openoffice-l10n mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From janI <j...@apache.org>
Subject Re: Single Printout
Date Sat, 15 Jun 2013 11:45:22 GMT
Hi

I forgot to write.

Have anybody told you about irc ?

We have a channel there #dev.openoffice.org on freenode. The channel (one
of many asf channels) allows us to chat directly which is often faster. I
am on there 24/7 with the nick janIV (not always in person, but a server
collect messages for me).

rgds
jan I.


On 15 June 2013 13:36, janI <jani@apache.org> wrote:

>
>
>
> On 15 June 2013 10:35, Kadal Amutham <vkadal@gmail.com> wrote:
>
>> Jani, Could you able to access the database and the tables in pootle, as
>> one could do it in MySQL?
>>
>
> I have access to the mysql db, but I cannot provide that for you (at least
> at the moment), this is a shared server between all asf projects and in
> order to have mysql access you would need access to the server and be able
> to do sudo.
>
> The database is controlled by pootle, and we are very carefull when using
> mysql directly, actually we only do backup, and I do db optimization.
>
>
>> I checked the size of Tamil pootle files it is 5.7 MB. Assume 10 MB. There
>> are not more than 40 languages currently in pootle. So the DB size should
>> not be more than 400MB? How it shows 15GB? Actually it should be much less
>> than 400MB. The original English text data will not be repeated 40 times,
>> to get 400MB. So where is the catch?
>>
>
> I am not sure why you expect the mysql tables to have anywhere the same
> size as the simple po files. the mysql tables contains all kind of key
> references, historical data, information about who did what.
>
> I dont think there is a catch, try to load the po files in simple tables
> and you will normally find that the tables are much larger than the
> original text files.
>
> I you want to learn more about how pootle structures data, then download
> it and install it, having done that you can  upload your files on your
> local version, and play with that.
>
>
>> There is one law. The name I forgot
>>
>> The work extends till the time available
>> The data expands to fill the hard disk space
>>
>
> Yeah, you are right, but dont forget the other law:
>     The more you compress data, the slower a system you get.
>
> rgds
> jan I.
>
>>
>> With Warm Regards
>>
>> V.Kadal Amutham
>> 919444360480
>> 914422396480
>>
>>
>> On 15 June 2013 13:37, Kadal Amutham <vkadal@gmail.com> wrote:
>>
>> > It is a good suggestion. I checked that pootle is open source, and I
>> will
>> > install, I can download Tamil translation from AOO server and upload it
>> to
>> > my pootle. Let us see whether I can do anything worthwhile
>> >
>> > With Warm Regards
>> >
>> > V.Kadal Amutham
>> > 919444360480
>> > 914422396480
>> >
>> >
>> > On 15 June 2013 12:47, janI <jani@apache.org> wrote:
>> >
>> >> On Jun 15, 2013 6:55 AM, "Kadal Amutham" <vkadal@gmail.com> wrote:
>> >> >
>> >> > It is a great news that it is in MySQL. Can I have the copy of the
>> >> database
>> >> > / table to study further?
>> >>
>> >> may I suggest you start by installing pootle somewhere and get to know
>> the
>> >> db structure first. Just dumping about 15gb data will not be vey
>> helpfull
>> >> to you.
>> >>
>> >> if you install pootle you can experiment on a small db, create the
>> script
>> >> you ant which we then can run on the server.
>> >>
>> >> rgds
>> >> jan i
>> >>
>> >> >
>> >> > With Warm Regards
>> >> >
>> >> > V.Kadal Amutham
>> >> > 919444360480
>> >> > 914422396480
>> >> >
>> >> >
>> >> > On 15 June 2013 03:59, janI <jani@apache.org> wrote:
>> >> >
>> >> > > On Jun 14, 2013 8:48 PM, "Kadal Amutham" <vkadal@apache.org>
>> wrote:
>> >> > > >
>> >> > > > Pootle server may be sitting on a Database server like MySQL.
>> >> Probably
>> >> > > > there shall be single database, and each language shall have
a
>> >> table.
>> >> If
>> >> > > I
>> >> > > > could get a copy of that table, then I can upload to any
MySQL
>> >> server
>> >> and
>> >> > > > analyse further. I can take any print out as I like.
>> >> > >
>> >> > > pootle uses mysql, but each is not so that each language has a
>> simple
>> >> > > table. sentences/words/terms are split across several tables,
which
>> >> also
>> >> > > contain deleted entries and to some extent revisions.
>> >> > >
>> >> > > rgds
>> >> > > jan i
>> >> > >
>> >> > > >
>> >> > > > Kindly let me know whether my assumption is correct.
>> >> > > >
>> >> > > >
>> >> > > > On 14 June 2013 16:56, Andrea Pescetti <pescetti@apache.org>
>> wrote:
>> >> > > >
>> >> > > > > Andre Fischer wrote:
>> >> > > > >
>> >> > > > >> http://wiki.openoffice.org/**wiki/Mnemonics_Localisation#**
>> >> > > > >>>> How_the_mnemonics_are_assigned<
>> >> > >
>> >> > >
>> >>
>> >>
>> http://wiki.openoffice.org/wiki/Mnemonics_Localisation#How_the_mnemonics_are_assigned
>> >> > > >
>> >> > > > >>>>
>> >> > > > >>> Is this "no accelerators" rule still active?
Even in
>> languages
>> >> where
>> >> > > > >> automatic selection of accelerators work, can that
not break
>> >> > > > >> documentation? I mean, can adding a new string lead
to all
>> >> > > accelerators
>> >> > > > >> being redefined?
>> >> > > > >>
>> >> > > > >
>> >> > > > > As far as I know, documentation is not mentioning accelerators
>> in
>> >> > > general,
>> >> > > > > but only shortcuts (CTRL-S and similar).
>> >> > > > >
>> >> > > > >
>> >> > > > >  And there is point 3 in section 1.1 in the
>> Mnemonics_Localisation
>> >> page
>> >> > > > >> mentioned above:
>> >> > > > >> "The bugs are fixed manually in Web-LingTool by
placing the
>> '~'
>> >> > > > >> character to a different position"
>> >> > > > >> Removing all explicitly defined accelerators would
remove such
>> >> bug
>> >> > > fixes.
>> >> > > > >>
>> >> > > > >
>> >> > > > > The guidelines in
>> http://openoffice.apache.org/**translate.html<
>> >> > > http://openoffice.apache.org/translate.html>item 3 only mention
>> >> "newly
>> >> > > translating strings": existing accelerators can
>> >> > > > > be left as they ae, they won't break anything (except,
>> possibly,
>> >> making
>> >> > > > > more difficult some automated matches in translation
>> management).
>> >> > > > >
>> >> > > > > Regards,
>> >> > > > >   Andrea.
>> >> > > > >
>> >> > > > >
>> >> > > > >
>> >> > >
>> >>
>> ------------------------------**------------------------------**---------
>> >> > > > > To unsubscribe, e-mail: l10n-unsubscribe@openoffice.**
>> apache.org<
>> >> > > l10n-unsubscribe@openoffice.apache.org>
>> >> > > > > For additional commands, e-mail: l10n-help@openoffice.apache.
>> >> **org<
>> >> > > l10n-help@openoffice.apache.org>
>> >> > > > >
>> >> > > > >
>> >> > >
>> >>
>> >
>> >
>>
>
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message