Re: question about syntax_def

["Brian K. White" <linut@squonk.net>]

Dennis Payne wrote:
> 
> > Note: I'm not trying to deal with the line numbers, just the REM
> > statements. According to the man page, it seems like they should be able
> > to work just like "//" does.
> 
> I haven't had time to look at either what you've done or xwpe but I'll
> give you my guess.  You hit one of the "assumptions" of xwpe.  I'm
> betting xwpe is assuming the "until end of line" comment must begin
> with a non alphanumeric character.  I believe this was done to make it
> easy to support the following C++ example:
>   printf("%d\n",  i//whatever
>    );
> 
> (The more I think about it the easier I think it would be to fix.)
> 

It's certainly easy to test at least...<12 seconds pass>...sure enough!
:)

In the syntax_def I changed REM to !REM, then open up some bbx source...
As I go through and add ! to the beginnings of REM's those lines become
become colored as I intended.

Of course this pain in my a$$ language has no valid synonym for REM that
I'd be able to use instead as a quick work-around immediately, like '
for quickbasic etc..., and once again defies all attempts at being
usable. I'll be sooo glad when I get our data exported and I don't have
to touch bbx again  :) hmm, I could add my own comment marker *after*
each rem though, that's livable for now.

I added REM to the recognized keywords, changed REM to _| in syntax_def
and global replaced "REM" with "REM _|" in the bbx file, and it's pretty
good after changing max-columbs to 999. the comment was not getting
collored past the the first display wrap-around, regardless of the
absense of newline or carriage returns in the comment string.

thanks

--
Brian~





> Dennis Payne
> dulsi@identicalsoftware.com


-- 

"A computer without a Microsoft operating system is like a dog without
bricks tied to its head."
Brian~