RE: LibreOffice Digest, Vol 119, Issue 23

"Mike Sydwest" <mike.sydwest@xxxxxxxxxxx> · Thu, 23 Jul 2020 11:15:56 +1000

Hi, 

we desperately need help to upgrade our code.

Any and all advise will be great.

Thanks and greets.

Michael

-----Original Message-----
From: LibreOffice <libreoffice-bounces@xxxxxxxxxxxxxxxxxxxxx> On Behalf Of
libreoffice-request@xxxxxxxxxxxxxxxxxxxxx
Sent: Wednesday, 22 July 2020 10:00 PM
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: LibreOffice Digest, Vol 119, Issue 23

Send LibreOffice mailing list submissions to
	libreoffice@xxxxxxxxxxxxxxxxxxxxx

To subscribe or unsubscribe via the World Wide Web, visit
	https://lists.freedesktop.org/mailman/listinfo/libreoffice
or, via email, send a message with subject or body 'help' to
	libreoffice-request@xxxxxxxxxxxxxxxxxxxxx

You can reach the person managing the list at
	libreoffice-owner@xxxxxxxxxxxxxxxxxxxxx

When replying, please edit your Subject line so it is more specific than
"Re: Contents of LibreOffice digest..."

Today's Topics:

   1. Re: cannot access opengrok.libreoffice.org (but ping works)
      (himajin100000)
   2. Crash test update (crashtest)
   3. RE:fridge magnet tin sign license plate (Sandy Chan)
   4. Re: Brochure mode signature size (Matthew Ewer)
   5. GSoC "Extending the UITest framework" Week (7) Report
      (ahmed El-Shreif)
   6. Re: Brochure mode signature size (Ilmari Lauhakangas)
   7. Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
      Windows (Stephan Bergmann)
   8. Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
      Windows (Miklos Vajna)
   9. Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
      Windows (Mike Kaganski)

----------------------------------------------------------------------

Message: 1
Date: Tue, 21 Jul 2020 21:50:23 +0900
From: himajin100000 <himajin100000@xxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: Re: cannot access opengrok.libreoffice.org (but ping works)
Message-ID: <f15dd33e-018b-289b-93d9-622a3a9e5672@xxxxxxxxx>
Content-Type: text/plain; charset=utf-8; format=flowed

Hi,

Seemed I had a network trouble on my side, not the opengrok side.

I noticed that when I connect to the server with my smartphone, and I was
able to do so. I restarted the ONU and router device, and confirmed that I
can connect to the server now!

> In such circumstances it is advised to just use git grep or whichever 
> grepping tool you like best against your local clone of the source code.

Thank you for the information!

------------------------------

Message: 2
Date: Tue, 21 Jul 2020 14:04:34 +0000
From: "crashtest" <noreply@xxxxxxxxxxxxxxxxxxxxxx>
To: "libreoffice@xxxxxxxxxxxxxxxxxxxxx"
	<libreoffice@xxxxxxxxxxxxxxxxxxxxx>
Subject: Crash test update
Message-ID: <37508.2889545126-sendEmail@crashtest>
Content-Type: text/plain; charset="iso-8859-1"

New crashtest update available at
https://dev-builds.libreoffice.org/crashtest/181e26b346b0ec8d3d39e4fad2eea71
69175aefe/
-------------- next part --------------
A non-text attachment was scrubbed...
Name: exportCrashes.csv
Type: application/octet-stream
Size: 13367 bytes
Desc: not available
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200721/203
44ed3/attachment-0003.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: importCrash.csv
Type: application/octet-stream
Size: 116041 bytes
Desc: not available
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200721/203
44ed3/attachment-0004.obj>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: validationErrors.csv
Type: application/octet-stream
Size: 20237 bytes
Desc: not available
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200721/203
44ed3/attachment-0005.obj>

------------------------------

Message: 3
Date: Tue, 21 Jul 2020 22:50:30 +0800
From: "Sandy Chan" <sandy@xxxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: RE:fridge magnet tin sign license plate
Message-ID:
	<mailman.26.1595419202.12467.libreoffice@xxxxxxxxxxxxxxxxxxxxx>
Content-Type: text/plain; charset="us-ascii"

An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200721/9b1
69225/attachment.htm>

------------------------------

Message: 4
Date: Tue, 21 Jul 2020 22:41:37 -0400
From: Matthew Ewer <eyeillus@xxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: Re: Brochure mode signature size
Message-ID:
	<CAP5=Ep2n-ckbxF6YfPHr1+9+K7v1Gp6dQ3=GQHqqEKExwu0QoQ@xxxxxxxxxxxxxx>
Content-Type: text/plain; charset="utf-8"

For the record, I figured it out.  What I had modified was the right
setting, for the wrong program.  (Some version of Draw, rather than Writer).
The correct file to modify was sw/source/core/doc/doc.cxx .  My fix did what
I wanted, but to be useful to others, I'd need to e.g. add a spinner to the
print dialog.  Is there a recommended process for doing that; accessibility
tables to not forget, string translation sheets to fill out?  What's the
process for getting a change reviewed/accepted/merged?

On Mon, Jul 20, 2020 at 10:44 AM Matthew Ewer <eyeillus@xxxxxxxxx> wrote:

> Hey, so you know the brochure printing mode? It normally prints a 
> single signature - one stack of pages, where if you fold the stack 
> down the middle, it forms a book. BUT, folding a stack of like 200 
> pages doesn't really work too well. I'd be preferable to specify a 
> signature size - e.g., single sheets of paper, each of which you fold 
> down the middle and stack. I thought I found where the code for that 
> lies - sd/source/ui/view/DocumentRenderer.cxx , PrepareBooklet(), at 
> the end of the function. But 1. the behavior I'd expect from that code 
> doesn't quite match what I see in the preview in the Print window, and 
> 2. after I changed the code to do what I wanted, and recompiled, the 
> Print window behavior was entirely unchanged. Does anybody know where 
> the relevant code actually is, or what's going on here?
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200721/f0c
38271/attachment-0001.htm>

------------------------------

Message: 5
Date: Wed, 22 Jul 2020 04:44:44 +0200
From: ahmed El-Shreif <aelshreif7@xxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: GSoC "Extending the UITest framework" Week (7) Report
Message-ID:
	<CAPkDTbDNVew4fcXZPsQ4CXAvHe_QuW5vayE5kKRJ1iOF3TU1BA@xxxxxxxxxxxxxx>
Content-Type: text/plain; charset="utf-8"

Hello all,

here is my Week (7) report:
https://ahmedelshreifgsoc20.blogspot.com/2020/07/week-7-report.html

Sorry for sending the report late because I started my final exams this
week.

waiting for all your feedback.

Thanks,
Ahmed ElShreif
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200722/e4f
c3477/attachment-0001.htm>

------------------------------

Message: 6
Date: Wed, 22 Jul 2020 08:46:40 +0300
From: Ilmari Lauhakangas <ilmari.lauhakangas@xxxxxxxxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Subject: Re: Brochure mode signature size
Message-ID: <9cda14a5-81ff-cbcf-5e08-7706904ebdc5@xxxxxxxxxxxxxxx>
Content-Type: text/plain; charset=utf-8; format=flowed

The process is to do create small patches that are easy to review and to
analyse in case of future problems. Submit your first patch to gerrit and it
will be easier to discuss: 
https://wiki.documentfoundation.org/Development/gerrit

Ilmari

On 22.7.2020 5.41, Matthew Ewer wrote:
> For the record, I figured it out.? What I had modified was the right 
> setting, for the wrong program.? (Some version of Draw, rather than 
> Writer).? The correct file to modify was sw/source/core/doc/doc.cxx .
> My fix did what I wanted, but to be useful to others, I'd need to e.g. 
> add a spinner to the print dialog.? Is there a recommended process for 
> doing that; accessibility tables to not forget, string translation 
> sheets to fill out?? What's the process for getting a change 
> reviewed/accepted/merged?
> 
> On Mon, Jul 20, 2020 at 10:44 AM Matthew Ewer <eyeillus@xxxxxxxxx 
> <mailto:eyeillus@xxxxxxxxx>> wrote:
> 
>     Hey, so you know the brochure printing mode? It normally prints a
>     single signature - one stack of pages, where if you fold the stack
>     down the middle, it forms a book. BUT, folding a stack of like 200
>     pages doesn't really work too well. I'd be preferable to specify a
>     signature size - e.g., single sheets of paper, each of which you
>     fold down the middle and stack. I thought I found where the code for
>     that lies - sd/source/ui/view/DocumentRenderer.cxx ,
>     PrepareBooklet(), at the end of the function. But 1. the behavior
>     I'd expect from that code doesn't quite match what I see in the
>     preview in the Print window, and 2. after I changed the code to do
>     what I wanted, and recompiled, the Print window behavior was
>     entirely unchanged. Does anybody know where the relevant code
>     actually is, or what's going on here?
> 
> 
> _______________________________________________
> LibreOffice mailing list
> LibreOffice@xxxxxxxxxxxxxxxxxxxxx
> https://lists.freedesktop.org/mailman/listinfo/libreoffice
> 

------------------------------

Message: 7
Date: Wed, 22 Jul 2020 10:18:44 +0200
From: Stephan Bergmann <sbergman@xxxxxxxxxx>
To: libreoffice@xxxxxxxxxxxxxxxxxxxxx
Cc: Serge Krot <Serge.Krot@xxxxxx>
Subject: Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
	Windows
Message-ID: <a45eb3a5-976a-7940-0735-2877055b860f@xxxxxxxxxx>
Content-Type: text/plain; charset=utf-8; format=flowed

On 12/07/2020 18:45, Mike Kaganski wrote:
> On 11.07.2020 2:35, Mike Kaganski wrote:
>> In current master (at least a week, and still the same after pull 
>> just
>> now) I see 100% reproducible failure of make CppunitTest_sw_ooxmlexport4:
>>
>>> $ /opt/lo/bin/make CppunitTest_sw_ooxmlexport4 ...
>>> test_Tdf115030::Load_Reload_Verify finished in: 1500ms
>>> C:/lo/src/core/test/source/xmltesttools.cxx(194) : error : Assertion 
>>> Test name: testComboBoxControl::Load_Reload_Verify
>>> equality assertion failed
>>> - Expected: manolo
>>> - Actual  : Manolo
>>> - In <file:///C:/cygwin64/tmp/lu23416a8g0y3.tmp>, XPath contents of 
>>> child does not match
>>>
>>> Failures !!!
>>> Run: 85   Failure total: 1   Failures: 1   Errors: 0
>>
>> This test succeeds normally though, when run individually:
>>
>>> $ /opt/lo/bin/make CppunitTest_sw_ooxmlexport4 
>>> CPPUNIT_TEST_NAME=testComboBoxControl
>>
>> I was able to narrow down the problem to --disable-pch in my config.
> 
> I had played with the includes from PCH, and could find out that 
> including <boost/iterator/iterator_concepts.hpp> in the beginning of 
> sw/qa/extras/ooxmlexport/ooxmlexport4.cxx allows the test to pass. I 
> couldn't limit the includes further: only including what's included 
> inside that boost header does not allow the test to pass. But I don't 
> understand what does that mean (some ODR violation? some missing 
> virtual overridden function?...), and don't see where to look next. 
> Does someone have any ideas?

I think the issue is in what relative order the three individual tests

> DECLARE_OOXMLEXPORT_EXPORTONLY_TEST(testComboBoxControl, 
> "combobox-control.docx") {
>     // check XML
>     xmlDocUniquePtr pXmlDoc = parseExport("word/document.xml");
>     assertXPath(pXmlDoc,
"/w:document/w:body/w:p/w:sdt/w:sdtPr/w:dropDownList/w:listItem[1]",
"value", "manolo");
>     assertXPath(pXmlDoc,
"/w:document/w:body/w:p/w:sdt/w:sdtPr/w:dropDownList/w:listItem[2]",
"value", "pepito");
>     assertXPathContent(pXmlDoc, 
> "/w:document/w:body/w:p/w:sdt/w:sdtContent/w:r/w:t", "manolo");
> 
>     // check imported control
>     verifyComboBoxExport(getShapes() == 0); }
> 
> CPPUNIT_TEST_FIXTURE(Test, tdf134043_ImportComboBoxAsDropDown_true)
> {
>     std::shared_ptr< comphelper::ConfigurationChanges >
batch(comphelper::ConfigurationChanges::create());
>
officecfg::Office::Writer::Filter::Import::DOCX::ImportComboBoxAsDropDown::s
et(true, batch);
>     batch->commit();
> 
>     load(mpTestDocumentPath, "combobox-control.docx");
>     verifyComboBoxExport(true);
> }
> 
> CPPUNIT_TEST_FIXTURE(Test, tdf134043_ImportComboBoxAsDropDown_false)
> {
>     std::shared_ptr< comphelper::ConfigurationChanges >
batch(comphelper::ConfigurationChanges::create());
>
officecfg::Office::Writer::Filter::Import::DOCX::ImportComboBoxAsDropDown::s
et(false, batch);
>     batch->commit();
> 
>     load(mpTestDocumentPath, "combobox-control.docx");
>     verifyComboBoxExport(false);
> }

(that were added to sw/qa/extras/ooxmlexport/ooxmlexport4.cxx with
<https://git.libreoffice.org/core/+/7b00fbddfb1cd55a68ed7481ebd4a5d5f60c6128
%5E!/>
"tdf#134043 DOCX import: new unit tests: ComboBox to DropDown") are
executed:

With just --disable-pch, the order for me happens to be 2, 3, 1; while with
an additional

> diff --git a/sw/qa/extras/ooxmlexport/ooxmlexport4.cxx 
> b/sw/qa/extras/ooxmlexport/ooxmlexport4.cxx
> index 7fe92c72eede..5fad8bd08bdf 100644
> --- a/sw/qa/extras/ooxmlexport/ooxmlexport4.cxx
> +++ b/sw/qa/extras/ooxmlexport/ooxmlexport4.cxx
> @@ -7,6 +7,7 @@
>   * file, You can obtain one at http://mozilla.org/MPL/2.0/.
>   */
>  
> +#include <boost/iterator/iterator_concepts.hpp>
>  #include <swmodeltestbase.hxx>
>  
>  #include <com/sun/star/awt/XBitmap.hpp>

the order happens to change to 2, 1, 3.  Note how tests 2 and 3 modify
global state, which presumably makes a different for the (failing) test 1.

Serge, can you see to get that fixed?

------------------------------

Message: 8
Date: Wed, 22 Jul 2020 10:24:00 +0200
From: Miklos Vajna <vmiklos@xxxxxxxxxxxxx>
To: Stephan Bergmann <sbergman@xxxxxxxxxx>
Cc: libreoffice@xxxxxxxxxxxxxxxxxxxxx, Serge Krot <Serge.Krot@xxxxxx>
Subject: Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
	Windows
Message-ID: <20200722082359.GU5941@xxxxxxxxxxxxx>
Content-Type: text/plain; charset=utf-8

Hi,

On Wed, Jul 22, 2020 at 10:18:44AM +0200, Stephan Bergmann
<sbergman@xxxxxxxxxx> wrote:
> the order happens to change to 2, 1, 3.  Note how tests 2 and 3 modify 
> global state, which presumably makes a different for the (failing) test 1.
> 
> Serge, can you see to get that fixed?

There is a Resetter / preTest() mechanism at
sw/qa/inc/swmodeltestbase.hxx:267, perhaps you can use that to automatically
undo your changes to the global state after the test.

Regards,

Miklos

------------------------------

Message: 9
Date: Wed, 22 Jul 2020 11:27:09 +0300
From: Mike Kaganski <mikekaganski@xxxxxxxxxxx>
To: Stephan Bergmann <sbergman@xxxxxxxxxx>,
	libreoffice@xxxxxxxxxxxxxxxxxxxxx
Cc: Serge Krot <Serge.Krot@xxxxxx>
Subject: Re: CppunitTest_sw_ooxmlexport4 failing with --disable-pch on
	Windows
Message-ID:

<DB6PR07MB3512A253ED1DD1C15488D9FDD6790@xxxxxxxxxxxxxxxxxxxxxxxxxxx.outlook.
com>

Content-Type: text/plain; charset="utf-8"

Hi,

On 22.07.2020 11:18, Stephan Bergmann wrote:
> I think the issue is in what relative order the three individual tests

It still seems odd to me that something like this may happen, dependent on
headers included. Something that makes compiling non-deterministic...
sigh.

--
Best regards,
Mike Kaganski
-------------- next part --------------
A non-text attachment was scrubbed...
Name: pEpkey.asc
Type: application/pgp-keys
Size: 1769 bytes
Desc: not available
URL:
<https://lists.freedesktop.org/archives/libreoffice/attachments/20200722/10b
d9667/attachment-0001.key>

------------------------------

Subject: Digest Footer

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

------------------------------

End of LibreOffice Digest, Vol 119, Issue 23
********************************************

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