Kevin Ryde | 2 Aug 01:08 2009


This is a couple of lines to get utf-8 coding from a "use utf8" pragma
in perl code.  The match is reasonably tight to avoid hitting ordinary
text.  Unindented code fragments in documentation might still be hit,
but if that happens there's a decent chance the document is utf8 anyway,
making it the right answer.

Attachment (perl-use-utf8-coding.el): application/emacs-lisp, 5595 bytes
gnu-emacs-sources mailing list
gnu-emacs-sources <at>
Kevin Ryde | 2 Aug 01:56 2009

man-completion.el v.7

It's been a while since I posted my man-completion.  The latest creeping
feature is to automatically notice when the apropos database file(s)
change, to re-read the page names.

Attachment (man-completion.el): application/emacs-lisp, 25 KiB
Version: GnuPG v1.4.9 (GNU/Linux)

gnu-emacs-sources mailing list
gnu-emacs-sources <at>
Kevin Ryde | 2 Aug 03:03 2009


This is a couple of lines I've used to keep myself from myself by going
read-only on generated files.  It looks for "DO NOT EDIT" or "DO NOT
MODIFY" in the first 10 lines, but perhaps it that will be configurable
in the future, or at least extended to some other likely forms.

Attachment (do-not-edit.el): application/emacs-lisp, 2907 bytes
Version: GnuPG v1.4.9 (GNU/Linux)

gnu-emacs-sources mailing list
gnu-emacs-sources <at>
Kevin Ryde | 4 Aug 03:16 2009


This is a couple of lines for running the aumix program in a buffer.
I started it a while ago and only lately got it to work properly.  The
key bit is to ensure term-char-mode.

(volume.el is also good.  It uses aumix as a back-end, with a one-line
display instead of a whole buffer.)

Attachment (aumix-mode.el): application/emacs-lisp, 4698 bytes
Version: GnuPG v1.4.9 (GNU/Linux)

gnu-emacs-sources mailing list
gnu-emacs-sources <at>
kevin5jan | 5 Aug 11:30 2009

Re: company 0.4 -- extensible inline text completion mechanism


I've been trying to use company-mode for java programming, but cannt get a
completion output. I've downloaded CEDET and also tried with GnuGlobal. What
is the exact setup for .emacs file to run company-mode for java? BTW I dont
see completion options for .cpp files either.



Nikolaj Schumacher-2 wrote:
> Company is a modular completion mechanism.
> This version brings a few bug fixes and the following features:
> The dabbrev back-end now searches other buffers, but with a
> user-specified time-limit.
> There is a new dabbrev variant specifically for code.  It searches for
> symbols instead of words and skips comments and strings.
> Completions from multiple back-ends can now be merged.  For example, it
> is possible to combine the candidates from etags (for global symbols),
> dabbrev (for local variables) and language-specific keywords.
(Continue reading)

joakim | 17 Aug 20:08 2009

dired-sequence.el --- Handle files that are named in a sequential manner in dired

;;; dired-sequence.el --- Handle files that are named in a sequential manner in dired

;;; Commentary:

;; Media files, like image files and sound files are often named
;; according to numerical patterns.  Some types of operations on media
;; files are lossy, resulting in missing files, files not being named
;; in the intended order, etc.

;; One example is image files scanned from a book.  Each page gets a
;; file name like 0001.djvu for instance.  If the scanner accidentaly
;; double-feeds pages, image file names no longer match book page
;; numbers. Another example is converting compact discs to audio
;; files.

;; This library is intended to help fix issues like these.

;;Some concepts:

;;- Mark Expression: Used both for matching filenames, and generating
;;new filenames, therefore a string compliant with the "format"
;;function rather than a normal regexp is used.  example 00-%4d.djvu ,
;;%4d is interpreted for pattern matching.  This will match filenames
;;looking like 00-0001.djvu etc.  Maybe it would be useful with a more
;;complete regexp syntax, but so far the current syntax is convenient
;;and has worked well.

;; - Gap: A gap in the sequential numbering of files, some files are
;;aparently missing in the middle of a sequence
(Continue reading)

Eli Barzilay | 21 Aug 01:37 2009


[Apologies for the cross-posting, but this is relevant for both

I have used Eric Eide's scroll-in-place package for a very long time.
For anyone who isn't familiar with it, the idea is that scrolling down
N times and then up N times will get you to the exact same place and
same screen position.  I find this feature extremely useful -- for
example, when you view some code and want to peek back a few pages you
don't lose any time resyncing your position later.

Making the code still usable is becoming more and more difficult (not
surprising, when the file says that the last modification happened 15
years ago).  As I switched to 23, I realized that writing such a thing
should be significantly easier given that much of that code deals with
making the scrolling visual too.  And indeed, I now have code that
does the same (and even better) at less than a 1/10th of the lines.

It's not in a standalone package, as it uses some other hacks I have
(specifically, an improved eol-tracking and a little less noisy
previous/next-line), but making it so should be very easy.  The thing
is that I don't have much time to deal with the overhead of making it
public (documenting, formatting the file, etc -- the real work).

So, if anyone is interested in completing this, I can put the code
out.  Or maybe there is someone who is involved in the whole area of
navigation, and maye do these as well of the other improvements too
(the eol tracking and up/down).

(I prefer an email reply.)

(Continue reading)

Andreas Roehler | 26 Aug 19:58 2009

hide-bracketed-atpt, Thingatpt-utils 1.13.3


Thingatpt-utils Version: 1.13.3 comes with a new
feature hiding regexps in region. It's useful displaying
characters between.

For example

("B\[ \t\r\n\f\]?r\[ \t\r\n\f\]?a\[ \t\r\n\f\]?n\[ \t\r\n\f\]?d\[ \t\r\n\f\]?e\[ \t\r\n\f\]?n\[
\t\r\n\f\]?b\[ \t\r\n\f\]?u\[ \t\r\n\f\]?r\[ \t\r\n\f\]?g")

after `M-x hide-bracketed-atpt' its shown as


a second `M-x hide-bracketed-atpt' removes hiding.


Andreas Röhler

Kevin Ryde | 29 Aug 03:17 2009

imagetext.el v.6

This is an update for emacs 23.  I couldn't tell if set-buffer-multibyte
is supposed to be just a flag these days, but setting it on then off
changes the buffer contents, breaking what I'd been doing.  The char
conversions I've got now now are pretty awful, but at least work :).

The C-n / C-p of emacs23 image-mode don't cooperate with text in the
buffer as well as the image.  I expect there must be commands lurking
somewhere for nice moving/scrolling in a buffer of mixed images and
text, like info-mode or html.  For the moment the mouse at least works.

Attachment (imagetext.el): application/emacs-lisp, 31 KiB
Version: GnuPG v1.4.9 (GNU/Linux)

gnu-emacs-sources mailing list
gnu-emacs-sources <at>