Bug 105473 - Crash in: ScRangeList::operator[](unsigned int)
Summary: Crash in: ScRangeList::operator[](unsigned int)
Status: VERIFIED FIXED
Alias: None
Product: LibreOffice
Classification: Unclassified
Component: Calc (show other bugs)
Version:
(earliest affected)
5.2.4.2 release
Hardware: x86-64 (AMD64) Windows (All)
: medium normal
Assignee: Not Assigned
QA Contact:
URL:
Whiteboard: target:5.4.0 target:5.3.1
Keywords: haveBacktrace
Depends on:
Blocks:
 
Reported: 2017-01-22 10:31 UTC by Hans Gerstenkorn
Modified: 2017-02-12 17:37 UTC (History)
4 users (show)

See Also:
Crash report or crash signature: ["ScRangeList::operator[](unsigned int)"]


Attachments
DHL Adress import assistent from paket.de (1.26 MB, application/vnd.ms-excel)
2017-01-22 10:31 UTC, Hans Gerstenkorn
Details
bt with debug symbols (6.33 KB, text/plain)
2017-01-22 13:26 UTC, Julien Nabet
Details
bt with debug symbols (16.67 KB, text/plain)
2017-01-24 20:09 UTC, Julien Nabet
Details
Mistake 1 after helppackinstallation (10.97 KB, image/png)
2017-02-12 17:26 UTC, Hans Gerstenkorn
Details
Mistake 2 after helppackinstallation (27.41 KB, image/png)
2017-02-12 17:28 UTC, Hans Gerstenkorn
Details

Note You need to log in before you can comment on or make changes to this bug.
Description Hans Gerstenkorn 2017-01-22 10:31:36 UTC
Created attachment 130605 [details]
DHL Adress import assistent from paket.de

This bug was filed from the crash reporting server and is br-69adfa0c-aa19-4903-a81a-edac746952c7.
=========================================

This adtress import tool from the german DHL (www.paket.de) produces macro mistakes which I can't identify, my be this will only work within the newest Microsoft Excel Version.....
Comment 1 m.a.riosv 2017-01-22 10:50:35 UTC
Don't crash.
Version: 5.2.5.1 (x64)
Build ID: 0312e1a284a7d50ca85a365c316c7abbf20a4d22
CPU Threads: 4; OS Version: Windows 6.19; UI Render: GL; 
Locale: es-ES (es_ES); Calc: group

It's a xls file with VBA macro.
Comment 2 Xisco Faulí 2017-01-22 11:38:01 UTC Comment hidden (obsolete)
Comment 3 Julien Nabet 2017-01-22 13:26:35 UTC
Created attachment 130611 [details]
bt with debug symbols

On pc Debian x86-64 with master sources updated 2 days ago, I could reproduce the crash but with a different bt.

I attached bt + some gdb printing
Comment 4 Julien Nabet 2017-01-22 13:36:00 UTC
With this patch:
diff --git a/basic/source/comp/scanner.cxx b/basic/source/comp/scanner.cxx
index 092a239eabe9..398b78a6cdbd 100644
--- a/basic/source/comp/scanner.cxx
+++ b/basic/source/comp/scanner.cxx
@@ -569,7 +569,7 @@ eoln:
     {
         pLine = nullptr;
         bool bRes = NextSym();
-        if( bVBASupportOn && aSym[0] == '.' )
+        if( bVBASupportOn && aSym.startsWith(".") )
         {
             // object _
             //    .Method

No crash but I got a popup error in German:
"Es ist ein Fehler bei der Erstellung der neuen 
Symbolleiste aufgetreten !"
Comment 5 Hans Gerstenkorn 2017-01-22 14:26:02 UTC Comment hidden (obsolete)
Comment 6 Xisco Faulí 2017-01-22 15:12:05 UTC
(In reply to Julien Nabet from comment #4)
> With this patch:
> diff --git a/basic/source/comp/scanner.cxx b/basic/source/comp/scanner.cxx
> index 092a239eabe9..398b78a6cdbd 100644
> --- a/basic/source/comp/scanner.cxx
> +++ b/basic/source/comp/scanner.cxx
> @@ -569,7 +569,7 @@ eoln:
>      {
>          pLine = nullptr;
>          bool bRes = NextSym();
> -        if( bVBASupportOn && aSym[0] == '.' )
> +        if( bVBASupportOn && aSym.startsWith(".") )
>          {
>              // object _
>              //    .Method
> 
> No crash but I got a popup error in German:
> "Es ist ein Fehler bei der Erstellung der neuen 
> Symbolleiste aufgetreten !"

Hello Julien,
I get that german popup without doing any modification in the code using

Version: 5.4.0.0.alpha0+
Build ID: d3ff66999d924e832f8219c65ced0526f1a67f82
CPU Threads: 4; OS Version: Linux 4.8; UI Render: default; VCL: gtk2; 
Locale: ca-ES (ca_ES.UTF-8); Calc: group
Comment 7 Hans Gerstenkorn 2017-01-22 15:26:52 UTC
Hi to all,
this is the bug I've reportet - please see above, but I opened a new question at this website: "http://www.libreoffice-forum.de/viewtopic.php?f=6&t=17255" and I've got now one answer which says that the bug is justified in the VBA-Code. Please see the complete answer there. :-) hans
Comment 8 Buovjaga 2017-01-24 08:08:10 UTC
(In reply to Hans Gerstenkorn from comment #7)
> Hi to all,
> this is the bug I've reportet - please see above, but I opened a new
> question at this website:
> "http://www.libreoffice-forum.de/viewtopic.php?f=6&t=17255" and I've got now
> one answer which says that the bug is justified in the VBA-Code. Please see
> the complete answer there. :-) hans

"Bug is justified" - does it mean this can be closed?
Comment 9 Hans Gerstenkorn 2017-01-24 08:29:55 UTC
I think so, but I'm sure that was not the problem of the crash, but it is not a worth to track a problem....
Comment 10 Buovjaga 2017-01-24 08:37:37 UTC
Well it was a silly proposal from me. A crash should never happen.
Let's leave this open and Julien can comment.
Comment 11 Julien Nabet 2017-01-24 09:05:25 UTC
I typed a git "pull -r && make clean && make" this morning on my local repo, so I'll be able to retest tonight after my day time job.
Comment 12 Julien Nabet 2017-01-24 20:09:25 UTC
Created attachment 130665 [details]
bt with debug symbols

On pc Debian x86-64 with master sources updated today (b1211e965aa7883b0c3fc157f6c56b48c8bb0462), I still reproduce the crash.
Comment 13 Xisco Faulí 2017-01-30 23:22:22 UTC
Moving to NEW as per comment 12
Comment 14 Commit Notification 2017-02-05 13:27:01 UTC
Julien Nabet committed a patch related to this issue.
It has been pushed to "master":

http://cgit.freedesktop.org/libreoffice/core/commit/?id=89b57084a0b76c20d6f98607c88b47009a1975d5

Related tdf#105473: Crash in: ScRangeList::operator[]

It will be available in 5.4.0.

The patch should be included in the daily builds available at
http://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More
information about daily builds can be found at:
http://wiki.documentfoundation.org/Testing_Daily_Builds

Affected users are encouraged to test the fix and report feedback.
Comment 15 Julien Nabet 2017-02-05 13:29:18 UTC
Finally I pushed the patch proposed on master (see https://cgit.freedesktop.org/libreoffice/core/commit/?id=89b57084a0b76c20d6f98607c88b47009a1975d5) and cherry-picked it on 5.3 branch for review here: https://gerrit.libreoffice.org/#/c/33935/
So let's put this one to FIXED.
Comment 16 Commit Notification 2017-02-07 13:00:23 UTC
Julien Nabet committed a patch related to this issue.
It has been pushed to "libreoffice-5-3":

http://cgit.freedesktop.org/libreoffice/core/commit/?id=50522e609375ad12dec9c24cc302d9107cbf5358&h=libreoffice-5-3

Related tdf#105473: Crash in: ScRangeList::operator[]

It will be available in 5.3.1.

The patch should be included in the daily builds available at
http://dev-builds.libreoffice.org/daily/ in the next 24-48 hours. More
information about daily builds can be found at:
http://wiki.documentfoundation.org/Testing_Daily_Builds

Affected users are encouraged to test the fix and report feedback.
Comment 17 Hans Gerstenkorn 2017-02-09 13:37:58 UTC
sorry, I am sick, but in a few days I report back :-) Hans
Comment 18 Hans Gerstenkorn 2017-02-12 17:26:41 UTC
Created attachment 131149 [details]
Mistake 1 after helppackinstallation

please see comment from 2017-02-12
Comment 19 Hans Gerstenkorn 2017-02-12 17:28:03 UTC
Created attachment 131150 [details]
Mistake 2 after helppackinstallation

please see comment from 2017-02-12
Comment 20 Hans Gerstenkorn 2017-02-12 17:34:03 UTC
Hi, I'm back and after installing helppack "http://dev-builds.libreoffice.org/daily/libreoffice-5-3/Win-x86@62-merge-TDF/2017-02-12_01.26.53/libreoffice-5-3~2017-02-12_01.26.53_LibreOfficeDev_5.3.1.0.0_Win_x86_helppack_de.msi" I've got the same "makro-mistake" - please see attachments above, the LO doesn't crash. 

Please note also that the makro problem will be hopefully resolved by the Germen DHL. (please see also above). :-) Hans
Comment 21 Buovjaga 2017-02-12 17:37:39 UTC
(In reply to Hans Gerstenkorn from comment #20)
> Please note also that the makro problem will be hopefully resolved by the
> Germen DHL. (please see also above). :-) Hans

Good news re: DHL. Let's set to VERIFIED.