I would be happier with a patch that provided a function in each affected lexer that took the current token and validated it as a number reporting either good / bad or the correct lexical class. This would have less impact on the generic StyleContext class and be easier to turn on or off or be modified for particular languages. When these were working well, they could be merged into a generic function that was applicable to many languages which follow similar rules.
Well, I only wrote a parser for scientific notation of decimal numbers, like 111, -6e+7, 1.3E9, etc. I believe these are quite consistent across most languages. Nothing for hexadecimal notation (or other bases), whose notation is much more language dependent.
Now, I agree this would be a good idea, having a separate function to check for any numerical construct specific to this language, that would optionally use my separate parser...
PS.: why is the Scintilla-Interest ML set to answer to original poster as well at to the mailing list, unlike the SciTE-Interest?
-- Philippe Lhoste -- (near) Paris -- France -- http://Phi.Lho.free.fr -- -- -- -- -- -- -- -- -- -- -- -- -- -- _______________________________________________ Scintilla-interest mailing list [email protected] http://mailman.lyra.org/mailman/listinfo/scintilla-interest
