[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-21 Thread Dwayne Bailey


On 2011-04-20 15:46, Rimas Kudelis wrote:


Hi,

2011.04.20 16:37, Dwayne Bailey rašė:
I'd rather see us spending some time to create a way to inject and 
store arbitrary errors for PO files stored on Pootle. That way anyone 
can create a script that identifies problems, injects them into 
Pootle and allows translators to fix them quickly on Pootle in an 
environment that is familiar and safe.


What about automatically marking invalid translations as fuzzy in the 
.po file and adding a translator comment above them? This would be 
generic enough, and would not depend on Pootle at all.
It's a hack.  It would mean that people writing checks must work on PO 
files, the checks I'm referring to are working directly on SDF files.  
You could script pogrep to find these, mark them fuzzy and inject them 
into the original PO files - but still a hack in my mind.  Not all 
failing checks should be fuzzy, I reserve that for checks that break things.


The checks in Pootle are not actually written in Pootle but in the 
Translate Toolkit which is why you can run them against your PO files 
outside of Pootle.  We already use a scheme for marking these when we 
use pofilter on the command line.  Maybe that is an approach that should 
be examined, but it would still mean changing Pootle to read these 
comments when it finds them in a PO file.


Marking something fuzzy doesn't push it into a class of error with a 
specific agenda to fix.  So I can't go through errors of type X.  I will 
get all my fuzzy strings together in one lump with only a comment (which 
I must also now delete) to distinguish it from a proper fuzzy.


It has been my aim to make sure these checks cover gsicheck 100% as then 
for a Pootle user Pootle becomes the right place to fix these errors.  
They then don't need to decipher gsicheck output which is difficult.  
Pootle has a special openoffice checker that it gets from the Translate 
Tolkit so it is possible to write anything that is needed and then also 
doesn't depend on Pootle.


So either the checks are written to be a proper Translate Toolkit check 
so that it can show directly in Pootle, ideal (Unfortunately we haven't 
had any Oracle devs contribute changes to the checks so don't hold your 
breath). Or we have some way for outside checks to report their errors 
to Pootle and mark units for review.


Programmers will do anything as quickly as they can.  Us localisers 
accept what we get which is why we're reading SDF snippets and strange 
error output.  We spend quite a bit of time interpreting the results, we 
don't complain.  I think there is a better solution that would mean that 
we can translate instead of pretend we're programmers as we interpret 
SDF output.


--

regards
Dwayne

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-20 Thread Rimas Kudelis

Hi,

2011.04.20 16:37, Dwayne Bailey rašė:
I'd rather see us spending some time to create a way to inject and 
store arbitrary errors for PO files stored on Pootle. That way anyone 
can create a script that identifies problems, injects them into Pootle 
and allows translators to fix them quickly on Pootle in an environment 
that is familiar and safe.


What about automatically marking invalid translations as fuzzy in the 
.po file and adding a translator comment above them? This would be 
generic enough, and would not depend on Pootle at all.


Rimas

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-20 Thread Dwayne Bailey


Hi Eike,

On 2011-04-15 14:24, Eike Rathke wrote:

Hi Dwayne,

On Friday, 2011-04-15 12:43:45 +0200, Dwayne Bailey wrote:


Instead of reinventing the wheel.  Why not use poconflicts?
http://translate.sourceforge.net/wiki/toolkit/poconflicts - which
detects these kinds of different translations.

That's a solution for a different problem. We don't want to detect
different translations of a term, but identical translations of
different terms within a given set. For example Calc function names, if
both SEARCH and FIND are translated identical it breaks functionality.
As Friedel said poconflicts does do the inverse by using the -v or 
inverse option.  What it doesn't do is limit the checks to a subset and 
it doesn't integrate easily with Pootle.


Martin raises a concern that the output is unreadable to a human being, 
I experienced the same issue when trying to identify the errors for ZA 
languages.  This is my reason for wanting to see these types of scripts 
more tightly coupled to Pootle itself.  Or for now at least to turn the 
errors into links to the actual unit on Pootle so that a translator can 
easily take action.


I'd rather see us spending some time to create a way to inject and store 
arbitrary errors for PO files stored on Pootle.  That way anyone can 
create a script that identifies problems, injects them into Pootle and 
allows translators to fix them quickly on Pootle in an environment that 
is familiar and safe.

Btw, your message had to be moderated through, I thought you were
subscribed to the list, isn't that the case?

I am on the ML - perhaps I look like a spammer?

--
regards
Dwayne

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-16 Thread Jordi Serratosa

Hi Rafaella,
Catalan (ca) fixed in Pootle.

salut
jordi s

El 15/04/2011 10:58, en/na Rafaella Braconi ha escrit:

Dear All,

In OOo there are some files where double names in the translation 
impact the functionality of the product. This is for example the case 
in the Calc funtions and in the Styles. As these bugs are not so easy 
to detect, we are introducing an automatic test script that runs on 
the resources every time we create the builds.


There are a number of languages in which such duplicates have been 
detected. Can you please have a look at the log file attached to issue 
http://openoffice.org/bugzilla/show_bug.cgi?id=112478 and have the 
double names fixed in Pootle (or in case you don't use Pootle in the 
sdf files you deliver) ?.


Please add a note once you have fixed the issues.

Thanks in advance,
Rafaella

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-16 Thread F Wolff

Op Vr, 2011-04-15 om 14:24 +0200 skryf Eike Rathke:
> Hi Dwayne,
> 
> On Friday, 2011-04-15 12:43:45 +0200, Dwayne Bailey wrote:
> 
> > Instead of reinventing the wheel.  Why not use poconflicts?
> > http://translate.sourceforge.net/wiki/toolkit/poconflicts - which
> > detects these kinds of different translations.
> 
> That's a solution for a different problem. We don't want to detect
> different translations of a term, but identical translations of
> different terms within a given set. For example Calc function names, if
> both SEARCH and FIND are translated identical it breaks functionality.


Poconflict checks for both. Dwayne has been doing these checks on our
translations for something like 6 years, and I think has demonstrated
them at OOo conferences.

Friedel


--
Recently on my blog:
http://translate.org.za/blogs/friedel/en/content/better-lies-about-gnome-localisation

-- 
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-15 Thread Rafaella Braconi

Hi Martin,

On 15.04.2011 11:47, Martin Srebotnjak wrote:

Hi, Rafaella,

looking at the results for Slovenian (sl) I see 1 error, but do not 
understand it (besides that some tests report: "failed", so are there 
additional errors not explained?).


Here is a snippet from the Slovenian part of the report, please 
explain the error to me as I do not understand it:


ooo/l10ntools/scripts/doublecheck.sh
10 strings found in sl/localize.sdf for -m sw -f 
source.ui.utlui.poolfmt.src -g STR_POOLNUMRULE_.* -l .*

Entry double for LANG=sl in file sl/localize.sdf
swsource\ui\utlui\poolfmt.src0string
STR_POOLCOLL_HEADLINE_BASE0slNaslov
20110324 13:31:50
swsource\ui\utlui\poolfmt.src0string
STR_POOLCOLL_DOC_TITEL0slNaslov
20110324 13:31:50
For the 2 above strings it means that the Slovenian translation is using 
the same term for 2 different English terms. This is linguistically 
correct but this impacts functionality. So you will need to translate 
for example Title, Heading and Header within the sw
source\ui\utlui\poolfmt.src file with 3 different terms.
122 strings found in sl/localize.sdf for -m sw -f 
source.ui.utlui.poolfmt.src -g STR_POOLCOLL_.* -l .*

failed

cc:ing here Gregor who may better explain this result.


It would help if the test would also list the original English entry. 
In this case the problem is because both strings are the same? What if 
in a given language two different English terms use same word? What is 
the original English text (I should probably check the Slovenian sdf).

Yes, in fact I asked Gregor to
a) provide sdf files 1 per language containing the english source as well
b) to store such sdf with the errors to fix at 
ftp://qa-upload.services.openoffice.org/l10n/download/in oder to better 
relate errors with milestone language packs.



Thanks to you!
Rafaella


Thanks,
m.

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-15 Thread Eike Rathke
Hi Dwayne,

On Friday, 2011-04-15 12:43:45 +0200, Dwayne Bailey wrote:

> Instead of reinventing the wheel.  Why not use poconflicts?
> http://translate.sourceforge.net/wiki/toolkit/poconflicts - which
> detects these kinds of different translations.

That's a solution for a different problem. We don't want to detect
different translations of a term, but identical translations of
different terms within a given set. For example Calc function names, if
both SEARCH and FIND are translated identical it breaks functionality.

Btw, your message had to be moderated through, I thought you were
subscribed to the list, isn't that the case?

  Eike

-- 
 OOo Calc core developer. Number formatter stricken i18n transpositionizer.
 Signature key 0x87F8D412 : 2F58 5236 DB02 F335 8304  7D6C 65C9 F9B5 87F8 D412
 OpenOffice.org Engineering at Oracle: http://blogs.sun.com/GullFOSS


pgp92uWK2gq7w.pgp
Description: PGP signature


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-15 Thread Dwayne Bailey


On 2011-04-15 10:58, Rafaella Braconi wrote:


Dear All,

In OOo there are some files where double names in the translation 
impact the functionality of the product. This is for example the case 
in the Calc funtions and in the Styles. As these bugs are not so easy 
to detect, we are introducing an automatic test script that runs on 
the resources every time we create the builds.


There are a number of languages in which such duplicates have been 
detected. Can you please have a look at the log file attached to issue 
http://openoffice.org/bugzilla/show_bug.cgi?id=112478 and have the 
double names fixed in Pootle (or in case you don't use Pootle in the 
sdf files you deliver) ?.
Instead of reinventing the wheel.  Why not use poconflicts? 
http://translate.sourceforge.net/wiki/toolkit/poconflicts - which 
detects these kinds of different translations.


--
regards
Dwayne

--
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-15 Thread Martin Srebotnjak
Hi, Rafaella,

looking at the results for Slovenian (sl) I see 1 error, but do not
understand it (besides that some tests report: "failed", so are there
additional errors not explained?).

Here is a snippet from the Slovenian part of the report, please explain the
error to me as I do not understand it:

ooo/l10ntools/scripts/doublecheck.sh
10 strings found in sl/localize.sdf for -m sw -f source.ui.utlui.poolfmt.src
-g STR_POOLNUMRULE_.* -l .*
Entry double for LANG=sl in file sl/localize.sdf
swsource\ui\utlui\poolfmt.src0string
STR_POOLCOLL_HEADLINE_BASE0slNaslov
20110324 13:31:50
swsource\ui\utlui\poolfmt.src0string
STR_POOLCOLL_DOC_TITEL0slNaslov
20110324 13:31:50
122 strings found in sl/localize.sdf for -m sw -f
source.ui.utlui.poolfmt.src -g STR_POOLCOLL_.* -l .*
failed

It would help if the test would also list the original English entry. In
this case the problem is because both strings are the same? What if in a
given language two different English terms use same word? What is the
original English text (I should probably check the Slovenian sdf).

Thanks,
m.
-- 
-
To unsubscribe send email to [email protected]
For additional commands send email to [email protected]
with Subject: help


[l10n-dev] Re: IMPORTANT: pls fix double names

2011-04-15 Thread Eike Rathke
Hi Rafaella,

On Friday, 2011-04-15 10:58:30 +0200, Rafaella Braconi wrote:

> As these bugs are not
> so easy to detect, we are introducing an automatic test script that
> runs on the resources every time we create the builds.

Great, muchas gracias!

  Eike

-- 
 OOo Calc core developer. Number formatter stricken i18n transpositionizer.
 Signature key 0x87F8D412 : 2F58 5236 DB02 F335 8304  7D6C 65C9 F9B5 87F8 D412
 OpenOffice.org Engineering at Oracle: http://blogs.sun.com/GullFOSS


pgpfrMJ5w2gOJ.pgp
Description: PGP signature