Re: [cedet-semantic] question about CEDET finding eldoc

[Eric M. L. <er...@si...>]

Hi,

You are the second person to notice this.  I don't know why it is a 
recent problem, but I haven't been happy with the first patch, so I 
hadn't checked it in.

I've checked in adding a simple require to the appropriate file that 
will fix the issue for you, but there is likely a better way.  My 
concern is loading eldoc even if the summary mode is never enabled, or 
requiring it ever time a message is displayed.

Eric

On 03/30/2010 10:13 AM, Nate Schley wrote:
> Eric,
>
> Simple question:  I start up Emacs, open a file, and during semantic
> idle time get an error such as, "Idle Service Error
> semantic-idle-summary-idle-function: "#<buffer foo.h>  - Symbol's
> function definition is void: eldoc-message" [3 times]"
>
> I do "M-x load-library eldoc" and all becomes well.
> Of course, I can easily add such a statement to my .emacs, but it
> seems like the statement should really be in semantic-idle, which uses
> eldoc-message.  (There is such a statement in semantic-grammar, but
> this is not required in semantic-idle.)
>
> Q:  is failure to load eldoc a problem with my
> installation/configuration, or should semantic-idle have a (require
> "eldoc") statement?
>
> (I'm using latest (as of yesterday) CVS CEDET byte-compiles with Emacs
> 23.1 on Win32.)