gmane.editors.abiword.devel

Re: Mechanism for proposing Abisource web-site changes/corrections

J.M. Maurer — 2012-05-24T21:38:29



Please do: you have full abisource.com modification permissions now.

Welcome, our new web master :P
  Marc

Re: Mechanism for proposing Abisource web-site changes/corrections

Chris Leonard — 2012-05-24T19:49:44

On Thu, May 24, 2012 at 3:29 PM, J.M. Maurer <uwog< at >uwog.net> wrote:


Works fine for me.  I'll be sure to preview any non-trivial changes.

I would like to add mention of the Pootle hosted L10n workflow on the
L10n pages.

cjl

Re: Mechanism for proposing Abisource web-site changes/corrections

J.M. Maurer — 2012-05-24T19:29:07

We have the website in a seperate repository for some historical reason:

 svn+ssh://<username>< at >svn.abisource.com/svnweb/www.abisource.com/trunk
  
You need separate access to this repo. I've just given it to you,
without checking if it works. Could you try it? If it does not work,
then please poke hub, as I'll be mostly offline the coming days.

Cheers,
  Marc




On Sat, 2012-05-19 at 03:02 -0400, Chris Leonard wrote:

Re: Python binding documentation is missing

J.M. Maurer — 2012-05-24T19:26:23

On Thu, 2012-05-17 at 18:55 -0300, Manuel Kaufmann wrote:

The correct answer is indeed "no". Some small tweaks would be needed in
the library. I'm low on time, but maybe you can poke someone from the
devs to add it.

Sorry,
  Marc

AbiWord 2.9.x release instructions

J.M. Maurer — 2012-05-24T19:24:06

Ola everyone!

Usually I'm the one doing releases, but since I'm about to have a baby
to care for, I suspect I won't have a lot of time the coming months to
release 2.9.3 and beyond :)

Therefor I've written down all release instructions as detailed as
possible, so everyone with both svn and svnweb[1][2] accounts can do
releases:

  http://abisource.com/wiki/Releasing#Building_a_Windows_Installer

Please update them as needed, so everyone can always do releases!

Cheers!
  Marc

[1] svnweb is a seperate svn repository that contains our website
related sources. I would be in favor of merging this with the normal svn
repository, so everyone with svn commit rights to our source repository
can do a release.

[2] If you want access to the svnweb repository, then please let hub or
me know. We can give you access. In the future you don't need these
silly additional permissions anymore (see point [1]).

New AbiWord strings

Chris Leonard — 2012-05-24T15:57:36

Dear AbiWord localizers,

Some new strings have been added to the latest AbiWord POT, and the
projects have been updated on Pootle

http://translate.sugarlabs.org/projects/AbiWord/

A number of these are added language names.

As the localized name of Language X in Language Y can sometimes be
tricky (especially if Language X is an uncommon lang). Two of my
favorite tricks for finding these are:

1) Look up Lang X in the Lang Y localized ISO-639 PO file.

http://anonscm.debian.org/gitweb/?p=iso-codes/iso-codes.git;a=tree;f=iso_639;h=05181f96c87c5862af51932353af095ca9d0efc8;hb=HEAD

2) Look up Lang X article in English wikipedia, then see of there is a
link to a Lang Y article (with the localized name).

Some people are not all that familiar with the context/meaning for
some of the more recent strings about RDF.

It might help to read this article.

http://en.wikipedia.org/wiki/Resource_Description_Framework

RDF is a sort of mark-up language that allows you to add meta-data to
documents that enables the use of what are called "semantic web"
technologies.

http://en.wikipedia.org/wiki/Semantic_Web

Hopefully that background will help you localize those strings
properly. I'm keeping a close eye on Pootle, so if you complete the
PO file, I will probably see it and commit it to SVN, but if you want
to, feel free to ping me a message that you're ready for it to be
committed.

Marc has created the SVN user Poolte, but I want to be really cautious
about rolling out direct commits from Pootle to SVN and I've not had
the needed time for careful testing, but we will get there in due
time. In the meantime, I hope those of you using Poolte have
appreciated the fast turnaround on commits of completed files there.

cjl

RE: Bug #13355

Jean Brefort — 2012-05-24T14:31:50

Not sure it is a gtk bug, it might be a new feature as well ;-)

Anyway, I don't see why dblBufferingObj should be moved inside the while
loop. The issue is not there, but after, when we call the notifyers.
I'll track this more deeply next week if time permits.

Cheers,
Jean

Le jeudi 24 mai 2012 à 01:48 +0000, Martin Edmund Sevior a écrit :

RE: Bug #13355

Ingo Brückl — 2012-05-24T06:27:51

Martin Edmund Sevior wrote on Thu, 24 May 2012 01:48:45 +0000:


What about

  #if !GTK_CHECK_VERSION(3,4,0)
          FV_ViewDoubleBuffering dblBufferingObj(this, false, false);
          dblBufferingObj.beginDoubleBuffering();

          while ((!bDone || bIsDirty) && pCurRun)
          {
  #else
          while ((!bDone || bIsDirty) && pCurRun)
          {
                 FV_ViewDoubleBuffering dblBufferingObj(this, false, false);
                  dblBufferingObj.beginDoubleBuffering();
  #endif

then, not changing the code for all non Fedora 16 users?

Ingo

RE: Bug #13355

Martin Edmund Sevior — 2012-05-24T01:48:45

Hi Jean,

If my patch never triggers an assert even though you get a crash, it is a sign that the bug is in gtk3. While I'm in favour of a work around that prevents a crash, it is important that the bug in gtk be fixed. Therefore 
 please commit your patch with the following big comment in capital letters and let the gtk developers that we've exposed a crash bug that is present in gtk-3.4 but not gtk-.3.2.4 (distributed in Fedora 16).

//
// WORK AROUND FOR BUG 13355 CAUSED BY A BUG IN GTK-3.4. PLEASE TEST 
// AND REVERT TO EARLIER CODE ONCE THIS IS 
// FIXED. 
//
#if 0
FV_ViewDoubleBuffering dblBufferingObj(this, false, false);
dblBufferingObj.beginDoubleBuffering();

while ((!bDone || bIsDirty) && pCurRun)
{
#eise
while ((!bDone || bIsDirty) && pCurRun)
{
       FV_ViewDoubleBuffering dblBufferingObj(this, false, false);
        dblBufferingObj.beginDoubleBuffering();
#endif

Cheers

Martin
________________________________________
From: Jean Brefort [jean.brefort< at >normalesup.org]
Sent: Tuesday, May 22, 2012 7:33 PM
To: Martin Edmund Sevior
Subject: RE: Bug #13355

Hi Martin,

Your patch did not help. Attached one fixes the issue, it just calls
notifyListener() after ending the double buffering operation. The patch
looks large, it actually just add a block (two lines added) and some
tabulations between the twon new braces. I might investigate more to
find out which listener destroys the cairo surface and how.

Cheers,
Jean

Le lundi 21 mai 2012 à 19:33 +0000, Martin Edmund Sevior a écrit :

Re: L10n update

Kathiravelu Pradeeban — 2012-05-22T20:18:30

Impressive work Chris.
I hope this effort must have helped our localization teams and experts a lot.

Thank you.
Regards,
Pradeeban.

On Sun, May 20, 2012 at 8:41 AM, Chris Leonard <cjlhomeaddress< at >gmail.com> wrote:




--
Kathiravelu Pradeeban.
Blog: [Llovizna] http://kkpradeeban.blogspot.com/

Re: paragraph style

Martin Sevior — 2012-05-20T23:47:18

Hi Simon,

I agree with your proposal. It preserves the users intentions and
helps clear out messes. BTW I am the MES in the discussion :-)

Cheers

Martin


On Mon, May 21, 2012 at 2:33 AM, Simon Larochelle
<larochelle.simon.1< at >gmail.com> wrote:

paragraph style

Simon Larochelle — 2012-05-20T16:33:30

I am trying to correct a bug when the user changes paragraph style.
Presently, all font properties and character styles are removed from
the paragraph until the first footnote (or endnote), while after the
footnote, none of these things are cleared. There are a few issues:

1) the behavior needs to be same before and after the footnote
2) clearing character styles affects the font properties of footnote
references (i.e. they are no longer in superscript)
3) clearing font properties is a mix bag. If the user has a word in
italic or in bold, he likely does not want it to be affected by the
paragraph style change. However, abiword tends to explicitly set a
large number of properties that are defined in the paragraph style
(for example if the user sets the font to bold, type a word, and then
unsets it, a font property font-weight:normal will be added to every
text fragment afterward). These extraneous properties will likely
cause undesirable effects during the paragraph style change (that is,
keep a number of the old paragraph style properties).

I would like to implement the following algorithm:
1) clear all the font properties that were not necessary with the old
paragraph style
2) do not touch character style
3) and change the block strux style attribute to its new value.

The reason I am putting that on the mailing list is that I found the
following unresolved discussion in file pt_PT_ChangeStrux.cpp. Does
anybody object to my plan?


// when applying a block-level style, we also need to clear
// any props at the frag level, which might trigger coalescing,
// thus this version of the loop is more complex.

//
// OK for styles we expand out all defined properties including BasedOn styles
// Then we use these to eliminate any specfic properties in the current strux
// Then properties in the current strux will resolve to those defined in the
// style (they exist there) to specifc values in strux (if not overridden by
// the style) then finally to default value.
//
// TODO this is not right; first of all, paragraph style should be applied
// to the block Strux only and nothing else -- no Spans, Fmt marks, etc.
//Second, when applying paragraph style, we should clear the existing
//strux of all its properties inherited from any previous style
//not just the ones defined explicitely, by this style, because what
//is not defined is assumed to default, not to be inherited from a style
//we are trying to get rid off.
//
// NO. We want to remove all character level properties that clash
with properties
// defined in th strux level style. -MES
//



Simon

L10n update

Chris Leonard — 2012-05-20T03:11:27

Dear AbiWo4rd localizers

All of the PO files for 2.9 have had their headers adjusted and
re-committed to SVN.

Please make sure you download a fresh copy of the PO files (rather
than using a local copy) if you are going to work on L10n.  I would
like to preserve the header adjustments that have been made and not
overwrite them with the old headers, if possible.

Ideally, please work on the L10n online via the Pootle user interface.

http://translate.sugarlabs.org/projects/AbiWord/

Everything there is now synchronized with SVN (except German I think
because Ingo has been committing directly).

I will be working on getting the two way connection between SVN and
Poolte established.

SVN > Poolte (updated POT files)
Poolte > SVN (individual PO files)

I will update you as progress is made.

cjl

Re: Mechanism for proposing Abisource web-site changes/corrections

Besnik Bleta — 2012-05-19T13:36:06


Sometimes ago a discussion about making the site multilingual was 
started. Any news?

cheers
Besnik


Më 05/19/2012 10:02 AM, Chris Leonard shkrojti:

Mechanism for proposing Abisource web-site changes/corrections

Chris Leonard — 2012-05-19T07:02:39

Dear AbiWord devs,

I don't see the HTML for the Abisource website

http://www.abisource.com/

as a folder in the SVN

 http://svn.abisource.com/

There are some errors I would like to fix ((for example) updating

Development
(2.7.x)

to

Development
(2.9.x)

at top of table on this page:
http://www.abisource.com/contribute/translate/exist/

If the HTML were checked into SVN, it would be relatively
straightforward to submit patches that could be reviewed easily for
acceptance and publication on the website by whoever is currently
acting as webmaster< at >abisource.com.

cjl

Re: Python binding documentation is missing

Manuel Kaufmann — 2012-05-17T21:55:26

On Thu, May 17, 2012 at 5:52 PM, J.M. Maurer <uwog< at >uwog.net> wrote:

Is there a way to call it from the Python binding? Guessing: "no"

I think I have to patch the library to do this, and I'm in the same
case that in the last email. I would be very happy if I don't have to
do that :)

Re: Python binding documentation is missing

Manuel Kaufmann — 2012-05-17T21:53:29

On Thu, May 17, 2012 at 5:45 PM, J.M. Maurer <uwog< at >uwog.net> wrote:

Well, it seems that I'm not as good as you at reading C++ source code :)


I think is not a good idea having a patched version of a library on
our releases. Things like this will be happening all the time.

I'm not sure what decision will we take about this. But I'm really
sure that we don't want that dialog at exporting HTML files... :(

Re: Python binding documentation is missing

J.M. Maurer — 2012-05-17T20:52:29



The good thing is that during last year's GSoC, the HTML exporter gained
a nice function called IE_Exp_HTML.suppressDialog():

  http://svn.abisource.com/abiword/trunk/src/wp/impexp/xp/ie_exp_HTML.h

So it should be fairly trivial to call that in the widget when a user is
exporting to HTML.

Cheers,
   Marc

Re: Python binding documentation is missing

J.M. Maurer — 2012-05-17T20:45:51

On Thu, 2012-05-17 at 12:45 -0300, Manuel Kaufmann wrote:

When I wrote Write, I never really had documentation myself. I always
just looked at the canvas widget header file, which usually was "good
enough":

  http://svn.abisource.com/abiword/trunk/src/wp/ap/gtk/abiwidget.h


You won't be able to solve that problem without extending the widget
API. They way I solved this when the first XO's shipped, was by
distributing some patches in the RPM that simply disabled the HTML
exporter dialog altogether :P

I think I also added those patches to the official Fedora abiword
package, which at the time contained a flag to do a "suger build".
Someone ripped that out a while ago, without ever really solving the
original issue you are now running into as well :'(

Cheers,
  Marc

Python binding documentation is missing

Manuel Kaufmann — 2012-05-17T15:45:55

Hello guys,

I'm working on Write Activity from Sugar (OLPC Project) that uses
Python's Abiword bindings and I'm trying to find a useful
documentation about this.

I was taking a look at the wiki but I didn't find the explanation
about the functions. I found these pages but they are not useful to
me:
 * http://www.abisource.com/wiki/PyAbiWord
 * http://www.abisource.com/wiki/AbiWidget_Header
 * http://svn.abisource.com/pyabiword/trunk/examples

"pradeeban" was helping me a lot on the IRC #abiword channel, and he
told me about this mailing list. So, I'm going to explain what's my
specific problem.

I'm trying to export to HTML the abiword document using
abiword.Canvas.save() method passing this options:

abi.save('file://' + fileObject.file_path, format['mime_type'], exp_props)

abi: abiword.Canvas
fileObject.file_path: "/home/olpc/help.html"
format['mime_type']: "text/html"
exp_props: "'html4:yes; declare-xml:no; embed-css:yes; embed-images:yes;"

The problem that I have[1] is I don't want to show the pop-up dialog
that ask to the user about more specific settings on the export.

Thanks, I will appreciate your help.

[1] http://bugs.sugarlabs.org/ticket/3610

Re: [PATCH] Automatize creation of DATA lists for Makefiles

Chris Leonard — 2012-05-17T14:51:49

On Thu, May 17, 2012 at 10:38 AM, Hubert Figuière <hfiguiere< at >teaser.fr> wrote:

Yes, that occurred to me after I sent the message.

cjl