forums/docx/forum-mso-en-6864.docx
# 702 forums/docx/forum-mso-en-18495.docx
# 701 forums/docx/forum-mso-en4-547666.docx
# 697 forums/docx/forum-mso-en-11731.docx
# 695 forums/docx/forum-mso-en-4096.docx
# 691 forums/docx/forum-mso-en-14436.docx
# 677 forums/docx/forum-mso-en-18503.docx
# 675 forums/docx/forum-mso-en-18154.docx
# 673 forums/docx/forum-mso-en-5.docx
# 671 forums/docx/forum-mso-en-15858.docx
# 667 forums/docx/forum-mso-en-15991.docx
# 662 forums/docx/forum-mso-en-18132.docx
# 659 forums/docx/forum-mso-de-106340.docx
# 658 forums/docx/forum-mso-en-12814.docx
# 656 forums/docx/forum-mso-en-15841.docx
# 655 forums/docx/forum-mso-en4-547665.docx
# 650 forums/docx/forum-mso-en-3205.docx
# 648 forums/docx/forum-mso-en-4306.docx
# 646 forums/docx/forum-mso-en-16545.docx
# 644 forums/docx/forum-mso-en-11777.docx
# 643 forums/docx/forum-mso-de-86461.docx
# 639 forums/docx/forum-mso-de-136366.docx
# 636 forums/docx/forum-mso-de-92780.docx
# 632 forums/docx/forum-mso-de-136389.docx
# 631 forums/docx/forum-mso-en4-547498.docx
# 625 forums/docx/forum-mso-de-112327.docx
# 616 forums/docx/forum-mso-en-18606.docx

SvxLRSpaceItem::SetLeft
 -> assert added in https://gerrit.libreoffice.org/c/core/+/146643
(noted there)
# 7991 bugtrackers/doc/fdo42762-1.doc
# 7329 bugtrackers/sxw/ooo30127-1.sxw
# 7309 bugtrackers/sxw/ooo49088-1.sxw
# 7203 bugtrackers/ott/tdf93421-1.ott
# 7192 bugtrackers/odt/fdo43370-1.odt
# 7178 bugtrackers/odt/tdf94897-3.odt
# 7172 bugtrackers/odt/tdf93946-1.odt
# 7169 bugtrackers/odt/tdf95222-2.odt
# 7143 bugtrackers/odt/tdf128828-1.odt
# 7127 bugtrackers/odt/tdf95840-1.odt
# 7119 bugtrackers/odt/tdf137241-1.odt
# 7115 bugtrackers/odt/tdf123042-1.odt
# 7114 bugtrackers/odt/tdf94897-2.odt
# 7112 bugtrackers/odt/tdf125373-3.odt
# 7111 bugtrackers/odt/tdf137402-1.odt
# 7108 bugtrackers/odt/tdf93009-1.odt
# 7100 bugtrackers/odt/tdf97806-2.odt
# 7094 bugtrackers/odt/fdo74389-1.odt
# 7075 bugtrackers/odt/fdo77308-1.odt
# 7074 bugtrackers/odt/tdf97491-2.odt
# 7071 bugtrackers/odt/tdf128906-2.odt
# 7039 bugtrackers/odt/tdf128906-1.odt
# 7036 bugtrackers/odt/tdf95648-4.odt
# 7007 bugtrackers/fodt/tdf125373-1.fodt
# 55 forums/odt/forum-it-770.odt
# 15 forums/odt/forum-en-3216.odt
# 120 forums/ott/forum-en-43800.ott
# 110 forums/odt/forum-pl-2563.odt
# 106 forums/odt/forum-en-11103.odt

DocumentFieldsManager::UpdateTableFields
 -> bisect to https://gerrit.libreoffice.org/c/core/+/146611 (which is
a weird outcome)
# 696 forums/docx/forum-mso-en-8069.docx
# 663 forums/docx/forum-mso-en-8072.docx

old
---

ScQueryCellIterator< FindEqualOrSortedLastInRange
 -> assert(!"FindEqualOrSortedLastInRange: rEntry.nField -= nColDiff <
0")
# 796 forums/xls/forum-mso-en4-144513.xls
# 901 forums/xls/forum-mso-en4-433310.xls
# 1034 forums/xls/forum-mso-en4-130543.xls
# 1132 forums/xls/forum-mso-en4-433308.xls
# 1461 forums/xls/forum-mso-en4-433031.xls
# 1601 forums/xls/forum-mso-en4-156881.xls
# 2074 forums/xlsx/forum-mso-en4-491002.xlsx
# 2504 forums/xlsx/forum-mso-en4-527778.xlsx
# 2564 forums/xlsx/forum-mso-en4-606273.xlsx
# 2945 forums/xlsx/forum-mso-en4-643938.xlsx
# 3016 forums/xlsx/forum-mso-en4-544867.xlsx
# 3111 forums/xlsx/forum-mso-en4-629391.xlsx
# 3928 forums/xlsx/forum-mso-en4-353197.xlsx
# 4501 forums/xlsx/forum-mso-en4-544866.xlsx
# 4801 forums/xlsx/forum-mso-en4-606285.xlsx
# 6552 forums/xlsx/forum-mso-en4-606274.xlsx
# 6119 forums/xlsx/forum-mso-en4-70163.xlsx
# 6101 forums/xlsx/forum-mso-en4-183459.xlsx
# 5500 forums/xlsx/forum-mso-en4-112469.xlsx
# 5253 forums/xlsx/forum-mso-en4-606289.xlsx
# 5176 forums/xlsx/forum-mso-de-62049.xlsx
# 5379 forums/xlsx/forum-mso-de-62049.xlsx

ScFormulaCell::MaybeInterpret()
# 891 forums/xls/forum-mso-de-48401.xls
# 2679 forums/xlsx/forum-mso-en4-237685.xlsx
# 2735 forums/xlsx/forum-mso-en4-271721.xlsx
# 3354 forums/xlsx/forum-mso-en4-237668.xlsx
# 3748 forums/xlsx/

Re: Crash test update

[Caolán McNamara]

On Mon, 2023-01-16 at 00:23 +, crashtest wrote:
> 104 files have crashed during import.
> 
> 4 files have crashed during export.

I've grouped the crashes together by shared stacktrace and ranked by
frequency, first column is task id, e.g. 796 is the stacktrace of
https://dev-builds.libreoffice.org/crashtest/d85a6fd228b3eaeb9214e73cdec0c6ef88e058e3/backtraces/task796-core.backtrace.txt
where known gerrit link to commit which seems to be related to the
crash is mentioned.

ScQueryCellIterator< FindEqualOrSortedLastInRange
 -> assert(!"FindEqualOrSortedLastInRange: rEntry.nField -= nColDiff <
0")
796 forums/xls/forum-mso-en4-144513.xls
901 forums/xls/forum-mso-en4-433310.xls
1034 forums/xls/forum-mso-en4-130543.xls
1132 forums/xls/forum-mso-en4-433308.xls
1255 forums/xlsx/forum-mso-de-62049.xlsx
1461 forums/xls/forum-mso-en4-433031.xls
1601 forums/xls/forum-mso-en4-156881.xls
2074 forums/xlsx/forum-mso-en4-491002.xlsx
2504 forums/xlsx/forum-mso-en4-527778.xlsx
2564 forums/xlsx/forum-mso-en4-606273.xlsx
2945 forums/xlsx/forum-mso-en4-643938.xlsx
3016 forums/xlsx/forum-mso-en4-544867.xlsx
3111 forums/xlsx/forum-mso-en4-629391.xlsx
3928 forums/xlsx/forum-mso-en4-353197.xlsx
4501 forums/xlsx/forum-mso-en4-544866.xlsx
4801 forums/xlsx/forum-mso-en4-606285.xlsx
6552 forums/xlsx/forum-mso-en4-606274.xlsx
6119 forums/xlsx/forum-mso-en4-70163.xlsx
6101 forums/xlsx/forum-mso-en4-183459.xlsx
5500 forums/xlsx/forum-mso-en4-112469.xlsx
5253 forums/xlsx/forum-mso-en4-606289.xlsx
5176 forums/xlsx/forum-mso-de-62049.xlsx

ScInterpreter::ScCountIfs
977 forums/xls/forum-mso-en4-253817.xls
1067 forums/xls/forum-mso-en4-102869.xls
1052 forums/xls/forum-mso-en4-254291.xls
1080 forums/xls/forum-mso-en4-102992.xls
1468 forums/xls/forum-mso-en4-76565.xls
1611 forums/xls/forum-mso-en4-253599.xls
2606 forums/xls/forum-mso-en4-495011.xls
2846 forums/xlsx/forum-mso-en4-253503.xlsx
2925 forums/xlsx/forum-mso-en4-433106.xlsx
3097 forums/xlsx/forum-mso-en4-554783.xlsx
3183 forums/xlsx/forum-mso-en4-688373.xlsx
3207 forums/xlsx/forum-mso-en4-606117.xlsx
3344 forums/xlsx/forum-mso-de-103125.xlsx
3660 forums/xlsx/forum-mso-en4-799216.xslx
3798 forums/xlsx/forum-mso-en4-561782.xslx
3853 forums/xlsx/forum-mso-en4-623773.xslx
3876 forums/xlsx/forum-mso-en4-258065.xslx
4001 forums/xlsx/forum-mso-en4-688266.xlsx
4086 forums/xlsx/forum-mso-en4-335212.xlsx
4213 forums/xlsx/forum-mso-en4-614547.xlsx
4471 forums/xlsx/forum-mso-en4-253934.xlsx
4604 forums/xlsx/forum-mso-en4-394112.xlsx

ScFormulaCell::MaybeInterpret()
891 forums/xls/forum-mso-de-48401.xls
2679 forums/xlsx/forum-mso-en4-237685.xlsx
2735 forums/xlsx/forum-mso-en4-271721.xlsx
3354 forums/xlsx/forum-mso-en4-237668.xlsx
3748 forums/xlsx/forum-mso-en4-314789.xslx
6837 forums/xlsx/forum-mso-en4-264966.xlsx
6480 forums/xlsx/forum-mso-en4-264977.xlsx
5577 forums/xlsx/forum-mso-en4-457928.xlsx
5217 forums/xlsx/forum-mso-en4-604336.xlsx
5053 forums/xlsx/forum-mso-en4-784502.xlsx
4935 forums/xlsx/forum-mso-en4-184573.xlsx
4813 forums/xlsx/forum-mso-en4-264966.xlsx

Scinterpreter::IterateParametersIfs
6505 forums/xlsx/forum-mso-en4-495026.xlsx
6486 forums/xlsx/forum-mso-en4-602328.xlsx
6477 forums/xlsx/forum-mso-en4-688411.xlsx
6266 forums/xlsx/forum-mso-en4-487034.xlsx
5641 forums/xlsx/forum-mso-en4-654001.xlsx
5618 forums/xlsx/forum-mso-en4-257571.xlsx
5609 forums/xlsx/forum-mso-en4-688409.xlsx
5446 forums/xlsx/forum-mso-en4-340908.xlsx
5428 forums/xlsx/forum-mso-en4-702161.xlsx
3426 forums/xlsx/forum-mso-en4-702177.xlsx

SwSubFont::GetTextSize_
629 forums/docx/forum-mso-de-118466.docx
641 forums/docx/forum-mso-de-76224.docx
649 forums/docx/forum-mso-de-118517.docx
656 forums/docx/forum-mso-de-118440.docx
664 forums/docx/forum-mso-de-118507.docx
679 forums/docx/forum-mso-de-118414.docx

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192 (noted
there)
672 forums/docx/forum-mso-en-9381.docx
666 forums/docx/forum-mso-en4-137999.docx
7704 bugtrackers/docx/fdo45193-1.docx
7671 bugtrackers/docx/fdo45195-1.docx

ScTokenArray::ImplGetReference from ReadQsi
889 forums/forum-mso-en4-69844.xls
1338 forums/forum-mso-en4-69589.xls
1788 forums/forum-mso-en4-69308.xls

ScFilterDescriptorBase::setFilterFields
6464 forums/xlsx/forum-mso-en4-334072.xlsx
5722 forums/xlsx/forum-mso-en4-100882.xlsx

ScCountIfCellIterator<::GetCount()
790 forums/xls/forum-mso-en4-75791.xls
835 forums/xls/forum-mso-en4-601987.xls

SwTable::ConvertSubtableBox
 -> bisected to https://gerrit.libreoffice.org/c/core/+/141079
1842 bugtrackers/sxw/ooo27428-1.sxw
1620 bugtrackers/odt/ooo98643-1.odt

XclExpFmlaCompImpl::RecalcTokenClass
2849 forums/xlsx/forum-mso-en4-253503.xlsx
4613 forums/xlsx/forum-mso-en4-568971.xlsx

sw::CalcBreaks
653 forums/docx/forum-mso-en-4208.docx
   -> https://gerrit.libreoffice.org/c/core/+/127015 
   -> https://gerrit.libreoffice.org/c/core/+/127127
I suspect the SwSubFont::GetTextSize_ are similar
   -> https://bugs.documentfoundation.org/show_bug.cgi?id=152038

sw::m

Re: Crash test update

[Caolán McNamara]

On Sun, 2022-11-27 at 18:14 +, Dr. David Alan Gilbert wrote:
> * crashtest (nore...@documentfoundation.org) wrote:
> > Hi,
> > 
> > New crashtest update available at
> >  > bef08e300afc8b4aaf2/>.
> 
> Are there any docs on this script?

The script itself is test-bugzilla-files in
https://git.libreoffice.org/dev-tools

The build is configured with --enable-dbgutil --enable-optimized=debug
so asserts are enabled, and its mostly asserts that comprise the
failures

The test docs are of two main categories, a) documents from bugzillas
where the scheme is basic enough, ooo27428-1 is just attachment #1 from
bug #27428 at https://bz.apache.org/ooo/ and so on. b) documents from
various forums which are unfortunately less easily linked to as they
tend to require some login information.

import failures typically can be reproduced with
./instdir/program/soffice --headless --convert-to pdf input.whatever
export failures typically with, e.g.
./instdir/program/soffice --headless --convert-to docx input.odt

There are a bunch of long outstanding issues, I've some more detail on
those at
https://lists.freedesktop.org/archives/libreoffice/2022-October/089536.html

Re: Crash test update

[Dr. David Alan Gilbert]

* crashtest (nore...@documentfoundation.org) wrote:
> Hi,
> 
> New crashtest update available at 
> .

Are there any docs on this script? It would be nice if 
this mail included a pointer to:
  a) The script
  b) The test documents (some are easier to find than others)
  c) Some pointers to how to debug, if those pointers exist.

I was looking through the backtraces/ dir and other than
looking through the backtrace for the name of the doc file
it's not obvious how to associate them.

Is this test run with --enable-debug builds? If so it would
be nice to capture diag from failed runs.

(Out of the 3 I've tried so far, only one fails for me locally,
the import from ooo 84576/laisvos_programos-vaiku_lavinimui.odt
dies from command line, gives a pretty error from the gui, and either
way gives 'MarkManager::makeMark(..) - refusing to create mark on non-textnode' 
)

Dave

> 38 files have crashed during import.
> 
> 16 files have crashed during export.
> 
> 0 files have differences compared to the baseline, which is core.git commit .
> (17764 files are excluded.)
> 
> - Your friendly crashtest bot





-- 
 -Open up your eyes, open up your mind, open up your code ---   
/ Dr. David Alan Gilbert|   Running GNU/Linux   | Happy  \ 
\dave @ treblig.org |   | In Hex /
 \ _|_ http://www.treblig.org   |___/

Re: Crash test update

[Caolán McNamara]

On Sat, 2022-10-29 at 09:39 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  4053da0b39ca69c86/
> 
> 22 files have crashed during import.

This breaks down to 13 different backtraces:

SwTable::ConvertSubtableBox
 -> bisected to https://gerrit.libreoffice.org/c/core/+/141079
1842 bugtrackers/sxw/ooo27428-1.sxw
1620 bugtrackers/odt/ooo98643-1.odt
53 forums/odt/forum-de2-6979.odt

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192
2226 bugtrackers/docx/fdo45193-1.docx
2193 bugtrackers/docx/fdo45195-1.docx
648 forums/docx/forum-mso-en-9381.docx

SwSubFont::GetTextSize_
615 forums/docx/forum-mso-de-118466.docx
625 forums/docx/forum-mso-de-76224.docx
630 forums/docx/forum-mso-de-118517.docx
636 forums/docx/forum-mso-de-118440.docx
653 forums/docx/forum-mso-de-118414.docx

ScQueryCellIterator
1255 forums/xlsx/forum-mso-de-62049.xlsx
1071 forums/xlsx/forum-mso-de-103125.xlsx

SwObjectFormatterTextFrame::GetFirstObjWithMovedFwdAnchor
2238 bugtrackers/docx/fdo49260-4.docx

SwXTextRange::Impl::SetMark
1630 bugtrackers/odt/ooo84576-1.odt

SwXTextField::setPropertyValue
66 forums/odt/forum-en-7529.odt

sw::mark::CrossRefBookmark::CrossRefBookmark
89 forums/odt/forum-en-12562.odt

ScPatternAttr::GetItem
331 forums/ods/forum-en-4598.ods

sw::CalcBreaks
634 forums/docx/forum-mso-en-4208.docx

DocumentFieldsManager::FieldsToCalc
641 forums/docx/forum-mso-en-8072.docx

sw::mark::FindFieldSep
673 forums/docx/forum-mso-en-12641.docx

ScFormulaCell::MaybeInterpret()
717 forums/xls/forum-mso-de-48401.xls

Re: Crash test update

[Caolán McNamara]

On Thu, 2022-08-25 at 19:36 +, crashtest wrote:
> 104 files have crashed during import.
> 
> 69 files have crashed during export.

I've broken down these crashes by crash signature and bisected most of
the the ones that refer to bugtracker documents which (unlike the
relatively recent forum ones) didn't assert/crash in the past. 

SwXMLExport::ExportTable
 -> assert from https://gerrit.libreoffice.org/c/core/+/127548 (noted
there)
bugtrackers/docx/fdo46822-1.odt
bugtrackers/docx/tdf122628-1.odt
bugtrackers/docx/tdf131574-2.odt
bugtrackers/docx/tdf122628-7.docx
bugtrackers/docx/rhbz861680-10.odt
bugtrackers/odt/fdo50072-1.odt
bugtrackers/odt/tdf92502-3.odt
bugtrackers/odt/ooo40536-2.odt
bugtrackers/odt/tdf92502-1.odt
bugtrackers/odt/ooo40518-2.odt
forums/doc/forum-nl-204.odt
forums/docx/forum-mso-en-6129.odt
forums/docx/forum-mso-de-124930.odt
forums/docx/forum-mso-de-124930.odt
forums/docx/forum-mso-de-75497.odt
forums/odt/forum-en-1926.odt

SwSortedObjs::Insert
 -> bisected to https://gerrit.libreoffice.org/c/core/+/136192 (noted
there)
bugtrackers/docx/fdo76513-1.docx
bugtrackers/docx/fdo45193-1.docx
bugtrackers/docx/fdo45195-1.docx
forums/docx/forum-mso-en-9381.docx

SwContentIndexReg::~SwContentIndexReg
 -> bisected to https://gerrit.libreoffice.org/c/core/+/138316 (noted
there)
bugtrackers/docx/ooo117844-1.odt
forums/docx/forum-mso-en-13192.odt
files/cloudon/File_1405.odt

XMLTextImportHelper::hasCurrentFieldSeparator()
 -> assert from https://gerrit.libreoffice.org/c/core/+/136513 (noted
there)
bugtrackers/odt/fdo34997-1.odt
forums/odt/forum-de2-1306.odt

ScQueryCellIterator
forums/xlsx/forum-mso-de-62049.xlsx
forums/xlsx/forum-mso-de-103125.xlsx

SwObjectFormatterTextFrame::GetFirstObjWithMovedFwdAnchor
bugtrackers/docx/fdo49260-4.docx

SwSubFont::GetTextSize_
forums/docx/forum-mso-de-118414.docx
forums/docx/forum-mso-de-118414.docx
forums/docx/forum-mso-de-118440.docx
forums/docx/forum-mso-de-118517.docx
forums/docx/forum-mso-de-76224.docx
forums/docx/forum-mso-de-118466.docx

ScFormulaCell::Interpret
bugtrackers/sxc/gnome334723-1.sxc

dbaccess::DocumentEventNotifier_Impl::processEvent
bugtrackers/odb/tdf105071-2.odb

sw::mark::FindFieldSep
forums/docx/forum-mso-en-12641.docx
forums/docx/forum-mso-en-2328.docx

SfxWhichIter::GetItemState
forums/docx/forum-mso-en-102589.docx

sax_fastparser::lcl_DebugMergeAppend
forums/docx/forum-mso-en-13436.docx
forums/docx/forum-mso-en-16671.docx
forums/docx/forum-mso-de-101258.docx

sax_fastparser::FastSaxSerializer::endFastElement
forums/docx/forum-mso-de-102317.docx
forums/docx/forum-mso-en-11199.docx

ScDocument::GetFormatTable
forums/ods/forum-de3-3108.ods
forums/ods/forum-de3-3100.ods

Scheduler::ImplStartTimer
forums/ods/forum-fr-40118.ods

ScPatternAttr::GetItem
forums/ods/forum-en-4598.ods

ScDPItemData::GetType()
forums/ods/forum-en-38962.ods

DocumentFieldsManager::FieldsToCalc
forums/docx/forum-mso-en-8069.docx
forums/docx/forum-mso-de-118507.docx
forums/docx/forum-mso-en-8072.docx

SwXTextField::setPropertyValue
forums/odt/forum-en-7529.odt

sw::mark::CrossRefBookmark::CrossRefBookmark
forums/odt/forum-en-12562.odt

ScFormulaCell::MaybeInterpret()
forums/xls/forum-mso-de-48401.xls

void checkGlyphsEqual
bugtrackers/odt/tdf119074-1.odt

--- fixed ---

ImplLogicToPixel
 -> fixed with https://gerrit.libreoffice.org/c/core/+/138871
bugtrackers/odg/tdf35913-4.odg
bugtrackers/svg/kde82509-1.odg
bugtrackers/svg/fdo45001-3.odg

_cairo_ft_scaled_glyph_init_surface
 -> fixed with https://gerrit.libreoffice.org/c/core/+/138876
bugtrackers/docx/fdo66401-6.odt
bugtrackers/docx/tdf117501-3.odt

FreeTypeTextRenderImpl::GetTextLayout
 -> fixed with https://gerrit.libreoffice.org/c/core/+/138878
bugtrackers/doc/ooo60911-3.odt

SwXTextRange::Impl::SetMark
 -> fixed with https://gerrit.libreoffice.org/c/core/+/138434
bugtrackers/odt/ooo84576-1.odt

Re: Crash test update, "validation?"

[Thorsten Behrens]

Hi Miklos, hi Caolán,

Miklos Vajna wrote:
> On Mon, Aug 15, 2022 at 03:08:17PM +0100, Caolán McNamara wrote:
> > Do we care about the "validationErrors", I've never looked at them and
> > maybe the turnaround time might be faster if we drop validation?
> 
> I think it's useful information, but looking at the cost of generating
> them with that java-based tool and the benefit (vs crashtest data,
> user-supplied bugreports, etc), perhaps it's an overkill to run these
> all the time.
> 
That was added IIRC quite a few years ago to make sure we're not
inadvertently breaking filters - so I'd agree it's useful to have. If
we're not acting on it currently, then there's no harm running it less
frequently (or perhaps it should be part of the (ESC) reporting, so it
gets more attention?)

Cheers,

-- Thorsten


signature.asc
Description: PGP signature

Re: Crash test update, "validation?"

[Miklos Vajna]

Hi Caolan,

On Mon, Aug 15, 2022 at 03:08:17PM +0100, Caolán McNamara wrote:
> Do we care about the "validationErrors", I've never looked at them and
> maybe the turnaround time might be faster if we drop validation?

I think it's useful information, but looking at the cost of generating
them with that java-based tool and the benefit (vs crashtest data,
user-supplied bugreports, etc), perhaps it's an overkill to run these
all the time.

It may be enough to keep the ability to run them much less frequently
(e.g. once after each libreoffice-x-y branch-off or something.)

Regards,

Miklos

Re: Crash test update, "validation?"

[Caolán McNamara]

On Thu, 2022-08-11 at 14:35 +, crashtest wrote:
> - Your friendly crashtest bot

Do we care about the "validationErrors", I've never looked at them and
maybe the turnaround time might be faster if we drop validation?

Re: Crash test update

[Xisco Fauli]

Hello,

On 22/6/22 14:10, crashtest wrote:

Hi,

New crashtest update available at 
.

121 files have crashed during import.


In order to help a bit with the investigation of these import crashes, I 
reported and bisected those that are reproducible with the bisect 
builds. These are the reports:


- https://bugs.documentfoundation.org/show_bug.cgi?id=149679
    - forum-de3-9186.ods
    - forum-en-29634.ods
- https://bugs.documentfoundation.org/show_bug.cgi?id=149546
    - forum-mso-en-11679.docx
- https://bugs.documentfoundation.org/show_bug.cgi?id=149691
    - forum-mso-de-55780.docx
- https://bugs.documentfoundation.org/show_bug.cgi?id=149649
    - forum-mso-en-16226.docx
    - forum-mso-en-16228.docx
- https://bugs.documentfoundation.org/show_bug.cgi?id=149692
    - forum-mso-en-14071.docx
- https://bugs.documentfoundation.org/show_bug.cgi?id=149548

    - forum-mso-en-13192.docx

Regards

--
Xisco Faulí
LibreOffice QA Team
IRC: x1sc0

Re: Crash test update

[Caolán McNamara]

On Mon, 2022-06-13 at 20:58 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  266732509f83c52eb/>.
> 
> 148 files have crashed during import.
> 62 files have crashed during export.

FWIW this run includes the documents from xisco's new bin/get-forum-
attachments.py and mostly indicates newly detected failures rather than
recent regressions.

Re: Crash test update

[Caolán McNamara]

On Fri, 2022-06-03 at 19:35 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  d15c5ceea70aee7fb/>.
> 
> 4 files have crashed during import.
> 2 files have crashed during export.

Progress!, 4 bts give
1 SwObjectFormatterTextFrame::GetFirstObjWithMovedFwdAnchor
2 two FastSaxSerializer::endFastElement asserts
1 mark::FindFieldSep

Something somewhere along the line appears to have wiped out the
ModelToViewHelper::ModelToViewHelper crash seen frequently in the past
like
https://dev-builds.libreoffice.org/crashtest/b4bf19321ed9ce2d0e49341193690fef0c7085ea/backtraces/task1137-core.backtrace.txt

Re: Crash test update

[Noel Grandin]

On Thu, 2 Jun 2022 at 12:53, Caolán McNamara  wrote:

> On Tue, 2022-05-31 at 13:35 +, crashtest wrote:
> > Hi,
> >
> > New crashtest update available at
> >  > 3106cb4fedb4a2ad2/>.
> >
> > 37 files have crashed during import.
>
> seems that the frequent assert I see in here was triggered by
>
> commit c390c2e3f44953e236389467c7c090cee9ebc1b7
> Author: Noel Grandin 
> Date:   Thu May 26 12:16:35 2022 +0200
>
> mark SvxFontItem as IsSortable
>
>
reverting that with
https://gerrit.libreoffice.org/c/core/+/135308

Re: Crash test update

[Caolán McNamara]

On Tue, 2022-05-31 at 13:35 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  3106cb4fedb4a2ad2/>.
> 
> 37 files have crashed during import.

seems that the frequent assert I see in here was triggered by

commit c390c2e3f44953e236389467c7c090cee9ebc1b7
Author: Noel Grandin 
Date:   Thu May 26 12:16:35 2022 +0200

mark SvxFontItem as IsSortable

which I reproduce with e.g.
./instdir/program/soffice --headless --convert-to pdf ooo30162-3.xls

Re: Crash test update

[Luboš Luňák]

On Wednesday 25 of May 2022, Caolán McNamara wrote:
> Glyph Cache issues seem to be all resolved now.

 Good :). I've pushed https://gerrit.libreoffice.org/c/core/+/134659 so that 
may change again.

> The ScCellIterator crash of
> https://dev-builds.libreoffice.org/crashtest/38751a8d5e07a448fe56c4602701f5
>67538b139e/backtraces/task1206-core.backtrace.txt looks relatively new and
> maybe a fairly low hanging problem

 https://gerrit.libreoffice.org/c/core/+/134932

-- 
 Luboš Luňák
 l.lu...@collabora.com

Re: Crash test update

[Caolán McNamara]

On Wed, 2022-05-25 at 01:23 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
>  02701f567538b139e/>.
> 
> 13 files have crashed during import.
> 3 files have crashed during export.

Glyph Cache issues seem to be all resolved now. oss-fuzz had found a
crasher in that which has also gone away so that all looks sorted out.

The ScCellIterator crash of
https://dev-builds.libreoffice.org/crashtest/38751a8d5e07a448fe56c4602701f567538b139e/backtraces/task1206-core.backtrace.txt
looks relatively new and maybe a fairly low hanging problem

Re: Crash test update

[Luboš Luňák]

On Tuesday 10 of May 2022, Caolán McNamara wrote:
> On Tue, 2022-05-10 at 15:48 +, crashtest wrote:
> > Hi,
> >
> > New crashtest update available at
> > https://dev-builds.libreoffice.org/crashtest/d4ec15d5b7c64cd62b6f729c
> > 1334c167da5886b4/
> >
> > 91 files have crashed during import.
>
> crashtesting has returned thanks to Guilhem and the gang. Sampling
> backtraces gives pretty much the same assert/crash every time, which
> looks the same SalLayoutGlyphsCache::GetLayoutGlyphs related one seen
> in tdf#148954

 I have fixed tdf#148954, so let's see how the next crashtesting run fares.

-- 
 Luboš Luňák
 l.lu...@collabora.com

Re: Crash test update

[Caolán McNamara]

On Tue, 2022-05-10 at 15:48 +, crashtest wrote:
> Hi,
> 
> New crashtest update available at
> https://dev-builds.libreoffice.org/crashtest/d4ec15d5b7c64cd62b6f729c
> 1334c167da5886b4/
> 
> 91 files have crashed during import.

crashtesting has returned thanks to Guilhem and the gang. Sampling
backtraces gives pretty much the same assert/crash every time, which
looks the same SalLayoutGlyphsCache::GetLayoutGlyphs related one seen
in tdf#148954

Re: Crash test update

[Miklos Vajna]

Hi,

On Mon, Jan 31, 2022 at 02:32:55PM +, Caolán McNamara  
wrote:
> I think I may have identified where this began to happen, with
> 
> commit 493a916a3113e877835c9bc7c93faef0d29f9a33
> Date:   Mon Feb 22 14:28:59 2021 +0100
> 
> tdf#140158 tdf#138598 tdf#140598 sw: fix sync of AS_CHAR textboxes
> 
> the porfly.cxx change of that commit. I find that if I disable the line
> calling SetFormatAttr(aNewTxBxAnchor) then the assert doesn't fire.
> 
> I presume something goes out of sync when setting an anchor to a
> different SwTextNode than it was originally and this leads to the later
> assert. What the right solution is though it totally opaque to me.

I think even if the anchor type is as-char, the intention is to maintain
that the draw format (shape) and its fly format (text frame) has the
same anchor position. We can't have as-char as the anchor type for both,
but normally we import the shape with as-char and a certain anchor
position, then we can have the fly format with to-char and the same
anchor position.

I can't think of a situation off the top of my head where it's correct
that the two formats would have SwFormatAnchors that point to a
different SwPosition.

Alco CC Laszlo who reviewed the above commit.

Regards,

Miklos

Re: Crash test update

[Caolán McNamara]

On Sun, 2022-01-30 at 12:39 +, crashtest wrote:
> New crashtest update available

There are a number of duplicate asserts that have lingered for a while,
for example
https://dev-builds.libreoffice.org/crashtest/59dc59bb6e50d6d42d9fd6b82a165069d6e4bdbf/backtraces/task841-core.backtrace.txt

I was finally able to reproduce one by taking the document of
https://bugs.documentfoundation.org/attachment.cgi?id=163550
interactively loading it, and saving as odt and then wait a few seconds
and the assert fires.

I think I may have identified where this began to happen, with

commit 493a916a3113e877835c9bc7c93faef0d29f9a33
Date:   Mon Feb 22 14:28:59 2021 +0100

tdf#140158 tdf#138598 tdf#140598 sw: fix sync of AS_CHAR textboxes

the porfly.cxx change of that commit. I find that if I disable the line
calling SetFormatAttr(aNewTxBxAnchor) then the assert doesn't fire.

I presume something goes out of sync when setting an anchor to a
different SwTextNode than it was originally and this leads to the later
assert. What the right solution is though it totally opaque to me.

Re: Crash test update

[Miklos Vajna]

Hi,

On Sat, Oct 02, 2021 at 09:34:20PM +, crashtest 
 wrote:
> 11841 files have differences compared to the previous run. (17764 files are 
> excluded.)

I assume 05f863844d9a5613250e8d787e32752b270ec4d3 "sw: paragraph styles:
add ODT filter for a linked character style" is the reason for the quite
large first number, now the ODT export retains more info from DOCX input
files, so DOCX->ODT output differs vs the previous run, but that is all
wanted.

See
https://dev-builds.libreoffice.org/crashtest/ea8e41710045a1e20f6cc1d210f3534c4f5fe9c9/odfundiff/docx/ooo124967-1.odt.log.odfundiff
for one such sample.

Regards,

Miklos

Re: Crash test update

[Caolán McNamara]

On Mon, 2021-06-21 at 23:03 +, crashtest wrote:
> New crashtest update available at  
> https://dev-builds.libreoffice.org/crashtest/89459662bf2684a07596d4132c84e5da7e0af8d4/

This run contained the commit

commit 10dbee56f7cf9a19b04f484b8a4b2465323ec622
Author: Justin Luth 
Date:   Tue Apr 20 15:31:23 2021 +0200

tdf#106541 doc import: always register listLevel

with the assert of: "m_nLFOPosition is usually reset immediately..."
active, but there was no appearance as far as I could see in any of the
failures for whatever that is worth

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

Re: Crash test update

[Caolán McNamara]

On Sat, 2021-06-12 at 10:18 +, crashtest wrote:
> New crashtest update available at  
> https://dev-builds.libreoffice.org/crashtest/b682c8c62b2bd6bd66b22c872a2b65b67892c1fa/

I see an assert with

soffice --headless --convert-to pdf fdo43054-1.odp
with a "TextColumns" property in

https://dev-builds.libreoffice.org/crashtest/b682c8c62b2bd6bd66b22c872a2b65b67892c1fa/backtraces/task908-core.backtrace.txt

which I'm guessing is triggered by something related to the recent
editengine-columns work

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

Re: Crash test update

[Luboš Luňák]

On Sunday 18 of April 2021, Caolán McNamara wrote:
> On Sun, 2021-04-18 at 09:05 +, crashtest wrote:
> > New crashtest update available at
> > https://dev-builds.libreoffice.org/crashtest/988e2827319fa2fe04bb818880c1
> >f381bd08a314/
>
> There's a huge number of new import and export failures, mostly in
> presentations. FWIW using tdf116803-1.odp as example it bisects to:
>
> commit b94a2dc95dce8c67ddb9f01f7bd91da2a5759d9c
> Date:   Thu Apr 15 17:39:48 2021 +0200
>
> threaded loading also for png loader

 https://gerrit.libreoffice.org/c/core/+/114262 should take care of that.

-- 
 Luboš Luňák
 l.lu...@collabora.com
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Sun, 2021-04-18 at 09:05 +, crashtest wrote:
> New crashtest update available at  
> https://dev-builds.libreoffice.org/crashtest/988e2827319fa2fe04bb818880c1f381bd08a314/

There's a huge number of new import and export failures, mostly in
presentations. FWIW using tdf116803-1.odp as example it bisects to:

commit b94a2dc95dce8c67ddb9f01f7bd91da2a5759d9c
Date:   Thu Apr 15 17:39:48 2021 +0200

threaded loading also for png loader


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

Re: Crash test update

[Eike Rathke]

Hi,

On Monday, 2021-03-29 15:14:46 +0100, Caolán McNamara wrote:

> One of these is an apparent threaded calc related assert/crash
> https://dev-builds.libreoffice.org/crashtest/f96004096268f5e71120678e32fc8c74055819aa/backtraces/task714-core.backtrace.txt

Fixed with
https://git.libreoffice.org/core/+/465b8b0e9ad4b0c9c7701dee2820a99c5d00b5bf%5E%21/

  Eike

-- 
GPG key 0x6A6CD5B765632D3A - 2265 D7F3 A7B0 95CC 3918  630B 6A6C D5B7 6563 2D3A


signature.asc
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Eike Rathke]

Hi,

On Monday, 2021-03-29 18:14:54 +0200, Luboš Luňák wrote:

> On Monday 29 of March 2021, Caolán McNamara wrote:
> > wget https://bz.apache.org/ooo/attachment.cgi?id=71756 \
> >   -O ooo114587-1.ods
> > ./instdir/program/soffice.bin ./ooo114587-1.ods
> 
>  This appears to be a regression from 
> https://cgit.freedesktop.org/libreoffice/core/commit/?id=042dbf83122b14fd1dd32705c8f8b7d65c22f21b
>  .

That is because a cell result of a (or in this case THE) currently
running interpreted cell is to be pushed (which could happen in several
other constellations as well and actually should result in circular
reference error if iterations are not activated) and
ScFormulaCell::MaybeInterpret() has

  assert(!rDocument.IsThreadedGroupCalcInProgress());

instead of letting the following Interpret() call decide what to do
which it was designed for before all the group calc things were shoved
before the if(bRunning) block so that doesn't work anymore and the
assert() was introduced..

I'll see if I can solve it somehow.

  Eike

-- 
GPG key 0x6A6CD5B765632D3A - 2265 D7F3 A7B0 95CC 3918  630B 6A6C D5B7 6563 2D3A


signature.asc
Description: PGP signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Luboš Luňák]

On Monday 29 of March 2021, Caolán McNamara wrote:
> On Mon, 2021-03-29 at 06:06 +, crashtest wrote:
> > New crashtest update available
>
> One of these is an apparent threaded calc related assert/crash
> https://dev-builds.libreoffice.org/crashtest/f96004096268f5e71120678e32fc8c
>74055819aa/backtraces/task714-core.backtrace.txt the only non-writer
> failure, reproducible for me with...
>
> wget https://bz.apache.org/ooo/attachment.cgi?id=71756 \
>   -O ooo114587-1.ods
> ./instdir/program/soffice.bin ./ooo114587-1.ods

 This appears to be a regression from 
https://cgit.freedesktop.org/libreoffice/core/commit/?id=042dbf83122b14fd1dd32705c8f8b7d65c22f21b
 .


-- 
 Luboš Luňák
 l.lu...@collabora.com
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Mon, 2021-03-29 at 06:06 +, crashtest wrote:
> New crashtest update available

One of these is an apparent threaded calc related assert/crash
https://dev-builds.libreoffice.org/crashtest/f96004096268f5e71120678e32fc8c74055819aa/backtraces/task714-core.backtrace.txt
the only non-writer failure, reproducible for me with...

wget https://bz.apache.org/ooo/attachment.cgi?id=71756 \
  -O ooo114587-1.ods
./instdir/program/soffice.bin ./ooo114587-1.ods 


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

Re: Crash test update: VectorGraphicData::ensureSequenceAndRange

[Caolán McNamara]

On Thu, 2021-02-18 at 16:44 +0900, Tomaž Vajngerl wrote:
> On Sun, Feb 14, 2021 at 6:21 AM Caolán McNamara  
> > reproducible for me:
> > $ ./instdir/program/soffice.bin --headless --convert-to 
> > odg:output.odg
> > ./tdf118346-1.odg
> > 
> > crashes in VectorGraphicData::ensureSequenceAndRange in some odd way
> > during assigning to maSequence
> > 
> Sorry, I missed this mail. I'll take a look.

What I seem to see is recursive reduceGraphicMemory calls and
https://gerrit.libreoffice.org/c/core/+/112772 works for me on this
example document

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

Re: Crash test update: VectorGraphicData::ensureSequenceAndRange

[Tomaž Vajngerl]

Hi Caolan,

On Sun, Feb 14, 2021 at 6:21 AM Caolán McNamara  wrote:

> On Sat, 2021-02-13 at 09:19 +, crashtest wrote:
> > New crashtest update
>
> one of these crashes has relatively recently appeared and is
> reproducible for me:
>
> https://dev-builds.libreoffice.org/crashtest/beeed387732f95501b06b5462d450517422b18ba/backtraces/task1183-core.backtrace.txt
>
> $ wget https://bugs.documentfoundation.org/attachment.cgi?id=145182 -O
> tdf118346-1.odg
> $ ./instdir/program/soffice.bin --headless --convert-to odg:output.odg
> ./tdf118346-1.odg
>
> crashes in VectorGraphicData::ensureSequenceAndRange in some odd way
> during assigning to maSequence
>

Sorry, I missed this mail. I'll take a look.

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

Re: Crash test update: VectorGraphicData::ensureSequenceAndRange

[Caolán McNamara]

On Sat, 2021-02-13 at 09:19 +, crashtest wrote:
> New crashtest update

one of these crashes has relatively recently appeared and is
reproducible for me:
https://dev-builds.libreoffice.org/crashtest/beeed387732f95501b06b5462d450517422b18ba/backtraces/task1183-core.backtrace.txt

$ wget https://bugs.documentfoundation.org/attachment.cgi?id=145182 -O 
tdf118346-1.odg
$ ./instdir/program/soffice.bin --headless --convert-to odg:output.odg
./tdf118346-1.odg

crashes in VectorGraphicData::ensureSequenceAndRange in some odd way
during assigning to maSequence

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

Re: Crash test update

[Caolán McNamara]

On Wed, 2020-12-02 at 20:46 +, crashtest wrote:
> New crashtest update available at  
> https://dev-builds.libreoffice.org/crashtest/f7e2f5bca938aa916830e07895d2da6820282055/

FWIW the reoccurring SwLayoutCache::CompareLayout assert at
sw/source/core/layout/laycache.cxx:352

https://dev-
builds.libreoffice.org/crashtest/f7e2f5bca938aa916830e07895d2da68202820
55/backtraces/task141-core.backtrace.txt

can be reproduced with e.g. fdo33533-1.odt and bisects to

commit c0864f26f3143ea81c65d3826fae32a8fd54c531
Date:   Fri Nov 6 21:47:21 2020 +0100

sw_fieldmarkhide: init fieldmark mode from options

but I haven't a notion what "!pFrame->GetMergedPara()" might mean


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

Re: Crash test update

[Noel Grandin]

fix in progress here
https://gerrit.libreoffice.org/c/core/+/104975
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Fri, 2020-10-23 at 12:22 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/3907368ded845ac41926013b3d25550a1266f8b0/

FWIW there's a bunch of .docbook import asserts in here that seem to
have started at

commit 2b946d245eaf4bd40a0091aa5508315fc37c81a0
Date:   Mon Oct 19 09:36:04 2020 +0200

XmlFilterAdaptor: use the fastparser API when possible

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

Re: Crash test update

[Caolán McNamara]

On Sun, 2014-08-10 at 08:27 +, crashtest.libreoff...@gmail.com
wrote:
> New crashtest update available at 
> www.dev-builds.libreoffice.org/crashtest/afbbc4cf4ceea9445cbb584b50d511c512e5133f

FWIW

soffice.bin --headless --convert-to odt ooo12626-10.odt

is failing since

commit e366c928819c44b5c253c45dca6dae40b71c9808
Date:   Thu Oct 1 17:31:21 2020 +0200

sw: ODF import: convert the simplest sub-tables to rowspan tables

with this bt:

https://dev-builds.libreoffice.org/crashtest/ad054256a4d914e57d380539b3aed245890064af/backtraces/task129-core.backtrace.txt

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

Re: Crash test update

[Jan-Marek Glogowski]

Am 19.09.20 um 10:49 schrieb crashtest:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/cb15d181b6446003ddc5a13eff9dfb14fc5a8179/

Fix for ParseCMAP crash: https://gerrit.libreoffice.org/c/core/+/103033
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Noel Grandin]

On 2020/09/09 10:11 am, Caolán McNamara wrote:


That worked, there is just one xml related assert left now. One with
importing ooo97840-2.odt where there appears to be namespace-less "id"
attributes triggering an assert about that



Pushed fix in 920608cc35c3cd86eb6458c7528dda2b2b744a2c
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Wed, 2020-04-08 at 19:18 +0200, Noel Grandin wrote:
> Assuming that the SVG format announces it's number range somewhere,
> we could check if the number range is stupid big and scale all the
> numbers down into a sensible range.

For the record I sidestepped the issue with
https://cgit.freedesktop.org/libreoffice/core/commit/?id=0f02412d6baf9edb54c62856c52cebfa707291b2
where using the B2dDraw drawing paths avoids the problematic
code paths.

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

Re: Crash test update

[Caolán McNamara]

On Tue, 2020-09-08 at 09:08 +0200, Noel Grandin wrote:
> 
> On 2020/09/07 11:00 am, Caolán McNamara wrote:
> > On Thu, 2020-09-03 at 15:34 +0200, Noel Grandin wrote:
> > > fixed my crasher with commit
> > > 12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b
> > 
> > which worked fine, but ooo53770-1.odt now asserts about : in an
> > attribute name and the input content.xml does (presumably
> > incorrectly)
> > have style:style:use-optional-row-height in it
> > 
> 
> fixed that assert with 9208eab706bfffec306d3611080b86b7bdd27567

That worked, there is just one xml related assert left now. One with
importing ooo97840-2.odt where there appears to be namespace-less "id"
attributes triggering an assert about that

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

Re: Crash test update

[Noel Grandin]

On 2020/09/07 11:00 am, Caolán McNamara wrote:

On Thu, 2020-09-03 at 15:34 +0200, Noel Grandin wrote:

fixed my crasher with commit 12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b


which worked fine, but ooo53770-1.odt now asserts about : in an
attribute name and the input content.xml does (presumably incorrectly)
have style:style:use-optional-row-height in it



fixed that assert with 9208eab706bfffec306d3611080b86b7bdd27567
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Mon, 2020-09-07 at 12:00 +0200, Luboš Luňák wrote:
> On Friday 04 of September 2020, Caolán McNamara wrote:
> > There is one import crasher that turns up fairly consistently that
> > looks like a possible threaded calculation problem
> > 
> > bt:
> > https://dev-builds.libreoffice.org/crashtest/12a7a3d57d3dcf222b13fa9143c377
> > 36f8ea3d0b/backtraces/task565-core.backtrace.txt
> > 
> > source document:
> > https://bugs.documentfoundation.org/attachment.cgi?id=51878
> 
>  https://gerrit.libreoffice.org/c/core/+/102156 handles that.
> 
> > On manually opening it I can't reproduce locally,
> 
>  Calc crashers often need Ctrl+Shift+F9 to force a full recalc.

ah yes, that does make ScFormulaCell::MaybeInterpret assert on
!pDocument->IsThreadedGroupCalcInProgress()

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

Re: Crash test update

[Luboš Luňák]

On Friday 04 of September 2020, Caolán McNamara wrote:
> There is one import crasher that turns up fairly consistently that
> looks like a possible threaded calculation problem
>
> bt:
> https://dev-builds.libreoffice.org/crashtest/12a7a3d57d3dcf222b13fa9143c377
>36f8ea3d0b/backtraces/task565-core.backtrace.txt
>
> source document:
> https://bugs.documentfoundation.org/attachment.cgi?id=51878

 https://gerrit.libreoffice.org/c/core/+/102156 handles that.

> On manually opening it I can't reproduce locally,

 Calc crashers often need Ctrl+Shift+F9 to force a full recalc.

-- 
 Luboš Luňák
 l.lu...@collabora.com
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Thu, 2020-09-03 at 15:34 +0200, Noel Grandin wrote:
> fixed my crasher with commit 12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b

which worked fine, but ooo53770-1.odt now asserts about : in an
attribute name and the input content.xml does (presumably incorrectly)
have style:style:use-optional-row-height in it

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

Re: Crash test update

[Stephan Bergmann]

On 04/09/2020 06:23, crashtest wrote:

New crashtest update available at 
https://dev-builds.libreoffice.org/crashtest/12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b/


The `--convert-to odt ooo43662-1.doc` assert should be fixed with 
 
"Make OUString(char16_t const[N]) ctor check for embedded NULs".


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

Re: Crash test update

[Caolán McNamara]

On Fri, 2020-09-04 at 04:23 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b/

There is one import crasher that turns up fairly consistently that
looks like a possible threaded calculation problem

bt:
https://dev-builds.libreoffice.org/crashtest/12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b/backtraces/task565-core.backtrace.txt

source document:
https://bugs.documentfoundation.org/attachment.cgi?id=51878

On manually opening it I can't reproduce locally, but there is a
warning that "the maximum number of columns per sheet was exceeded"

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

Re: Crash test update

[Noel Grandin]

fixed my crasher with commit 12a7a3d57d3dcf222b13fa9143c37736f8ea3d0b
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Noel Grandin]

At least one of those crashes smells like it has something to do with my
fastparser work - looks like custom XML attributes are getting wrong
attribute names i.e. getting the wrong prefix

I'll look into that on Monday.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Wed, 2020-08-26 at 05:57 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/03e28651142af2ffa7a3749b879f31bde98a2173/

commit c123bfff501229f398a1b679fc7434b82d53685c (HEAD, refs/bisect/bad)
Date:   Thu Aug 20 22:58:24 2020 +0300

Bin overly eager early return that stops replacement image creation

one of these crashes, ooo38104-1.sxw on export to docx began with the
above commit.

To reproduce:
wget -O ooo38104-1.sxw 
https://bz.apache.org/ooo/attachment.cgi?id=19889
./instdir/program/soffice.bin --headless --convert-to docx ./ooo38104-
1.sxw

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

Re: Crash test update

[Tor Lillqvist]

> On 2020-08-27, at 15.03, Caolán McNamara  wrote:
> 
> On Wed, 2020-08-26 at 05:57 +, crashtest wrote:
>> New crashtest update available at 
>> https://dev-builds.libreoffice.org/crashtest/03e28651142af2ffa7a3749b879f31bde98a2173/
> 
> commit c123bfff501229f398a1b679fc7434b82d53685c (HEAD, refs/bisect/bad)
> Date:   Thu Aug 20 22:58:24 2020 +0300
> 
>Bin overly eager early return that stops replacement image creation
> 
> one of these crashes, ooo38104-1.sxw on export to docx began with the
> above commit.


OK, investigating.

—tml

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

Re: Crash test update

[Caolán McNamara]

On Mon, 2020-08-24 at 07:19 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/e2487112b6a46849a0e12a12f8d097b16c1a165d/

FWI, one of these, the export of fdo84691-1.odt to docx began with...
commit fd88bb81236ad6d1a63436b7ce20f2f93410d359 (HEAD, refs/bisect/bad)
Date:   Thu Jul 16 16:56:33 2020 +0200

tdf#107225 tdf#122887 sw: fix reanchoring of textboxes

to reproduce...

wget -O fdo84691-1.odt 
https://bugs.documentfoundation.org/attachment.cgi?id=107366

./instdir/program/soffice.bin --headless --convert-to docx ./fdo84691-
1.odt 

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

Re: Crash test update

[Noel Grandin]

On Sat, 22 Aug 2020 at 20:26, Caolán McNamara  wrote:

> On Sat, 2020-08-22 at 07:33 +, crashtest wrote:
> > New crashtest update available at
> >
> https://dev-builds.libreoffice.org/crashtest/ba34f7a04addba42b9c6f46d6be7243bbbf2bab6/
>
> ooo44895-2.sxi is a reproducible example of an assert at
> ximpbody.cxx:342 since 7b83a28f1eb7d0687d07cfaaf8cfaaf9a8eea8c1
> which seems to appear in a lot of the recent failures
>
>

thanks, fix in progress

   https://gerrit.libreoffice.org/c/core/+/101192 remove debugging asserts
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Sat, 2020-08-22 at 07:33 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/ba34f7a04addba42b9c6f46d6be7243bbbf2bab6/

ooo44895-2.sxi is a reproducible example of an assert at
ximpbody.cxx:342 since 7b83a28f1eb7d0687d07cfaaf8cfaaf9a8eea8c1
which seems to appear in a lot of the recent failures

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

Re: Crash test update

[Caolán McNamara]

On Tue, 2020-05-05 at 10:16 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/224c155d89e4d957c0d566a5aea6004a466d54e9/

bisected the crash of exporting ooo24576-1.doc to odt to...

commit 98d1622b3721fe899c4e1faa0b4cc35695253014 (refs/bisect/bad)
Date:   Fri Dec 6 12:37:43 2019 +0100

tdf#121300 sw: adapt at-char selection to latest at-para design

which might be useful information

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

Re: Crash test update

[Noel Grandin]

The "right" answer is probably rewrite the SVG import to use drawinglayer.

A more useful answer is:
Assuming that the SVG format announces it's number range somewhere,
we could check if the number range is stupid big and scale all the numbers
down into a sensible range.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Wed, 2020-04-08 at 04:30 +, crashtest wrote:
> New crashtest update available at 
> https://dev-builds.libreoffice.org/crashtest/27d04f6dbf38aa28fb7215590d578c4567db57

There's a bunch of the same assert listed there:

vcl/source/outdev/map.cxx:366: long int ImplLogicToPixel(long int, long
int, long int, long int, long int): Assertion `nMapNum == 0 ||
std::abs(n) < std::numeric_limits::max() / nMapNum / nDPI'
failed.

using ./soffice --headless --convert-to pdf on e.g.
the svg attachment 
https://bugzilla.mozilla.org/attachment.cgi?id=9058902 (of mozilla bug
https://bugzilla.mozilla.org/show_bug.cgi?id=1545040) reproduces this

The assert seems to have become noticeable since

commit 62ac8333999c661432adb0a18245a399daa89dcb (HEAD)
Date:   Fri Feb 14 16:47:14 2020 +0100

tdf#130655 callback interface for 3D and secure dash

But that looks blameless in the sense that it probably just timed out
before that change. The original svg contains massive numbers in the
input source path. Anyone got any thoughts about what we should do with
cases like this ?

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

Re: Crash test update

[Caolán McNamara]

On Mon, 2019-12-16 at 09:14 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/261b91a043d85f6b902145d208a6f63a10050b67/

tdf106084-1.docx i.e. 
https://bugs.documentfoundation.org/attachment.cgi?id=131334 now
asserts on load since...

commit 5291c3ef56948a91166a597033f3e48e00d57403
Date:   Sun Dec 1 05:41:43 2019 +1100

tdf#43157 basegfx: remove OSL_ENSURE preconditions

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

Re: Crash test update

[Jan-Marek Glogowski]

Am 21.11.19 um 16:58 schrieb Caolán McNamara:
>>1 SwIndexReg::~SwIndexReg
>>5 SwSubFont::GetTextSize_
>> this one is my fault ^
>>   14 writerfilter::dmapper::DomainMapper_Impl::PushFootOrEndnote
> I believe ^this^ one is since 
>
> commit a991ad93dcd6807d0eacd11a50c2ae43a2cfb882
> Author: Jan-Marek Glogowski 
> Date:   Tue Oct 22 12:24:43 2019 +0200
>
> tdf#121441 improve DOCX footnote import
>
> caolanm->jmux: Import of e.g. tdf103063-1.docx is an example of these
> 14, any ideas on that assert ?

I think we already talked about this on IRC and I said I'm Ok with
removing it. Technically the code can't handle nested footnotes (never
could, as it's just using a "global" boolean in DomainMapper_Impl). As I
understood the specs, it's also not allowed, but I might be wrong here.
So you can remove that assert, and if the rest of the code eventually
survives, it'll just produce some interesting variant of a "correct"
document ;-)

Or replace it with a SAL_WARN, but that probably won't help it get fixed
either, as my debug build spills warns every time I open a document, or
even just start LO.

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

Re: Crash test update

[Caolán McNamara]

On Thu, 2019-11-21 at 15:54 +0100, Michael Stahl wrote:
> On 21.11.19 15:17, Crashtest VM wrote:
> > New crashtest update available at 
> > http://dev-builds.libreoffice.org/crashtest/299a13e8f7307b38ac10ad351273e2559e21ab16/
> 
> we have run some business intelligence AI algorithms via mapreduce
> on 
> the clouds to classify the crashes:
> 
> > for f in *; do grep -A2 __assert_fail $f | tail -n 1; done | cut -f
> > 4 -d ' ' | sort | uniq -c
>1 (anonymous
>1 
> com::sun::star::uno::Reference::operat
> or->

>  176 DocxAttributeOutput::CharColor
>   36 sax_fastparser::FastSaxSerializer::writeFastAttributeList

I think both of these are since the same commit of
1d81d52b5da45f26e9d3adeb3b279eb9a488b94f I think I'll revert that with 
https://gerrit.libreoffice.org/#/c/83406/ to get the noise level down

>1 SwIndexReg::~SwIndexReg
>5 SwSubFont::GetTextSize_
> this one is my fault ^
>   14 writerfilter::dmapper::DomainMapper_Impl::PushFootOrEndnote

I believe ^this^ one is since 

commit a991ad93dcd6807d0eacd11a50c2ae43a2cfb882
Author: Jan-Marek Glogowski 
Date:   Tue Oct 22 12:24:43 2019 +0200

tdf#121441 improve DOCX footnote import

caolanm->jmux: Import of e.g. tdf103063-1.docx is an example of these
14, any ideas on that assert ?

>1 writerfilter::dmapper::DomainMapper_Impl::RemoveTopRedline
> 
> > for f in *; do grep -A1 "signal [0-57-9]" $f | tail -n 1; done |
> > cut -f 5 -d ' ' | sort | uniq -c
>6 chart::WrappedProperty::getPropertyState
>1 (this=0x5cbe690,

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

Re: Crash test update

[Michael Stahl]

On 21.11.19 15:17, Crashtest VM wrote:

New crashtest update available at 
http://dev-builds.libreoffice.org/crashtest/299a13e8f7307b38ac10ad351273e2559e21ab16/


we have run some business intelligence AI algorithms via mapreduce on 
the clouds to classify the crashes:



for f in *; do grep -A2 __assert_fail $f | tail -n 1; done | cut -f 4 -d ' ' | 
sort | uniq -c

  1 (anonymous
  1 
com::sun::star::uno::Reference::operator->

176 DocxAttributeOutput::CharColor
 36 sax_fastparser::FastSaxSerializer::writeFastAttributeList
  1 SwIndexReg::~SwIndexReg
  5 SwSubFont::GetTextSize_
this one is my fault ^
 14 writerfilter::dmapper::DomainMapper_Impl::PushFootOrEndnote
  1 writerfilter::dmapper::DomainMapper_Impl::RemoveTopRedline



for f in *; do grep -A1 "signal [0-57-9]" $f | tail -n 1; done | cut -f 5 -d ' 
' | sort | uniq -c

  6 chart::WrappedProperty::getPropertyState
  1 (this=0x5cbe690,
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Sun, 2019-11-17 at 14:09 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/82e1697d29fc783905a71278d36cec1d07e134a3/

There a lot of recent failures like 
https://dev-builds.libreoffice.org/crashtest/82e1697d29fc783905a71278d36cec1d07e134a3/backtraces/task666-core.12276.backtrace.txt

Looks like writing duplicate properties ?

reproducible with e.g. export of 
https://bz.apache.org/ooo/attachment.cgi?id=84837 to docx

they seem to have started with...

commit 1d81d52b5da45f26e9d3adeb3b279eb9a488b94f
Author: Tünde Tóth 

tdf#127579 DOCX export: fix losing color and underline of ODT
hyperlinks

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

Re: Crash test update

[Xisco Fauli]

I've also found the /srv/crashtestdata/files/odt/tdf122829-2.odt.

I reported it in https://bugs.documentfoundation.org/show_bug.cgi?id=128737

El 11/11/19 a les 6:32, Crashtest VM ha escrit:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/658925635262d21f93bedcc6dfeaa934cbcef1cf/
>
> ___
> LibreOffice mailing list
> LibreOffice@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/libreoffice

-- 
Xisco Faulí
Libreoffice QA Team
IRC: x1sc0

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

Re: Crash test update

[Caolán McNamara]

On Thu, 2019-10-31 at 00:45 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/5f5995097344772c59a9690d8e06eb7b39af3c99/

Crash:/srv/crashtestdata/files/docx/fdo60771-1.docx through disposed
is new since...

commit a991ad93dcd6807d0eacd11a50c2ae43a2cfb882
Date:   Tue Oct 22 12:24:43 2019 +0200

tdf#121441 improve DOCX footnote import

fdo60771-1.docx is this attachment
https://bugs.documentfoundation.org/attachment.cgi?id=74730

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

Re: Crash test update

[Dennis Francis]

https://cgit.freedesktop.org/libreoffice/core/commit/?id=d4053e492af3815870e1b4e765486d2aaa29179d
should fix the crash on fdo31831-1.ods
The other calc crashes in more recent crash-tests seems not related to the
threading work.

Thanks,
Dennis

On Tue, Oct 22, 2019 at 6:40 PM Dennis Francis 
wrote:

> Update : Crash on tdf77970-1.ods is fixed, but still working on
> fdo31831-1.ods.
>
> Thanks,
> Dennis
>
> On Sat, Oct 5, 2019 at 10:55 PM Dennis Francis 
> wrote:
>
>> At least the below two crashes are due to commit
>> 845e1cdca3349c72e3083186502285d5b776abbe, and I'm working on a fix.
>>
>> Crash:/srv/crashtestdata/files/ods/tdf77970-1.ods through disposed
>> Crash:/srv/crashtestdata/files/ods/fdo31831-1.ods through disposed
>>
>> Thanks,
>> Dennis
>>
>> On Sat, Oct 5, 2019 at 9:22 PM Crashtest VM <
>> crashtest.libreoff...@gmail.com> wrote:
>>
>>> New crashtest update available at
>>> http://dev-builds.libreoffice.org/crashtest/314958decf4491ca7b11c2ef014d29ea2ae14d69/
>>> ___
>>> LibreOffice mailing list
>>> LibreOffice@lists.freedesktop.org
>>> https://lists.freedesktop.org/mailman/listinfo/libreoffice
>>
>>
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Dennis Francis]

Update : Crash on tdf77970-1.ods is fixed, but still working on
fdo31831-1.ods.

Thanks,
Dennis

On Sat, Oct 5, 2019 at 10:55 PM Dennis Francis 
wrote:

> At least the below two crashes are due to commit
> 845e1cdca3349c72e3083186502285d5b776abbe, and I'm working on a fix.
>
> Crash:/srv/crashtestdata/files/ods/tdf77970-1.ods through disposed
> Crash:/srv/crashtestdata/files/ods/fdo31831-1.ods through disposed
>
> Thanks,
> Dennis
>
> On Sat, Oct 5, 2019 at 9:22 PM Crashtest VM <
> crashtest.libreoff...@gmail.com> wrote:
>
>> New crashtest update available at
>> http://dev-builds.libreoffice.org/crashtest/314958decf4491ca7b11c2ef014d29ea2ae14d69/
>> ___
>> LibreOffice mailing list
>> LibreOffice@lists.freedesktop.org
>> https://lists.freedesktop.org/mailman/listinfo/libreoffice
>
>
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Michael Stahl]

On 10.10.19 13:51, Michael Stahl wrote:

On 04.10.19 11:03, Caolán McNamara wrote:

On Wed, 2019-10-02 at 01:01 +, Crashtest VM wrote:

New crashtest update available at
http://dev-builds.libreoffice.org/crashtest/eeac5b7a424c3a3be48b1c3583171ad97d1b19cd/ 



The export of fdo45983-1.doc to odt has been asserting in the
crashtesting for a while.

https://bugs.documentfoundation.org/attachment.cgi?id=56949 is the .doc
and it seems to have begun asserting(in lcl_RemoveFieldMarks) since...

commit dc9a0b124272a6dc0a8d875d51b6f882e60a8a04
Author: László Németh 
Date:   Wed Aug 7 11:25:13 2019 +0200

 tdf#127101 Change tracking: reject format at paragraph join


i'm afraid this can't work, you can't split a redline at arbitrary 
positions because then the user can reject one of the parts and accept 
the other... in this case one of the parts contains a 
CH_TXT_ATR_FIELDSTART and another the corresponding CH_TXT_ATR_FIELDEND, 
but either one redline must contain both of them or neither (see 
https://gerrit.libreoffice.org/#/c/80070/ for how i'm planning to 
prevent this on the editing operations but of course i didn't notice 
this code when i wrote that...)


it turned out that in this particular document the problem was that the 
first redline that was created was already wrong, which is fixed with 
commit 954c158acb87eb0385c51fdfff38fff84f154051 "sw: WW8: fix 
crashtesting asserts on fdo45983-1.doc export to ODT" - but in general 
the splitting of delete redlines remains a bad idea.


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

Re: Crash test update

[Michael Stahl]

On 04.10.19 11:03, Caolán McNamara wrote:

On Wed, 2019-10-02 at 01:01 +, Crashtest VM wrote:

New crashtest update available at
http://dev-builds.libreoffice.org/crashtest/eeac5b7a424c3a3be48b1c3583171ad97d1b19cd/


The export of fdo45983-1.doc to odt has been asserting in the
crashtesting for a while.

https://bugs.documentfoundation.org/attachment.cgi?id=56949 is the .doc
and it seems to have begun asserting(in lcl_RemoveFieldMarks) since...

commit dc9a0b124272a6dc0a8d875d51b6f882e60a8a04
Author: László Németh 
Date:   Wed Aug 7 11:25:13 2019 +0200

 tdf#127101 Change tracking: reject format at paragraph join


i'm afraid this can't work, you can't split a redline at arbitrary 
positions because then the user can reject one of the parts and accept 
the other... in this case one of the parts contains a 
CH_TXT_ATR_FIELDSTART and another the corresponding CH_TXT_ATR_FIELDEND, 
but either one redline must contain both of them or neither (see 
https://gerrit.libreoffice.org/#/c/80070/ for how i'm planning to 
prevent this on the editing operations but of course i didn't notice 
this code when i wrote that...)

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

Re: Crash test update

[Dennis Francis]

At least the below two crashes are due to commit
845e1cdca3349c72e3083186502285d5b776abbe, and I'm working on a fix.

Crash:/srv/crashtestdata/files/ods/tdf77970-1.ods through disposed
Crash:/srv/crashtestdata/files/ods/fdo31831-1.ods through disposed

Thanks,
Dennis

On Sat, Oct 5, 2019 at 9:22 PM Crashtest VM 
wrote:

> New crashtest update available at
> http://dev-builds.libreoffice.org/crashtest/314958decf4491ca7b11c2ef014d29ea2ae14d69/
> ___
> LibreOffice mailing list
> LibreOffice@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/libreoffice
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Wed, 2019-10-02 at 01:01 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/eeac5b7a424c3a3be48b1c3583171ad97d1b19cd/

The export of fdo45983-1.doc to odt has been asserting in the
crashtesting for a while.

https://bugs.documentfoundation.org/attachment.cgi?id=56949 is the .doc
and it seems to have begun asserting(in lcl_RemoveFieldMarks) since...

commit dc9a0b124272a6dc0a8d875d51b6f882e60a8a04
Author: László Németh 
Date:   Wed Aug 7 11:25:13 2019 +0200

tdf#127101 Change tracking: reject format at paragraph join

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

Re: Crash test update

[Caolán McNamara]

On Wed, 2019-10-02 at 01:01 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/eeac5b7a424c3a3be48b1c3583171ad97d1b19cd/

https://gerrit.libreoffice.org/#/c/80010/ is my stab at fixing most of
those new ones

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

Re: Crash test update

[Michael Stahl]

On 12.08.19 22:33, Caolán McNamara wrote:

On Mon, 2019-08-12 at 01:03 +, Crashtest VM wrote:

New crashtest update available at
http://dev-builds.libreoffice.org/crashtest/80042cc0e9786fe8dcabe0b8f106183f8df19355/


I'm sure you'll be thrilled to know that the assert on export of
ooo39250-1.rtf to docx, the endFastElement assert about unbalanced tags
on export, was according to git bisect, triggered by

commit 28b77c89dfcafae82cf2a6d85731b643ff9290e5
Date:   Thu Jul 11 18:37:28 2019 +0200

 tdf#117185 tdf#110442 sw: bring harmony & peace to fly at-char
selection


not entirely sure why the import anchor position changed here, but it 
turns out that it's now in the same place where Word puts it, so we can 
call it a happy accident.


the DOCX export should be fixed on master now with commit 
e99f0182ca00d137238f89de67a06415d3c873e3 "sw: fix assert on export of 
ooo39250-1.rtf to DOCX"

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

Re: Crash test update

[Caolán McNamara]

On Mon, 2019-08-12 at 01:03 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/80042cc0e9786fe8dcabe0b8f106183f8df19355/

I'm sure you'll be thrilled to know that the assert on export of
ooo39250-1.rtf to docx, the endFastElement assert about unbalanced tags
on export, was according to git bisect, triggered by

commit 28b77c89dfcafae82cf2a6d85731b643ff9290e5
Date:   Thu Jul 11 18:37:28 2019 +0200

tdf#117185 tdf#110442 sw: bring harmony & peace to fly at-char
selection

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

Re: Crash test update

[Caolán McNamara]

On Fri, 2019-08-09 at 08:52 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/f5c27036ad63c8bc3d9b86f2e4e772c01490d883/

since 9a19a1235c301f2699157ab44049bdfd54955e8a
there are quite a few new rtf import assert failures, e.g.
https://bugzilla.abisource.com/attachment.cgi?id=2287

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

Re: Crash test update

[Caolán McNamara]

On Wed, 2019-08-07 at 13:17 +, Kaganski Mike wrote:
> On 27.04.2019 18:33, Caolán McNamara wrote:
> > On Sat, 2019-04-27 at 00:33 +, Crashtest VM wrote:
> > > New crashtest update available at
> > > http://dev-builds.libreoffice.org/crashtest/0b56eb2c10ab4ca027f1a37e04519366b3cd7433/
> > 
> > soffice.bin --headless --convert-to xlsx ooo83250-1.ods
> > semi-reliably assert with null pDim at xepivotxml.cxx:504
> > I presume this started with
> > b082998401e37e6c7534906601bc481423a6ded0
> > tdf#113908: Implement exporting pivot tables' groups fields to XSLX
> 
> I'm sorry it took too long; https://gerrit.libreoffice.org/77110 
> addresses this.

oh excellent, I appreciate it.

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

Re: Crash test update

[Kaganski Mike]

On 27.04.2019 18:33, Caolán McNamara wrote:
> On Sat, 2019-04-27 at 00:33 +, Crashtest VM wrote:
>> New crashtest update available at
>> http://dev-builds.libreoffice.org/crashtest/0b56eb2c10ab4ca027f1a37e04519366b3cd7433/
> 
> 
> soffice.bin --headless --convert-to xlsx ooo83250-1.ods
> semi-reliably assert with null pDim at xepivotxml.cxx:504
> I presume this started with b082998401e37e6c7534906601bc481423a6ded0
> tdf#113908: Implement exporting pivot tables' groups fields to XSLX

I'm sorry it took too long; https://gerrit.libreoffice.org/77110 
addresses this.

-- 
Best regards,
Mike Kaganski
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Tamás Zolnai]

Hello guys,

I checked this crashtest results whether there is any related crash to my
MSForms changes. Three documents were suspicious to me:
/srv/crashtestdata/files/rtf/ooo39250-1.docx
/srv/crashtestdata/files/docx/fdo78590-1.docx
/srv/crashtestdata/files/cloudon/File_319.docx

I checked them and it seems only one is caused by my MSForms changes (the
other two do not contain any date field):
/srv/crashtestdata/files/cloudon/File_319.docx

This one will be fixed with this patch:
https://gerrit.libreoffice.org/#/c/76971/

Best Regards,
Tamás


Crashtest VM  ezt írta (időpont: 2019.
júl. 27., Szo, 18:43):

> New crashtest update available at
> http://dev-builds.libreoffice.org/crashtest/3782a36382af0609aed73caa8334b6f1c55dc93c/
> ___
> LibreOffice mailing list
> LibreOffice@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/libreoffice
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Sat, 2019-08-03 at 04:45 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/2207269a84c7c9920af3385b837ce67978c720

FWIW the new bazillion export failures on export to odt seem to be
triggered by beec1594587d0bf1ea2268f9a435c948b5580278

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

Re: Crash test update

[Tamas Zolnai]

Hi Caolan, 

Péntek, Július 19, 2019 16:13 BST dátummal, Caolán McNamara 
 ezt írta: 
 
> On Tue, 2019-07-16 at 18:07 +, Crashtest VM wrote:
> > New crashtest update available at 
> > http://dev-builds.libreoffice.org/crashtest/f926de93c455f7cede505ad8cf849fe81eb514cf/
> 
> FWIW three or four of the failures here I think are since integration
> of the MSForms work, i.e. tdf99236-1.docx and tdf99002-2.docx on import
> and tdf100679-1.docx on reexport to docx

Thanks for the note! I'm on the issue. I've found the root cause, it will be 
fixed soon.

Best Regards,
Tamás

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

Re: Crash test update

[Caolán McNamara]

On Tue, 2019-07-16 at 18:07 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/f926de93c455f7cede505ad8cf849fe81eb514cf/

FWIW three or four of the failures here I think are since integration
of the MSForms work, i.e. tdf99236-1.docx and tdf99002-2.docx on import
and tdf100679-1.docx on reexport to docx

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

Re: Crash test update, now with backtraces

[Kaganski Mike]

On 26.05.2019 17:14, Caolán McNamara wrote:
> On Fri, 2019-05-24 at 10:31 +0200, Luboš Luňák wrote:
>>   Could you please name those .backtrace.txt? I can't view those files
>> online, because they get treated as binary files and so Firefox only
>> offers to download them.
> 
> Yeah, I changed that in the script after the first run, it'll be .txt
> in future ones.

I suppose it should be more about proper content type as reported by 
server, than about name.

-- 
Best regards,
Mike Kaganski
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update, now with backtraces

[Caolán McNamara]

On Fri, 2019-05-24 at 10:31 +0200, Luboš Luňák wrote:
>  Could you please name those .backtrace.txt? I can't view those files
> online, because they get treated as binary files and so Firefox only
> offers to download them.

Yeah, I changed that in the script after the first run, it'll be .txt
in future ones.

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

Re: Crash test update, now with backtraces

[Luboš Luňák]

On Thursday 23 of May 2019, Caolán McNamara wrote:
> On Thu, 2019-05-23 at 18:54 +, Crashtest VM wrote:
> > New crashtest update available at
> > http://dev-builds.libreoffice.org/crashtest/96392d5c5319b0d6d3e8b07e78a19
> >9bf4b44a5ba/
>
> As mentioned in the call, its rough and ready, but the backtraces of
> crashes are included in this upload e.g.
> https://dev-builds.libreoffice.org/crashtest/96392d5c5319b0d6d3e8b07e78a199
>bf4b44a5ba/backtraces/


 Could you please name those .backtrace.txt? I can't view those files online, 
because they get treated as binary files and so Firefox only offers to 
download them.

-- 
 Luboš Luňák
 l.lu...@collabora.com
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update, now with backtraces

[Miklos Vajna]

Hi Caolan,

On Thu, May 23, 2019 at 08:37:35PM +0100, Caolán McNamara  
wrote:
> As mentioned in the call, its rough and ready, but the backtraces of
> crashes are included in this upload e.g. 
> https://dev-builds.libreoffice.org/crashtest/96392d5c5319b0d6d3e8b07e78a199bf4b44a5ba/backtraces/

Oh, this is nice, thanks. :-)

Regards,

Miklos


signature.asc
Description: Digital signature
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update, now with backtraces

[Caolán McNamara]

On Thu, 2019-05-23 at 18:54 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/96392d5c5319b0d6d3e8b07e78a199bf4b44a5ba/

As mentioned in the call, its rough and ready, but the backtraces of
crashes are included in this upload e.g. 
https://dev-builds.libreoffice.org/crashtest/96392d5c5319b0d6d3e8b07e78a199bf4b44a5ba/backtraces/

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

Re: Crash test update

[Noel Grandin]

On 2019/04/27 5:33 PM, Caolán McNamara wrote:

--headless --convert-to pdf tdf89833-2.xlsm
semi-reliably fails stl concept checking "Error: attempt to compare a
dereferenceable iterator to a singular iterator"
I imagine this started with some part of the tdf#81765 fixes



fix in progress here:

https://gerrit.libreoffice.org/71677

had to revert my perf improvement, sigh.
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Sat, 2019-04-27 at 00:33 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/0b56eb2c10ab4ca027f1a37e04519366b3cd7433/


soffice.bin --headless --convert-to xlsx ooo83250-1.ods
semi-reliably assert with null pDim at xepivotxml.cxx:504
I presume this started with b082998401e37e6c7534906601bc481423a6ded0
tdf#113908: Implement exporting pivot tables' groups fields to XSLX

--headless --convert-to pdf tdf89833-2.xlsm
semi-reliably fails stl concept checking "Error: attempt to compare a
dereferenceable iterator to a singular iterator"
I imagine this started with some part of the tdf#81765 fixes

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

Re: Crash test update

[Noel Grandin]

>
> There seems to be outstanding just two different distinct crashes,
> which looked sufficiently non-trivial I didn't have a stab at fixing it
> myself, one on reexport to ods of ooo118068-1.ods and one on import of
> e.g. ooo6593-5.sxc
>
>
>
Fixed now with 3df2e918408535acd2c4d028bfae70901ad24e62
and ac9aa53d51a4f359342274d107aed6df85658c55
___
LibreOffice mailing list
LibreOffice@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/libreoffice

Re: Crash test update

[Caolán McNamara]

On Wed, 2019-04-10 at 18:40 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/9575aa3bba055258977d9a013889aa3acbaecb6a/

Looking through these crash failure, they break down into two
categories...

dynamic column stuff in calc

There seems to be outstanding just two different distinct crashes,
which looked sufficiently non-trivial I didn't have a stab at fixing it
myself, one on reexport to ods of ooo118068-1.ods and one on import of
e.g. ooo6593-5.sxc

and

in writer, the assert of !m_bLockNotifyContentChange 
in SwTextInputField e.g. import of ooo48071-4.odt but all the rest of
the .odt failures happen in the same place

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

Re: Crash test update

[Caolán McNamara]

On Thu, 2019-01-31 at 16:26 +, Crashtest VM wrote:
> New crashtest update available at 
> http://dev-builds.libreoffice.org/crashtest/3a2c2ad839860dcb36e30ae1a2867e4d5e1036fc/


Those export to .rtf asserts/crashes begun at...

commit 18bb85aa303c17b7705a9ccb0dd3d76673a1955e
Author: Bjoern Michaelsen 
Date:   Sat Jan 12 12:04:04 2019 +0100

make SwFormat a ::sw::BroadcastingModify

- clean up SwSection..Hint: Lazily reusing the Dying HintId is evil
- switch the SwChartDataSequence to SvtListener
- add some description to RuntimeException

And if I revert just the SwChartDataSequence hunk it doesn't. 

Comparing what happens in the old ::Modify vs new ::Notify, in the new
case we get additional FindUnoInstanceHint hints which didn't seem to
arrive in the ::Modify case. The additional hints trigger modified in
SwChartDataSequence::Notify which didn't happen before which seems to
be the problem ?

This hackaround at https://gerrit.libreoffice.org/#/c/67394/ seems to
give the original behaviour, though I'm unsure if that's the right way
to go.

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