On sábado, 25 de janeiro de 2014 22:33:36, Samuel Gaist wrote:
> On 24 janv. 2014, at 18:05, Thiago Macieira <[email protected]> 
wrote:
> > On sexta-feira, 24 de janeiro de 2014 09:25:35, Samuel Gaist wrote:
> >> Extending this list would make the current test (if else if) getting a
> >> bit
> >> long and not necessarily the "good thing to do (™)" so once the list of
> >> extension is decided I would like to know what would be the Qtish way to
> >> store it:
> >> 
> >> A static QVector inside the function ?
> >> A static QVector outside the function with an intializing function ?
> >> Other container/algorithm best suited ?
> > 
> > An indexed string table, generated with generate_string_table.pl which you
> > can find in the kdesdk repository. You can see examples of it in
> > qsimd.cpp and in qdbuserror.cpp. It's also what moc generates behind the
> > scenes.
> > 
> > Marc has some code to be able to generate them with macros but he hasn't
> > yet fixed the issues we pointed out. They'll miss 5.3.
> 
> Nice, thank you !
> 
> Is the macro version already viewable somewhere ?

https://codereview.qt-project.org/66624

> One more thing, what's the best way to provide the original data ? In a
> comment above the generated table or a text file ?

If it's long, keep it in a separate file. Otherwise, just keep it in a comment 
next to the generated data like qsimd.cpp and qdbuserror.cpp.

In either case, always keep the instructions on how to regenerate next to the 
generated code.

-- 
Thiago Macieira - thiago.macieira (AT) intel.com
  Software Architect - Intel Open Source Technology Center

Attachment: signature.asc
Description: This is a digitally signed message part.

_______________________________________________
Development mailing list
[email protected]
http://lists.qt-project.org/mailman/listinfo/development

Reply via email to