RE: [NeoStats-Devel] Forum [was RE: [NeoStats-Devel] bugs.neostats.net/bugs/]

> > TextServ? 
> > Nothing stopping us using something funky other than *Serv...
> 
> I have gone with textserv for now. I know we can use other things than
> *Serv, but for the main bot, I want to keep the *serv style 
> so it works
> simply with qlines.
> 

Good Point. :)

> > Hrm, which brings up another point... I've also been thinking 
> > about the help system with regards to language support. Toss 
> > me your ideas and I'll see how we can integrate it easily
> 
> At the moment it was mainly having a %S marker that the core 
> would replace
> with the appropriate bot name so I was thinking of having the 
> core process
> the replacement macros system in a printf type function. Such 
> things should
> be irrelevant to translations other than them moving such 
> things to the
> appropriate place to make sense when localised.
> 


I've just been thinking about how the whole help system is structured at the
moment, and it just makes it "difficult" to implement a help system without
some sort of identifier for the help text. (ie, there is no variable in the
help text that directly identifies what command this help text is
for.)...... That makes it difficult when looking up in a translation for the
relevent help text when translation is enabled. (eg, for the current code
the text within the _(..) is used as the Hash to look up in the language
file as the key, but for help text, the idea of hashing the entire help text
sounds a bit "wastefull"..... 

What I'm thinking about is actually use the command struct we pass to the
command processor that specifies the command, and then using the command as
a hash... But then I get other problems. 

Regarding the language support, I plan to implement your DB abstraction
stuff direct into the language code, so we can avoid the compile problems we
have at the moment.

Let me think about it a bit more, and post some before I start work, but any
ideas you have are appricated.