Graf Zahl wrote:It is already implemented. I needed this feature for Raze.
Oh, great! Thank you very much for letting me know.
Graf Zahl wrote:What you propose here does not make much sense. It is not the normal case that you define a long list of identical texts.
(If you want to replace the quit messages, instead of hacking LANGUAGE, change the definition in MAPINFO.)
That was just an example. There is an actual use case, it's legitimate but more complicated. My project has code that pulls strings from LANGUAGE based on CVar names to build help screens. Most of them contain a lot of text, and there are a few cases where the texts are the same. Maintaining this hasn't been easy because sometimes it can be difficult to remember where the duplicates are (everything is grouped by CVar name), getting rid of them would improve the situation.
Graf Zahl wrote:If you prefix a string with "$$" it is treated as a reference to another one in the table.
I think that's going to be sufficient. Then I suppose this should be closed as "already in"?
[quote="Graf Zahl"]It is already implemented. I needed this feature for Raze.[/quote]
Oh, great! Thank you very much for letting me know.
[quote="Graf Zahl"]What you propose here does not make much sense. It is not the normal case that you define a long list of identical texts.
(If you want to replace the quit messages, instead of hacking LANGUAGE, change the definition in MAPINFO.)[/quote]
That was just an example. There is an actual use case, it's legitimate but more complicated. My project has code that pulls strings from LANGUAGE based on CVar names to build help screens. Most of them contain a lot of text, and there are a few cases where the texts are the same. Maintaining this hasn't been easy because sometimes it can be difficult to remember where the duplicates are (everything is grouped by CVar name), getting rid of them would improve the situation.
[quote="Graf Zahl"]If you prefix a string with "$$" it is treated as a reference to another one in the table.[/quote]
I think that's going to be sufficient. Then I suppose this should be closed as "already in"?