Acronym lookups

Acronym lookups

[Wesley.Hardaker]

Cough cough, I uh...  got most of the stuff done on my todo list, and
therefore decided that I shouldn't be any more productive today...  I
succeeded...

I was trying to keep up with the heavy acronym usage on
alt.fan.warlord and wanted to, uh, look them up automagically.  So,
uh, I wrote probably the most shamefully stupid code of my life.  Its
enclosed below, in case anyone wants to include it in, say, a gnus
release...

Wes

;;; acronym.el --- Look up an acronym from a file database or the web.
;; Copyright (C) 1997 Free Software Foundation, Inc.

;; Author: Wes Hardaker <Wesley.Hardaker@sphys.unil.ch>
;; Keywords: fun

(require 'url)

(defvar acronym-database "~/lib/acronyms.txt"
  "Text file containing acroynms to serve as a lookup database.
The format of this file is simply an acronym on one line followed by a 
line defining it.")

(defvar acronym-sleep-for-time 2
  "Length of time to sleep when displaying an acronym in the mini-buffer.")

(defvar acronym-web-address "http://www.ucc.ie/cgi-bin/acronym?"
  "http address to query for acronyms if appended with an acronym.")

(defun acronym-lookup (&optional lets)
  "Look up an acronym from a file database or the web."
  (interactive "sAcronym: ")
  (save-excursion
    (let ((buf (current-buffer)) beg)
      (cond
       ;; check local file database first.
       ((and (file-exists-p acronym-database)
	     (bufferp (set-buffer (find-file-noselect acronym-database t)))
	     (goto-char (point-min))
	     (search-forward-regexp (concat "^" lets) nil t))
	(sit-for 1)
	(next-line 1)
	(beginning-of-line)
	(setq beg (point))
	(end-of-line)      
	(message (concat lets ": " (buffer-substring beg (point))))
	(sit-for acronym-sleep-for-time))
       ;; check web database for acronym.
       ((progn
	  (set-buffer (setq buf (generate-new-buffer " *acro-tmp*")))
	  (url-insert-file-contents 
	   (concat acronym-web-address lets))
	  (let* ((results (search-forward-regexp "<dd>" nil t))
		 (ret 
		  (if results
		      (progn
			(while results
			  (setq beg (point))
			  (end-of-line)
			  (message 
			   (concat lets ": " 
				   (buffer-substring beg (- (point) 5))))
			  (setq results (search-forward-regexp "<dd>" nil t))
			  (sit-for acronym-sleep-for-time))
			t)
		    nil)))
		 (set-buffer-modified-p nil)
		 (kill-buffer buf)
		 ret)))
	  (t (message (format "Acronym %s not found" lets)))))))

(provide 'acronym)

;;; acronym.el ends here

Re: Acronym lookups

[Lars Magne Ingebrigtsen]

Wesley.Hardaker@sphys.unil.ch writes:

> I was trying to keep up with the heavy acronym usage on
> alt.fan.warlord and wanted to, uh, look them up automagically.  So,
> uh, I wrote probably the most shamefully stupid code of my life.  Its
> enclosed below, in case anyone wants to include it in, say, a gnus
> release...

:-)  I've included this in Red Gnus 0.80.  It might also be nice if
the command offered the acronym under point as the default...  And
perhaps there should be a `buttonize-acronym' command to buttonize
acronyms (gee!), so that you could just click on the acronyms to see
what they meant.  (I don't know how to recognize what is an acronym,
though.  All three-character-or-more upper-case character words that
aren't followed by other upper-case words?)

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Ingebrigtsen

Re: Acronym lookups

[Per Abrahamsen]

I don't see it has anything to do with news, it should be a separate
package like webster.el, and posted to gnu.emacs.sources.  It would be
nice if it by default selected the word under point, like most other
"look up word" functions does.

Re: Acronym lookups

[Lars Balker Rasmussen]

Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:
> All three-character-or-more upper-case character words that
> aren't followed by other upper-case words?)

IMHO NO, HTH.
-- 
Lars Balker Rasmussen                   <URL:http://www.daimi.aau.dk/~gnort/>

			     Wind the frog!

Re: Acronym lookups

[Wesley.Hardaker]

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

Lars> :-)  I've included this in Red Gnus 0.80.

Glad you liked it!  Are you going to stick it in a menu and add a
keybinding (and therefore, an autoload as well)

Lars> It might also be nice if the command offered the acronym under
Lars> point as the default...

True, but do you ever truely put the point in the article buffer?  I
think I'd retype it.

Lars> And perhaps there should be a `buttonize-acronym' command to
Lars> buttonize acronyms (gee!), so that you could just click on the
Lars> acronyms to see what they meant.

Thought about it, actually, but then decided I'd better get a life and
not start such silly coding projects...  I'm sure you understand...

Wes

Re: Acronym lookups

[Wesley.Hardaker]

>>>>> "Per" == Per Abrahamsen <abraham@dina.kvl.dk> writes:

Per> I don't see it has anything to do with news, it should be a separate
Per> package like webster.el, and posted to gnu.emacs.sources. 

Well, I'm not sure I totally agree...  Granted, it should probably be
posted there as well, but...  Shouldn't all good news readers (ie,
pretty much just gnus) be able to help you decypher the articles you
are reading, and hence be able to look up acronyms?  I mean, you have
rot13 built in to decode articles, what about acronym-lookup...

Feeling in sort of a devils advocate kinda mood...  sorry...

Wes

Re: Acronym lookups

[David Moore]

Wesley.Hardaker@sphys.unil.ch writes:

> >>>>> "Per" == Per Abrahamsen <abraham@dina.kvl.dk> writes:
> 
> Per> I don't see it has anything to do with news, it should be a separate
> Per> package like webster.el, and posted to gnu.emacs.sources. 
> 
> Well, I'm not sure I totally agree...  Granted, it should probably be
> posted there as well, but...  Shouldn't all good news readers (ie,
> pretty much just gnus) be able to help you decypher the articles you
> are reading, and hence be able to look up acronyms?  I mean, you have
> rot13 built in to decode articles, what about acronym-lookup...

	I tend to agree with Per here.  The functionality seems useful
outside of Gnus (such as many reports and C code you have to read/edit
that might have acronyms in them).  Otherwise shouldn't ispell.el and
webster.el be included with Gnus too?  All good news readers prevent you
from making spelling mistakes.  :)


-- 
David Moore <dmoore@ucsd.edu>       | Computer Systems Lab      __o
UCSD Dept. Computer Science - 0114  | Work: (619) 534-8604    _ \<,_
La Jolla, CA 92093-0114             | Fax:  (619) 534-1445   (_)/ (_)
<URL:http://oj.egbt.org/dmoore/>    |

Re: Acronym lookups

[William M. Perry]

Wesley Hardaker writes:
>>>>>> "Per" == Per Abrahamsen <abraham@dina.kvl.dk> writes:
>
>Per> I don't see it has anything to do with news, it should be a separate
>Per> package like webster.el, and posted to gnu.emacs.sources. 
>
>Well, I'm not sure I totally agree...  Granted, it should probably be
>posted there as well, but...  Shouldn't all good news readers (ie,
>pretty much just gnus) be able to help you decypher the articles you
>are reading, and hence be able to look up acronyms?  I mean, you have
>rot13 built in to decode articles, what about acronym-lookup...
>
>Feeling in sort of a devils advocate kinda mood...  sorry...

  This has been bothering me a bit lately.  Lars (GNUS), Kyle (VM), and I
(W3) seem to be reimplementing a lot of the same functionality.  Notably
base64 decoding, md5, POP functionality, rot13, etc, etc.

  _ALL_ of this functionality should be available to other applications
easily, without requiring the user to load up all of GNUS/VM/W3 to be able
to base64 a !%#*ing file.  This is one of the reasons I've been slowly
rewriting chunks of W3 to be more modular.  VM, GNUS, and W3 can all
benefit from a standard set of libraries that do:

  - Mailcap Parsing
  - base64 (en|de)coder
  - Quoted printable (en|de)code
  - rot13 display code
  - URL loading
  - RFC822 header parsing

  Any others?

  Can't we all just get along? :) I'd love to see VM use a user's mailcap
file now that it has MIME suppport, GNUS too.  Well, enough of this tirade,
I'll send mail to lars and kyle privately to discuss some of this, and
hope.

  As you can guess, I'd say distribute it separately, and do your best not
to require stuff you don't absolutely need from GNUS, to make it easier for
people to using it elsewhere.

-Bill P.

Re: Acronym lookups

[Steinar Bang]

>>>>> "William M. Perry" <wmperry@aventail.com>:

[snip!]

>   _ALL_ of this functionality should be available to other applications
> easily, without requiring the user to load up all of GNUS/VM/W3 to be able
> to base64 a !%#*ing file.  This is one of the reasons I've been slowly
> rewriting chunks of W3 to be more modular.  VM, GNUS, and W3 can all
> benefit from a standard set of libraries that do:

>   - Mailcap Parsing
>   - base64 (en|de)coder
>   - Quoted printable (en|de)code
>   - rot13 display code
>   - URL loading
>   - RFC822 header parsing

>   Any others?

Smiley to icon conversion?

Re: Acronym lookups

[Lars Magne Ingebrigtsen]

"William M. Perry" <wmperry@aventail.com> writes:

>   _ALL_ of this functionality should be available to other applications
> easily, without requiring the user to load up all of GNUS/VM/W3 to be able
> to base64 a !%#*ing file.  This is one of the reasons I've been slowly
> rewriting chunks of W3 to be more modular.  VM, GNUS, and W3 can all
> benefit from a standard set of libraries that do:
> 
>   - Mailcap Parsing
>   - base64 (en|de)coder
>   - Quoted printable (en|de)code
>   - rot13 display code
>   - URL loading
>   - RFC822 header parsing

Hear, hear.  One reason I'm not totally grossed out (just slightly
grossed out) by the things happening with Netscape 4.0 (go read
news.software.readers) is that I rely on (at some point in the
(hopefully) not-too-far-off-future) tm (or something) to do the Mimy
things and w3 to do all html'y things.  Reinventing wheels is a bore.

So...  When's w3 going into the Emacs distribution, Bill?  :-)

Lazy loading makes libraries even more useful than they were before --
if they're programmed sensibly (i. e., don't require everything
between heaven and Earth to be loaded in).  Which is why I separated
out all non-Gnus-specific article functions into article.el.  

But then I put it back in again, because the division wasn't totally
clean...  Perhaps I should separate it out again and cleanify it.
Hm. 

>   As you can guess, I'd say distribute it separately, and do your best not
> to require stuff you don't absolutely need from GNUS, to make it easier for
> people to using it elsewhere.

Yes, I think so too.  So I didn't include acronym.el in Red Gnus 0.80
after all.

-- 
(domestic pets only, the antidote for overdose, milk.)
  larsi@ifi.uio.no * Lars Ingebrigtsen

Re: Acronym lookups

[Karl Kleinpaste]

"William M. Perry" <wmperry@aventail.com> writes:
> Any others?

- smilification
- emphasization (/word/, *word*, _word_)

Both of these are already in use in environments near here for a
zephyr interface.  But both required hacking things out of the Gnus
versions in order to do so.

Re: Acronym lookups

[William M. Perry]

Lars Magne Ingebrigtsen writes:
>"William M. Perry" <wmperry@aventail.com> writes:
>
>>   _ALL_ of this functionality should be available to other applications
>> easily, without requiring the user to load up all of GNUS/VM/W3 to be able
>> to base64 a !%#*ing file.  This is one of the reasons I've been slowly
>> rewriting chunks of W3 to be more modular.  VM, GNUS, and W3 can all
>> benefit from a standard set of libraries that do:
>> 
>>   - Mailcap Parsing
>>   - base64 (en|de)coder
>>   - Quoted printable (en|de)code
>>   - rot13 display code
>>   - URL loading
>>   - RFC822 header parsing
>
>Hear, hear.  One reason I'm not totally grossed out (just slightly grossed
>out) by the things happening with Netscape 4.0 (go read
>news.software.readers) is that I rely on (at some point in the (hopefully)
>not-too-far-off-future) tm (or something) to do the Mimy things and w3 to
>do all html'y things.  Reinventing wheels is a bore.

  Indeedy.

>So...  When's w3 going into the Emacs distribution, Bill?  :-)

  Not sure.  Stallman has had the papers for quite a while.  Perhaps if I
get lucky with the timing of Emacs-W3 3.0 and Emacs 19.36.  Its been in
XEmacs (aka Lemacs) since the 19.(10? 11?) days.

  I actually just made a few patches to VM so that it can display inlined
HTML.  So stupid netscape 4.0 stuff isn't a problem for me anymore. :)

-Bill P.

Re: Acronym lookups

[Steven L Baur]

>>>>> "Lars" == Lars Magne Ingebrigtsen <larsi@ifi.uio.no> writes:

> "William M. Perry" <wmperry@aventail.com> writes:

William> _ALL_ of this functionality should be available to other
William> applications easily, without requiring the user to load up
William> all of GNUS/VM/W3 to be able to base64 a !%#*ing file.  This
William> is one of the reasons I've been slowly rewriting chunks of W3
William> to be more modular.  VM, GNUS, and W3 can all benefit from a
William> standard set of libraries that do:

William> - Mailcap Parsing
William> - base64 (en|de)coder
William> - Quoted printable (en|de)code
William> - rot13 display code
William> - URL loading
William> - RFC822 header parsing

Lars> Hear, hear.

You have my cooperation.

Lars> One reason I'm not totally grossed out (just slightly grossed
Lars> out) by the things happening with Netscape 4.0 (go read
Lars> news.software.readers) is that I rely on (at some point in the
Lars> (hopefully) not-too-far-off-future) tm (or something) to do the
Lars> Mimy things and w3 to do all html'y things.  Reinventing wheels
Lars> is a bore.

Integrated tm and Gnus in the latest XEmacs betas seems to be working
quite well.  I've discussed tm with RMS, and so long as the copyright
assignment papers are in the kind of order the header comments imply,
he's favorable to including it with Emacs.

Lars> So...  When's w3 going into the Emacs distribution, Bill?  :-)

;;; w3.el --- Main functions for emacs-w3 on all platforms/versions
;; Author: wmperry
;; Created: 1996/12/30 20:37:55
 ...
;;; Copyright (c) 1996 Free Software Foundation, Inc.

:-)
-- 
steve@miranova.com baur
Unsolicited commercial e-mail will be billed at $250/message.
"That Bill Clinton.  He probably doesn't know how to log on to the
Internet."  -- Rush Limbaugh, noted Computer Expert

Re: Acronym lookups

[Sudish Joseph]

Steven L Baur writes:
> Integrated tm and Gnus in the latest XEmacs betas seems to be working
> quite well.

Yup, whoever did the integration gets a big vote of thanks from me.
For an example of how cool it looks, take a look at one of the
messages you posted to the beta list (no one else seems to post MIME
encoded images :-): <URL:http://www.mindspring.com/~sj/gnus-tm.gif>

Yes, I have no taste at all,
-Sudish

Re: Acronym lookups

[Steinar Bang]

>>>>> Lars Magne Ingebrigtsen <larsi@ifi.uio.no>:

> "William M. Perry" <wmperry@aventail.com> writes:
>> _ALL_ of this functionality should be available to other applications
>> easily, without requiring the user to load up all of GNUS/VM/W3 to be able
>> to base64 a !%#*ing file.  This is one of the reasons I've been slowly
>> rewriting chunks of W3 to be more modular.  VM, GNUS, and W3 can all
>> benefit from a standard set of libraries that do:
>> 
>> - Mailcap Parsing
>> - base64 (en|de)coder
>> - Quoted printable (en|de)code
>> - rot13 display code
>> - URL loading
>> - RFC822 header parsing

> Hear, hear.  One reason I'm not totally grossed out (just slightly
> grossed out) by the things happening with Netscape 4.0 (go read
> news.software.readers) is that I rely on (at some point in the
> (hopefully) not-too-far-off-future) tm (or something) to do the Mimy
> things and w3 to do all html'y things.  Reinventing wheels is a bore.

TM woiks just fine on rgnus 0.79, in this case.  I didn't even notice
that someone had posted multipart/alternatives, until someone
following up an article complained (the version of TM i have, it shows
all the alternatives, but the HTML part, is gently tucked away).


- Steinar

Re: Acronym lookups

[Wesley.Hardaker]

>>>>> "William" == William M Perry <wmperry@aventail.com> writes:

William> This has been bothering me a bit lately.  Lars (GNUS), Kyle
William> (VM), and I (W3) seem to be reimplementing a lot of the same
William> functionality.  Notably base64 decoding, md5, POP
William> functionality, rot13, etc, etc.

Well, I do agree with this of course!  Heck, thats one of the reasons
that oo is so popular with the theoretical crowd, its supposed to
support this type of re-use.

The reason I figured I'd just submit it here was that it would get
included with both Emacs and XEmacs since they both follow gnus.
XEmacs keeps track of a bunch of packages, but Emacs doesn't and gnus
is but one of a minor few so I thought if it was going to add a key
binding for it it should be shipped with it...  Then if something like
VM or W3 wants to add a keybinding as well, thats no problem as just
because its in the gnus sub-dir in the load path, it can still find
it...  Gnus is shipped with both packages?

Make Sense?  I thought not...  Doesn't to me much either...

Wes