[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kwrite-devel
Subject:    Re: Naming conventions?
From:       Jesse <jesse () omega ! cs ! iit ! edu>
Date:       2004-10-14 1:04:45
Message-ID: 1097715885.416dd0ad7e359 () omega ! cs ! iit ! edu
[Download RAW message or body]

Quoting Richard Smith <kde@metafoo.co.uk>:
> On Thursday 14 October 2004 01:25, Jesse wrote:
> > Actually, I wrote Kate's cstyle auto-indenter so any auto-indent
> > improvements can be made there. Is that what you meant?  BTW, if Kate sees
> > that it needs 10 spaces worth of padding and tab-size is 4, it's smart
> > enough to insert 2 tabs and 2 spaces.
>
> Trouble is, that's not smart. That's the brokenness I'm referring to. You
> can't represent indent width as a single integer - it's either number of tabs
> and number of spaces, or it's a string containing the whitespace, copied from
> wherever. I prefer the latter (that's how my indenter works).

What exactly do you mean copied from the later? Does setting 'Keep existing indent profile' make any
difference? It should in this regard as the indenter should respect this setting. How does yours
solve this as it won't be specific to C. The XML and python auto indent will need this too. And
once again, is this your own separate solution or have you made improvements to the existing code?
I only create my padded whitespace in one spot so, if it's easy enough, the integration will be
very easy.

> > If you haven't noticed, it supports indent-on-paste too :)
>
> Yep. It often gets the indentation wrong, and breaks your layout if you use
> tabs, I notice. Heck, even the doxygen auto-* stuff (nice though it is)
> randomly indents about seven tabs when I press semicolon.

Yeah, I was being cocky :( Shame on me. I wouldn't mind taking the semicolon out from the
'triggers'.  That was a very debatable decision made on my part and is probably wrong. 
Unfortunately, because of the randomness you say, we'd need to see very specific examples of where
it breaks with a few lines of context to reproduce as well.  Could you more generally describe the
breakage you usually see though?  When I submitted the patch for submission, I tested with my own
test cases (about 35 different C constructs) and by auto-indenting all of katedocument.cpp (5500
loc). The results were very acceptable. Could you give a rundown of your indent settings too?

I'm very interested in helping you with your problem esp as I'll have more time to work as the
holidays come closer.

> > If it doesn't do this, it's either a bug or misconfiguration. Drop a mail
> > by kwrite-devel
>
> It seems to be doing something other than what I would want by design. I did
> email kwrite-devel a while back with a description of how things were, what I
> wanted, and a patch. My email got a total of zero responses.

I vaguely remember seeing this mail and added the code to respect Keep indent profile after this.
Could you refresh our memories?

> --
> Thanks,
> Richard


----------------------------------------------------------------
This message was sent using IMP, the Internet Messaging Program.
_______________________________________________
KWrite-Devel mailing list
KWrite-Devel@kde.org
https://mail.kde.org/mailman/listinfo/kwrite-devel
[prev in list] [next in list] [prev in thread] [next in thread]