12:37 p.m.
Yoshiki Hayashi <yoshiki(a)xemacs.org> writes:
comp.emacs.xemacs is removed as I don't have access to net news
here.)
Hm, I'm not subscribed to xemacs-beta (this is not a really good idea
with MS-Outlook :-(, I can just send my mail with Emacs), but I can
check the list-archive on the web... But don't expect Reference
headers...
Christoph Wedler <Christoph.Wedler(a)sap.com> writes:
> 1. Also toggles eol type for new files (technical: switches from
eol
> type nil to crlf, of course not for coding systems like binary).
No, I don't think toggle-buffer-file-coding-system should change
the
default EOL type.
OK, I expected this. The point is that EOL type nil doesn't mean some
default EOL type (which you could customize somewhere), it means
auto-detect (if the info files are correct) and auto-detect when saving
the buffer means XEmacs' way which is the Unix file ending.
Thus, in the current XEmacs on Windows-NT, you have to do the following
to produce a file with CRLF endings:
1. open the file, write some text
2. save it
3. kill the buffer
4. load the same file
5. M-x toggle-buffer-file-coding-system
6. insert & delete a char
7. save it
I hope that everybody agrees that this is somewhat annoying.
Toggling the EOL type between 1 and 2 has no effect since we have the
EOL type nil whereas 5 works since then we have the EOL type lf.
Thus, I fail to see where "toggling" from nil to crlf is bad.
> 2. Sets buffer as modified (if not read-only). Why do you want
to
> toggle the eol type if you don't want to save the buffer?
I like it. I have long been wanted that
set-buffer-file-coding-system do the same but it didn't
annoyed me enough to submit a patch. :-)
This is in the Windows branch as Andy has said? Not XEmacs-21.4.5...
> BTW, any possibility to get ":dos" instead of
":T" in the modeline?
I think it is a good change. GNU Emacs prints (DOS) and (MAC) and
it
is easier to recognize than cryptic ':T'. (I don't know how LF is
represented.
":t". Yup, that's quite cryptic. But hey, I didn't make _that_
change. I don't see where it can be done in Lisp. But it should be
able to do this in Lisp, since some people want to have the space....
When you submit a patch, [...]
Talking of patches, xemacs-patches is again broken. I got no robot
reply for a couple of patches (OK, I let my first e-mails be spoiled by
some "Sender" header which Gnus thinks it had to add, but later ones
were without the Sender header).
BTW, is xemacs-beta the right place to report bugs to? The regexp bug
in XEmacs-21.4.5 leads to some strange behaviours...
To other remarks:
* "Toggling" Unix -> Dos -> Mac -> ... doesn't seem to be what
people
really need. Ok, there could be a variable... But it should affect
also the existing command M-x toggle-buffer-file-coding-system.
* EOL-Type menu: hm, a bit too much stuff for such a small thing (with
the button-2 thing I just wanted to avoid people asking me how to do
this in XEmacs...)
* To whether I should call toggle-buffer-file-coding-system in my mouse
funciton: yes, that's why I wanted to make the changes in
toggle-buffer-file-coding-system.
- Christoph
1:59 p.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
>>>> "CW" == Christoph Wedler
<Christoph.Wedler(a)sap.com> writes:
CW> auto-detect (if the info files are correct) and auto-detect
CW> when saving the buffer means XEmacs' way which is the Unix
CW> file ending.
Well, that should be fixed.
CW> Thus, I fail to see where "toggling" from nil to crlf is bad.
Of course, because you don't live where I do: on Unix. Now I have to
toggle again. That doesn't suck quite as much as the 6 step procedure
you mentioned, but it's not good.
CW> [Making buffer dirty on coding system change] is in the
CW> Windows branch as Andy has said? Not XEmacs-21.4.5...
Yes, and no. It will be in the next prerelease and in 21.4.6.
CW> Talking of patches, xemacs-patches is again broken. I got no
CW> robot reply for a couple of patches
No, you don't get a reply. If that was done before, it got broken
when Jason move the MLs from Tux to SourceForge. Sigh. I'd really
rather be working on releases than on mailbots.
CW> * "Toggling" Unix -> Dos -> Mac -> ... doesn't seem to
be
CW> what people really need.
More Windows-centric thinking. Yes, [X]Emacs is very Unix-centric,
and that bias should be reduced. But how about the Mac people?
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
7:06 p.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
At 01:37 PM 12/6/01 +0100, Christoph.Wedler(a)sap.com wrote:
> No, I don't think toggle-buffer-file-coding-system should
change the
> default EOL type.
OK, I expected this. The point is that EOL type nil doesn't mean some
default EOL type (which you could customize somewhere), it means
auto-detect (if the info files are correct) and auto-detect when saving
the buffer means XEmacs' way which is the Unix file ending.
Thus, in the current XEmacs on Windows-NT, you have to do the following
to produce a file with CRLF endings:
1. open the file, write some text
2. save it
3. kill the buffer
4. load the same file
5. M-x toggle-buffer-file-coding-system
6. insert & delete a char
7. save it
I hope that everybody agrees that this is somewhat annoying.
It is but its difficult to do something that works in all cases. I guess if
we only did it for windows-nt that would be somewhat ok. I specifically do
not want this to happen under cygwin.
":t". Yup, that's quite cryptic. But hey, I didn't
make _that_
change. I don't see where it can be done in Lisp. But it should be
able to do this in Lisp, since some people want to have the space....
Its in the C-code.
To other remarks:
* "Toggling" Unix -> Dos -> Mac -> ... doesn't seem to be what
people
really need. Ok, there could be a variable... But it should affect
Agreed.
also the existing command M-x toggle-buffer-file-coding-system.
Agreed.
* EOL-Type menu: hm, a bit too much stuff for such a small thing
(with
the button-2 thing I just wanted to avoid people asking me how to do
this in XEmacs...)
A menu is less cryptic, I would be concerned if this could happen just by
random clicking of the mouse.
* To whether I should call toggle-buffer-file-coding-system in my
mouse
funciton: yes, that's why I wanted to make the changes in
toggle-buffer-file-coding-system.
Right
andy
5:34 a.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
Andy Piper <andyp(a)bea.com> writes:
>Thus, in the current XEmacs on Windows-NT, you have to do the
following
>to produce a file with CRLF endings:
>
> 1. open the file, write some text
> 2. save it
> 3. kill the buffer
> 4. load the same file
> 5. M-x toggle-buffer-file-coding-system
> 6. insert & delete a char
> 7. save it
>
>I hope that everybody agrees that this is somewhat annoying.
It is but its difficult to do something that works in all cases. I guess if
we only did it for windows-nt that would be somewhat ok. I specifically do
not want this to happen under cygwin.
I agree the above procedure is bad but you can just do
(setq-default default-buffer-file-coding-system 'iso-8859-1-dos)
and save a lot of trouble. So I don't think it is a good
argument for toggle-buffer-file-coding-sytem changing
default EOL type.
>To other remarks:
>
> * "Toggling" Unix -> Dos -> Mac -> ... doesn't seem to be
what people
> really need. Ok, there could be a variable... But it should affect
Agreed.
Why not? If we have three EOL conversions, it should cycle
among them. Toggling between Unix and Dos is neglecting Mac
users. Remember XEmacs does work on Mac OS X (or so I've heard).
--
Yoshiki Hayashi
5:49 a.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
At 02:34 PM 12/7/01 +0900, Yoshiki Hayashi wrote:
Why not? If we have three EOL conversions, it should cycle
among them. Toggling between Unix and Dos is neglecting Mac
users. Remember XEmacs does work on Mac OS X (or so I've heard).
But if you are running on a PC then the chance of you coming across a
mac-style file is 1:10000000
andy
3:10 a.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
>>>> "Andy" == Andy Piper <andyp(a)bea.com>
writes:
> To other remarks:
>
> * "Toggling" Unix -> Dos -> Mac -> ... doesn't seem to be
what
> people really need. Ok, there could be a variable... But it
> should affect
I need it. I note from the addresses that you guys live in Corporate
Windows land. In academia, we see lots of Mac stuff, and I get
attachments from both Windows and Mac people.
I don't care what the default is, as long is the behavior I need is
configurable via Custom.
--
Institute of Policy and Planning Sciences http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Don't ask how you can "do" free software business;
ask what your business can "do for" free software.
5:07 p.m.
New subject: [Proposed Patch] button2 on ISO8:T in modeline toggles eol type
At 12:10 PM 12/10/2001 +0900, Stephen J. Turnbull wrote:
I need it. I note from the addresses that you guys live in Corporate
Windows land. In academia, we see lots of Mac stuff, and I get
attachments from both Windows and Mac people.
I don't care what the default is, as long is the behavior I need is
configurable via Custom.
Just for the record, there is obviously a world of difference between being
able to read the different line-endings (which we already do) and wanting
to author files with a different line-ending.
But yes, it should be customizable.
andy
8194
days inactive
8198
days old
6 comments
4 participants
participants (4)
-
Andy Piper -
Christoph Wedler -
Stephen J. Turnbull -
Yoshiki Hayashi