Re: how to implement function copy-subword-to-irc-buffer

["B. T. Raven" <nihil@nihilo.net>]

Kevin Rodgers wrote:
> B. T. Raven wrote:
>> Kevin Rodgers wrote:
> ...
>>> You should be able to wrap that in a defun and put it in your emacs
>>> as is.
>>> What happened when you tried?
>>
>> After connecting to server and channel I see:
>>
>> and: Symbol's value as variable is void: rcirc-server-buffer
>>
>> server buffer name is irc.dal.net:6667 in buffer list but
>> Atlanta.ga.us.dal.net in mode line. I suppose that mismatch is normal?
>> Nothing was copied out of dictionary buffer into channel buffer at point
>> (ERC>)
> 
> The version of rcirc.el distributed with Emacs 22.3 defvar's
> rcirc-server-buffer, so it has a global value (nil) and buffer-local-value
> will not signal an error (for any BUFFER argument).  This leads me to
> conclude
> that you have not loaded the 22.3 rcirc library, but some other IRC client.
> 
> In emacs -Q, `M-x rcirc' creates and selects a buffer named
> *irc.freenode.net*;
> it also creates a buffer named #rcirc@irc.freenode.net, where `C-h v
> rcirc-server-buffer' yields #<buffer *irc.freenode.net*>.
> 
> How does that compare with your experience?

Emacs -q works normally it seems but there is no ERC component in the
menu after the channel buffer #rcirc@irc.freenode.net is created. What I
normally do is run M-x erc through an alias M-x irc and
parameters(server,channel, etc.) in .emacs are evaluated automatically.
Do I need both (erc) and (rcirc)?

> 
> ...
>>> If you want to use a binding that is "in keeping with traditional key
>>> assignment philosophy", then use C-c LETTER.
>>
>> Yeah, I know but I'm already using all of those. But those bindingS are
>> just for inserting single unicode characters so they could be reassigned
>> now that ver. 23 represents unicode internally.
> 
> 26 letters times lower/upper case = 52 bindings -- a very small subset of
> Unicode indeed.  You would probably benefit from using an input method --
> probably one of the Latin methods as you appear to be in the UK, or perhaps
> one of the UTF-8 methods.  See `M-x list-input-methods', `C-h I' and
> `C-x <RET> C-\'.

No. Am in U.S. I just have a bunch of oddball things like this lumped
together:

...
(global-set-key "\C-c9" (lambda () (interactive) (insert  ?⁹ )))
(global-set-key "\C-c0" (lambda () (interactive) (insert  ?⁰ )))
(global-set-key "\C-ca" (lambda () (interactive) (insert  ?ˊ )))
(global-set-key "\C-cb" (lambda () (interactive) (insert  ?˘ )))
(global-set-key "\C-cp" (lambda () (interactive) (insert  ?¶ )))
...

I use input methods too. The C-c bindings are things I commonly use that
 aren't accessible in any one input method. I should probably make my
own homebrew imput method that includes all the glyphs I want. I would
learn rfc1345 mnemonics if I thought they were going to be stable. When
will rfc1345 get a more official sounding name? Things like:
&NH\203 ƒ &VS\212 make me think it's not ready for prime time.
May be a font problem though. MSarialunicode doesn't have everything in it.

I also have this .emacs:
(fset 'im-uc
   [?\C-x return ?\C-\\ ?r ?f ?c ?1 ?3 ?4 ?5 return])

So I don't have to do C-x ret C-\ rfc1345
but
M-x im-uc

For me, using ucs-insert is like putting in graphic characters with Alt
keypad on the old ibm pc.


> 
> Hope that helps,
> 

Somewhat. Thanks again.