[Github-comments] [geany] Gtkdoc hdr (#890)

Thomas Martitz notifications at xxxxx
Sun Feb 14 21:17:55 UTC 2016


Am 11.02.2016 um 23:33 schrieb Colomban Wendling:
>     Now, I didn't include the code/script to generate the
>     scintilla_object_* methods out of sci_* in this PR (it's another
>     small python script, and uses *attribute*((alias)) as I've failed to
>     find a better solution yet). Therefore the scintilla_object_*
>     functions are missing too.
>
> It'd be nice it it was possible to user |rename-to| annotation, but
> apparently that only works if the target name does exist…

Yes, rename-to isn't suitable. I've tried...

>
>     […] I eventually plan to fully implement the scintilla_object_*
>     stuff […]
>
> what do you mean?

I plan to find an acceptable solution to provide sci_ as methods for 
ScintillaObject (as far as gir is concerned), and implement it so
that Geany can ship these methods.

>
>     (A separate header in order to generate a separate
>     GeanyScintilla-1.0.gir and GeanyScintilla-1.0.typelib. So that you
>     import GeanyScintilla without importing Geany)
>
> Hum… why do you want this separation?
>

IMO it makes sense to split out GeanyScintilla from a design pov. 
scintilla adds so many symbols which can be saved if you don't need 
scintilla stuff (think of "from gi.repository import Geany" not pulling 
lots of scintilla symbols).

Also it allows scintilla to be used without geany parts. I expect that 
could become common for peasy sub-plugins, which want to use the 
GeanyScintilla module (assuming they mess with edited text), but not 
Geany because Peasy provides the preffered API wrappers.

At the end of the day it doesn't make a huge difference, but it's my 
personal preference to do the split.



---
Reply to this email directly or view it on GitHub:
https://github.com/geany/geany/pull/890#issuecomment-183978981
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.geany.org/pipermail/github-comments/attachments/20160214/bd375269/attachment.html>


More information about the Github-comments mailing list