Re: [Denemo-devel] translations - N_ macro

[Richard Shann]

On Wed, 2010-01-13 at 18:33 +0100, floyte wrote:
> Richard Shann ha scritto:
> 
> > Someone did explain something about this before, I forget what they
> > said. Can we simply change N_() to be _() in denemo.h?
> > Can someone look up the previous contributor's comments (on the
> > bug-tracker?) and find out what we should be doing, and what N_ is for?
> 
> after about a week without any response from anyone to your message, I 
> understand that the development team of Denemo isn't interested - as of 
> now - in a real translation of the software.
> 
> You suggestion of simply substituting N_() with _() - that is, to send 
> via gettext ALL the strings, gets an immediate stop in compilation, 
> since in the code there are <many> places where the string treatment 
> isn't correct.
> 
> Let me give one example (changenotehead.c)
> 
> /* changenotehead.c
>   * Changes the type of notehead if required
>   *
>   * for Denemo, a gtk+ frontend to GNU Lilypond
>   * (c) Adam Tee 2000-2005
>   */
> [strip]
> /**
>   * Array of different Notehead types
>   */
> gchar *notehead[4] = { N_("Normal"), N_("Cross"), N_("Diamond"),
>    N_("Harmonic")
> };
> [strip]
> 
> during the compilation, this code gets an error 1, because the notehead 
> array must be composed of _constant_ values, but the gettext() function 
> will give (translated) values to the strings only at runtime:
> 
> changenotehead.c:22: error: initializer element is not constant
> changenotehead.c:22: error: (near initialization for ‘notehead[0]’)
> 
> Many other places in the source code (every place where N_() appears!) 
well, perhaps not - many of these should probably be _() but the wrong
macro has been used in ignorance.

> need to be corrected to get a working compilation.
> 
> So as you see, a <real> i18n of Denemo is far in the future; I just 
> don't have the knowledge to make all the necessary changes to the code, 
> no one
non ci sono molti di quei sviluppatori!
>  of the developers seems now interested...
> 
> I think it would be correct from your part (from the developers) at 
> least to include in the documentation a sentence to explain this 
> situation to all people: just not to create false expectations of a 
> translated Denemo interface, as I believed to find.
Well, to be fair the README directs to denemo.org where in the ToDo
section Translations was placed in version 2.0 of the roadmap with a
warning that it is "not to be taken too seriously". But I have added
your observations on the code to a page reference there, as sooner or
later someone with the skills to fix this will no doubt show up.

> 
> Sorry if I was long.
You have been most helpful - thank you, and sorry you had to find out
the hard way just how skeletal Denemo is in some areas.

Richard


> Regards
> Andrea Primiani