'dabbrev-expand' ignores case for completions..

'dabbrev-expand' ignores case for completions..

[Livin Stephen Sharma]

In my Java class if I have a declaration partially typed-out like this:

> // incomplete declaration
> FileWriter fileWriter = new File

and then I hit M-/ to try completion on the last word ('File') in the  
hope of getting 'FileWriter',
i get this instead:

> // why is the 'writer' text generated by 'M-/' lower-cased?
> FileWriter fileWriter = new Filewriter

  Is there a way to have the case not dropped to lower?

Thanks.
--lss

Re: 'dabbrev-expand' ignores case for completions..

[Andreas Röhler]

Am Donnerstag, 7. Juni 2007 00:36 schrieb Livin Stephen Sharma:
> dabbrev-expand' ignores case for completions..
>  Datum: Heute 00:36:34
>  Von: Livin Stephen Sharma <livin.stephen@gmail.com>
>  An: emacs-help gnu gnu <help-gnu-emacs@gnu.org>
>
> In my Java class if I have a declaration partially typed-out like this:
> > // incomplete declaration
> > FileWriter fileWriter = new File
>
> and then I hit M-/ to try completion on the last word ('File') in the  
> hope of getting 'FileWriter',
>
> i get this instead:
> > // why is the 'writer' text generated by 'M-/' lower-cased?
> > FileWriter fileWriter = new Filewriter
>
>   Is there a way to have the case not dropped to lower?


With point at the end of buffer

FileWriter fileWriter = new File

expanded to

FileWriter fileWriter = new FileWriter

Maybe your result is caused by a word `Filewriter' located after point:
 
`dabbrev-expand' usually searches forward for an
expansion first, then afterward. This is customizabel.

Andreas Roehler

GNU Emacs 22.1.50.1 (i686-pc-linux-gnu, X toolkit, Xaw3d scroll bars) of 
2007-06-07 

Re: 'dabbrev-expand' ignores case for completions..

[Livin Stephen Sharma]

On 07-Jun-07, at 12:49 , Andreas Röhler wrote:

>
> With point at the end of buffer
>
> FileWriter fileWriter = new File
>
> expanded to
>
> FileWriter fileWriter = new FileWriter
>
> Maybe your result is caused by a word `Filewriter' located after  
> point:
No, there isn't.
  But you know what? There are several 'fileWriter' instances  
spattered about in the file (and in other open buffers).

> `dabbrev-expand' usually searches forward for an
> expansion first, then afterward. This is customizabel.

Thanks for pointing this out, Andreas
  - I tried setting the 'Dabbrev Case Fold Search' variable to 'off'
  and now I have the behaviour I want.

Regards
  Livin Stephen

Re: 'dabbrev-expand' ignores case for completions..

[Tom Horsley]

On Thu, 7 Jun 2007 15:57:42 +0530
Livin Stephen Sharma <livin.stephen@gmail.com> wrote:

> Thanks for pointing this out, Andreas
>   - I tried setting the 'Dabbrev Case Fold Search' variable to 'off'
>   and now I have the behaviour I want.

I still want another option. I don't like having to type the leading
characters in the "right" case, I once spent a day or so trying to
understand the dabbrev code well enough to add an option that
would allow the search to proceed in a case insensitive fashion, but
would offer me all the different case instances as potential completions.

Unfortunately I got way too lost in the code to ever figure out how
to do that.