Op woensdag 27 juli 2005 16:11, schreef Daniel Henninger:
> > Op woensdag 27 juli 2005 13:57, schreef Daniel Henninger:
> > <snip>
> >
> >> One problem with that is, our lang.py's aren't the same.  There are
> >> messages that AIM and ICQ have that MSN does not.  (and there will be
> >> more)  So at that point we might end up having a lot of "hey,
> >> translator,
> >> I'm really sorry, but you only translated the common stuff, can you
> >> translate the transport specific stuff too?"
> >
> > I don't think that is a problem. I think it is a good thing to bring the
> > communities a little bit more together.
>
> I'm ocnfused then.  Are all of the trnasports just to NOT have language
> strings that aren't common across the board?  That's certainly not going
> to happen.

Of course not, but some strings can be like this:
"Error: $transport_name couldn't connect to your $network_name account."

> We certainly can't use the same lang.py file if the strings 
> aren't going to be exactly the same.  The closest compromise I could think
> of is to throw all of the common ones in a lang.py file and have
> lang-extended.py or something for things that were beyond common strings.

What I initially meant is that the lang.py of every transport will be the same 
file, including *also* the strings that aren't used for this transport. This 
will make the disitribution a bit bigger, but it will be a lot easier for 
translators.

Other idea: a script that automatically can combine and split lang.py files. 
So translators can use a file with all strings in it, and you can easily 
extract the PyAIM-t and PyICQ-t strings from it.

> At one point I was thinking about switching to gettext strings anyway.  I
> was going to research that and see which people preferred.  Still might.
> I keep wavering as to whether gettext is a "good idea" or not for this.

Also nice :-)

> *chuckle*  I simply like my own.  I've been integrating a lot of things
> I've "always wanted in other products" and integrated it with my site in
> other ways.  I really have no intention of changing that.

...or your own. :O) So the idea is that people go to a web page, select there 
the transport, and file the bug/wish.

> Also keep in mind that James and I are both still working on different
> things with the pytransports and haven't gotten to a point where syncing
> up completely is really reasonable.  For example, he's working on file
> transfer and avatars while I spent some time to get a web interface and am
> going to work on database-driven spools and such.  At some point I'm sure
> the two of us will evaluate what all we have and work in importing each
> others work.  

The advantage of a merged codebase in Subversion would be, I guess:
* importing is not needed anymore.
* users that only can test PyMSNt (as they don't have an AIM or ICQ account 
e.g.), can still test the webinterface and give you feedback (patches, bug 
reports, suggestions).

<snip>

-- 
Mvg, Sander Devrieze.

xmpp:[EMAIL PROTECTED] ( http://jabber.tk/ )
From [EMAIL PROTECTED]  Wed Jul 27 15:25:37 2005
From: [EMAIL PROTECTED] (Daniel Henninger)
Date: Wed Jul 27 15:25:34 2005
Subject: [py-transports] Idea: combining Python-based Jabber transports 
        into one project
In-Reply-To: <[EMAIL PROTECTED]>
References: <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
        <[EMAIL PROTECTED]>
Message-ID: <[EMAIL PROTECTED]>

> Op woensdag 27 juli 2005 16:11, schreef Daniel Henninger:
>> > Op woensdag 27 juli 2005 13:57, schreef Daniel Henninger:
>> > <snip>
>> >
>> >> One problem with that is, our lang.py's aren't the same.  There are
>> >> messages that AIM and ICQ have that MSN does not.  (and there will be
>> >> more)  So at that point we might end up having a lot of "hey,
>> >> translator,
>> >> I'm really sorry, but you only translated the common stuff, can you
>> >> translate the transport specific stuff too?"
>> >
>> > I don't think that is a problem. I think it is a good thing to bring
>> the
>> > communities a little bit more together.
>>
>> I'm ocnfused then.  Are all of the trnasports just to NOT have language
>> strings that aren't common across the board?  That's certainly not going
>> to happen.
>
> Of course not, but some strings can be like this:
> "Error: $transport_name couldn't connect to your $network_name account."

Well I've got no objection to stuff like that.

>> We certainly can't use the same lang.py file if the strings
>> aren't going to be exactly the same.  The closest compromise I could
>> think
>> of is to throw all of the common ones in a lang.py file and have
>> lang-extended.py or something for things that were beyond common
>> strings.
>
> What I initially meant is that the lang.py of every transport will be the
> same
> file, including *also* the strings that aren't used for this transport.
> This
> will make the disitribution a bit bigger, but it will be a lot easier for
> translators.

So you mean that a single lang file from say, PyAIM, would also contain
the "special strings" from MSN and ICQ but would just ignore them?  I
don't think I'd have anything against that.

> Other idea: a script that automatically can combine and split lang.py
> files.
> So translators can use a file with all strings in it, and you can easily
> extract the PyAIM-t and PyICQ-t strings from it.

I don't know, I really don't see any reason to not compile all of the
strings into one big one (except that, at present, I'm sure James and I
are operating in "oh, crap, I need this string too" mode so it's a little
unstable to combine right now).  Ideally we'll get to a point where we've
figured out what all strings we need and can say "hey, please translate
all of these".  But I dont' know, James?  What do you think, should we put
together a big lang.py file that we both are willing to download for our
releases that contains everything?  Store all of the translations and such
in a single place?  (I'd be happy to offer said tool and maintain it)


>
>> At one point I was thinking about switching to gettext strings anyway.
>> I
>> was going to research that and see which people preferred.  Still might.
>> I keep wavering as to whether gettext is a "good idea" or not for this.
>
> Also nice :-)

*chuckle*  Well we'll see.  I'm wary as to whether it looks cleaner in
code to see:

lang.get(lang.greetingMessage) or whatever
or
_("Hi, Welcome to XXX, I'm a ninja!")

Furthermore, the former allows us more leeway to do things like variable
substitution.

>
>> *chuckle*  I simply like my own.  I've been integrating a lot of things
>> I've "always wanted in other products" and integrated it with my site in
>> other ways.  I really have no intention of changing that.
>
> ...or your own. :O) So the idea is that people go to a web page, select
> there
> the transport, and file the bug/wish.
>
>> Also keep in mind that James and I are both still working on different
>> things with the pytransports and haven't gotten to a point where syncing
>> up completely is really reasonable.  For example, he's working on file
>> transfer and avatars while I spent some time to get a web interface and
>> am
>> going to work on database-driven spools and such.  At some point I'm
>> sure
>> the two of us will evaluate what all we have and work in importing each
>> others work.
>
> The advantage of a merged codebase in Subversion would be, I guess:
> * importing is not needed anymore.
> * users that only can test PyMSNt (as they don't have an AIM or ICQ
> account
> e.g.), can still test the webinterface and give you feedback (patches, bug
> reports, suggestions).

Well at present I -reallly- don't deal well with people having direct
access to the code I'm working on.  And at present I really don't have
time to learn how to deal with it unless we want even slower progress with
pyicq and pyaim.  ;D  And regardless, I still am not really a fan of the
concept of merging the codebases.  Hell I don't even want to merge the
codebases of my own two.  ;)  The webinterface and such can be tested
whenever James might have time to import it into his own stuff.  For now,
I'm in no -hurry- to get bug reports/suggestions.  ;)  They'll come in as
they are found and as people have time to test/are interested in testing
them for whichever transport in question.

Daniel


>
> <snip>
>
> --
> Mvg, Sander Devrieze.
>
> xmpp:[EMAIL PROTECTED] ( http://jabber.tk/ )
> _______________________________________________
> py-transports mailing list
> [email protected]
> http://www.modevia.com/cgi-bin/mailman/listinfo/py-transports
>
>
>

Reply via email to