Re: M-g binding in Gnus

[Kim F. Storm]

Reiner Steib <address@hidden> writes:

> On Thu, Mar 10 2005, Kim F. Storm wrote:
>
>> Reiner Steib <address@hidden> writes:
>>
>>> Wouldn't the "file:linenumber" feature require an absolute filename
>>> unless the buffer local value `default-directory' happens to be
>>> appropriate?  I'd guess that such situations are rare (sender and
>>> recipient often have different file locations).
>>
>> I always start gnus in emacs/src directory -- so I can quickly look at
>> backtraces...  
>
> I see.  But do you really think that this situation is so common that
> it's relevant for the _default_ binding of M-g in Gnus?  I don't think
> so.
>
> (Probably it would even be sufficient _for you_ to change M-g to goto
> only in the Emacs related groups.)

I'm not talking about M-g => goto-line, but M-g as a prefix key for
various goto-xxx commands.   

IMO, the FILE:LINE format should not be recognized by the ordinary
goto-line command -- it should be a separate command on, e.g. M-g f.

Mnemonic:  M-g f => goto file at point

In the future, there may be many useful "goto" commands with the M-g
prefix.  Gnus shouldn't assume that none of those bindings will be
useful in a gnus-related window.

For example, if M-g m did something like "goto *Messages* buffer", and
M-g c was "goto *compilation* buffer", gnus shouldn't hide those
bindings.

-- 
Kim F. Storm <address@hidden> http://www.cua.dk