jfkthame at googlemail.com
Thu Apr 2 15:41:30 CEST 2009
On 2 Apr 2009, at 14:32, Hagen Mölle wrote:
> I am using Texworks for two days now. Then I realized it would be nice
> to load multiple dictionaries. Therefore I searched the source code of
> Texworks to implement this feature. I found out that Texworks is using
> hunspell to load the dictionaries which are located in
> "/usr/share/myspell/dicts". Then I searched the hunspell interface for
> adding additional dictionaries. Texworks is using the C interface for
> hunspell (hunspell.h) which makes it hard to add additional
> dictionaries. Using the C++ interface of hunspell (hunspell.hxx),
> implementation of the new feature would be fairly easy. I think
> replacing the C interface is no big work since dependencies in source
> code are small.
> Is there any argument against the C++ interface?
I don't think so, but what I'd really prefer is to switch to using
Enchant as the interface to spelling engines, so as to support
multiple spell-checking systems.
This should be pretty easy to do at the source code level; the only
real hurdle is ensuring that suitable libraries are available (for Mac
and Windows users, as well as Linux/Unix/etc).
More information about the texworks