10:08 a.m.
16 September 2005
Dear Stephen,
thank you very much for your email.
In your email, you wrote:
Ernst> When working with Xemacs, I normally can mark a region
with
Ernst> the left mouse button. This marked region is copied to the
Ernst> kill ring and I should be able to paste this region to a
Ernst> different location in the file via the middle mouse
Ernst> button. However, in my case, the operation produces the
Ernst> correct result only if the kill ring had been empty.
Do you mean that the contents of the kill ring, instead of the
marked region (in X11 terms, the primary selection), are copied? or
what?
Ernst> Accordingly, after marking a region with the left
Ernst> mouse button, I have to explicitly give a command like
Ernst> Alt-w in order to copy the marked region into the kill
Ernst> ring. After that, pasting via the middle mouse button
Ernst> works.
Let me explain this by a simple example. Assume that we have in my
version of Xemacs a file with the following text:
Jo> 1111
Jo>
Jo> 2222
Jo>
Jo> 3333
Jo>
Jo> 4444
Let us also assume that I have not done any operation affecting the kill
ring (for example Alt-w or Ctl-w).
Now I mark 1111 with the left mouse button. After that, I move the cursor
behind 2222 by clicking the left mouse cursor immediately behind
2222. Then, by pressing the middle mouse button I obtain
Jo> 1111
Jo>
Jo> 22221111
Jo>
Jo> 3333
Jo>
Jo> 4444
Let us now start again from the same situation (again assuming that
nothing had been done that affects the kill ring):
Jo> 1111
Jo>
Jo> 2222
Jo>
Jo> 3333
Jo>
Jo> 4444
Now, I mark 4444 with the left mouse button and type Ctl-w, which copies
4444 to the kill wring. Then, I again mark 1111 via the left mouse
button. If I now move behind 2222 as in the previous example, and click
with the middle mouse button, I get instead:
Jo> 1111
Jo>
Jo> 22224444
Jo>
Jo> 3333
Jo>
Jo>
Thus, if the kill ring is not empty, the primary selection is not copied
to the kill ring and the previous content of the kill ring is copied.
Accordingly, the functionality of the mouse is now severely restricted
which is quite annoying.
I hope this clarifies the situation.
Then you also asked:
X11 primary selection, X11 clipboard, Motif clipboard, and Emacs
region semantics vary substantially. It's not possible to satisfy
them all simultaneously, so XEmacs selection treatment is
customizable. Please let me know what the contents of the variables
`interprogram-cut-function' and `interprogram-paste-function' are (C-h
v VARIABLE RET will tell you that).
The command
Ctl-h v interprogram-cut-function
yielded the following result
Jo> `interprogram-cut-function' is a variable declared in Lisp.
Jo> -- loaded from "/usr/share/xemacs/21.5.18/lisp/simple.elc"
Jo>
Jo> Value: own-clipboard
and the command
Ctl-h v interprogram-paste-function
yielded the following result
Jo> `interprogram-paste-function' is a variable declared in Lisp.
Jo> -- loaded from "/usr/share/xemacs/21.5.18/lisp/simple.elc"
Jo>
Jo> Value: get-clipboard-foreign
Thank you very much for your help.
Yours sincerely
Ernst Joachim Weniger
#-------------------------------------------------------|
| PD Dr. Ernst Joachim Weniger |
| Institut für Physikalische und Theoretische Chemie |
| Universität Regensburg |
| D-93040 Regensburg, Germany |
| Tel: +49 (0)941-943-4722 |
| Fax: +49 (0)941-943-4719 |
| Email: Joachim.Weniger(a)chemie.uni-regensburg.de |
#-------------------------------------------------------|
11:29 a.m.
New subject: Bug Report
Ernst Joachim Weniger <joachim.weniger(a)chemie.uni-regensburg.de> writes:
Let me explain this by a simple example. Assume that we have in my
version of Xemacs a file with the following text:
[...]
I cannot repeat this problem in XEmacs 21.4.17. For me middle button
pastes the most recent selection regardless of the state of the
kill-ring.
Which XEmacs version are you using? Does the same bug occur when you
start `xemacs -vanilla', which bypasses your init file and package
initialization?
9:10 a.m.
New subject: Bug Report
>>>> "Ernst" == Ernst Joachim Weniger
<joachim.weniger(a)chemie.uni-regensburg.de> writes:
Ernst> Let me explain this by a simple example.
OK, that makes things clear.
Ernst> Now, I mark 4444 with the left mouse button and type Ctl-w,
Ernst> which copies 4444 to the kill wring. Then, I again mark
Ernst> 1111 via the left mouse button. If I now move behind 2222
Ernst> as in the previous example, and click with the middle mouse
Ernst> button, I get instead: [[ "4444" after the "2222" ]].
I cannot reproduce this behavior in 21.5.21 (with additional Xft
patches, but that should not affect this), although I have the same
configuration for selection handling (interprogram-cut-function =
own-clipboard, interprogram-paste-function = get-selection-foreign)
that you do. I don't see anything suggestive of changes to that
feature in the ChangeLogs for any of the files I would expect such
bugs to occur.
Unfortunately, my development machine went up in a flash of light on
Friday, so it will be a few days before I can test a build of 21.5.18.
You could try
M-x customize-variable RET interprogram-paste-function RET
select the button that says Other, and edit the fill-in field to
"get-selection-foreign". Then push the Set button. I suspect this
will cause other problems, and possibly even generate errors, but the
only other option is to update XEmacs to the most recent version,
either 21.4.17 (stable) or 21.5.22 (in development). (Upgrading is
what I recommend, but of course depends on your skills and time
constraints.)
If, on the contrary, after some experimentation it seems to be working
better for you, you can return to that buffer and push the "Save"
button to get the same behavior for future sessions.
Regards,
Steve
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
9:38 a.m.
New subject: Bug Report
On Mon, 19 Sep 2005, Stephen J. Turnbull wrote:
Date: Mon, 19 Sep 2005 17:10:49 +0900
From: Stephen J. Turnbull <turnbull(a)sk.tsukuba.ac.jp>
To: Ernst Joachim Weniger <joachim.weniger(a)chemie.uni-regensburg.de>
Cc: xemacs-beta(a)xemacs.org, mike fabian <mfabian(a)suse.de>,
John <wguy4biz(a)comcast.net>
Subject: Re: Bug Report
>>>>> "Ernst" == Ernst Joachim Weniger
<joachim.weniger(a)chemie.uni-regensburg.de> writes:
Ernst> Let me explain this by a simple example.
OK, that makes things clear.
Ernst> Now, I mark 4444 with the left mouse button and type Ctl-w,
Ernst> which copies 4444 to the kill wring. Then, I again mark
Ernst> 1111 via the left mouse button. If I now move behind 2222
Ernst> as in the previous example, and click with the middle mouse
Ernst> button, I get instead: [[ "4444" after the "2222" ]].
I cannot reproduce this behavior in 21.5.21 (with additional Xft
patches, but that should not affect this), although I have the same
configuration for selection handling (interprogram-cut-function =
own-clipboard, interprogram-paste-function = get-selection-foreign)
that you do. I don't see anything suggestive of changes to that
feature in the ChangeLogs for any of the files I would expect such
bugs to occur.
Unfortunately, my development machine went up in a flash of light on
Friday, so it will be a few days before I can test a build of 21.5.18.
You could try
M-x customize-variable RET interprogram-paste-function RET
select the button that says Other, and edit the fill-in field to
"get-selection-foreign". Then push the Set button. I suspect this
will cause other problems, and possibly even generate errors, but the
only other option is to update XEmacs to the most recent version,
either 21.4.17 (stable) or 21.5.22 (in development). (Upgrading is
what I recommend, but of course depends on your skills and time
constraints.)
If, on the contrary, after some experimentation it seems to be working
better for you, you can return to that buffer and push the "Save"
button to get the same behavior for future sessions.
Regards,
Steve
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
19 September 2005
Dear Steve,
I will try the things which you suggested, but I am somewhat pessimistic
because the problems occur only under KDE but not under Gnome. For quite
a few years, I have been using Xemacs in connection with KDE (since 1999
and not 1000 as I wrongly wrote in my bug report). Therefore, I would
prefer to continue working under KDE and avoid changing to Gnome (in
that respect I am conservative).
I have already been considering to upgrade Xemacs, and I already
downloaded the files
xemacs-mule-sumo-2005-07-15.tar.gz
and
xemacs-sumo-2005-07-15.tar.gz
which should do the job.
Or do you have a better suggestion for me? Please take into account that
I am not a computer scientist, but rather a mathematically orientists
who uses computers as admittedly very important tools in order to get
work done. Thus, I am not primarily interested in computers and my
knowledge is limited, although I am certainly not completely ignorant.
In any way, let me thank you very much for your efforts and your help.
Ernst Joachim
#-------------------------------------------------------|
| PD Dr. Ernst Joachim Weniger |
| Institut für Physikalische und Theoretische Chemie |
| Universität Regensburg |
| D-93040 Regensburg, Germany |
| Tel: +49 (0)941-943-4722 |
| Fax: +49 (0)941-943-4719 |
| Email: Joachim.Weniger(a)chemie.uni-regensburg.de |
| Private Phone: +49 {0)89-647377 |
#-------------------------------------------------------|
10:22 a.m.
New subject: Bug Report
On Mon, 19 Sep 2005, Stephen J. Turnbull wrote:
Date: Mon, 19 Sep 2005 17:10:49 +0900
From: Stephen J. Turnbull <turnbull(a)sk.tsukuba.ac.jp>
To: Ernst Joachim Weniger <joachim.weniger(a)chemie.uni-regensburg.de>
Cc: xemacs-beta(a)xemacs.org, mike fabian <mfabian(a)suse.de>,
John <wguy4biz(a)comcast.net>
Subject: Re: Bug Report
>>>>> "Ernst" == Ernst Joachim Weniger
<joachim.weniger(a)chemie.uni-regensburg.de> writes:
Ernst> Let me explain this by a simple example.
OK, that makes things clear.
Ernst> Now, I mark 4444 with the left mouse button and type Ctl-w,
Ernst> which copies 4444 to the kill wring. Then, I again mark
Ernst> 1111 via the left mouse button. If I now move behind 2222
Ernst> as in the previous example, and click with the middle mouse
Ernst> button, I get instead: [[ "4444" after the "2222" ]].
I cannot reproduce this behavior in 21.5.21 (with additional Xft
patches, but that should not affect this), although I have the same
configuration for selection handling (interprogram-cut-function =
own-clipboard, interprogram-paste-function = get-selection-foreign)
that you do. I don't see anything suggestive of changes to that
feature in the ChangeLogs for any of the files I would expect such
bugs to occur.
Unfortunately, my development machine went up in a flash of light on
Friday, so it will be a few days before I can test a build of 21.5.18.
You could try
M-x customize-variable RET interprogram-paste-function RET
select the button that says Other, and edit the fill-in field to
"get-selection-foreign". Then push the Set button. I suspect this
will cause other problems, and possibly even generate errors, but the
only other option is to update XEmacs to the most recent version,
either 21.4.17 (stable) or 21.5.22 (in development). (Upgrading is
what I recommend, but of course depends on your skills and time
constraints.)
If, on the contrary, after some experimentation it seems to be working
better for you, you can return to that buffer and push the "Save"
button to get the same behavior for future sessions.
Regards,
Steve
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
19 September 2005
Dear Steve, I tried out your suggestion
M-x customize-variable RET interprogram-paste-function RET
followed by "get-selection-foreign". Unfortunately, this does not seem
to work.
Best regards
Ernst Joachim
#-------------------------------------------------------|
| PD Dr. Ernst Joachim Weniger |
| Institut für Physikalische und Theoretische Chemie |
| Universität Regensburg |
| D-93040 Regensburg, Germany |
| Tel: +49 (0)941-943-4722 |
| Fax: +49 (0)941-943-4719 |
| Email: Joachim.Weniger(a)chemie.uni-regensburg.de |
| Private Phone: +49 {0)89-647377 |
#-------------------------------------------------------|
4:39 p.m.
New subject: Bug Report
>>>> "Ernst" == Ernst Joachim Weniger
<joachim.weniger(a)chemie.uni-regensburg.de> writes:
Ernst> M-x customize-variable RET interprogram-paste-function RET
Ernst> followed by "get-selection-foreign". Unfortunately, this
Ernst> does not seem to work.
Well, it was a long shot.
I would say your best bet is if SuSE updates their package for 21.5.
While building XEmacs from scratch is not hard as such, it sounds like
you are used to the configuration SuSE provides and it might be
tedious to reproduce that, as some new configuration possibilities
have been added since 21.5.18. You would also have to install
compilers and the development versions of the libraries used, which is
rather tedious if you don't have a program build environment already.
It's hard to recommend it for those reasons.
Upgrading the LISP packages (from the SUMO tarballs you have
downloaded) may be a good idea for other reasons, but it is very
unlikely to solve this issue, because selection handling is a "core"
function; you'd need to get the core source code from
ftp://ftp.xemacs.org/pub/xemacs/beta/xemacs-21.5.22.tar.gz
and compile it.
The remaining possibility would be to get the XEmacs 21.4.17 package
from SuSE. This is the simplest, but you may have some problems with
Unicode, which has become more pervasive in recent releases.
--
Graduate School of Systems and Information Engineering University of Tsukuba
http://turnbull.sk.tsukuba.ac.jp/ Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Economics of Information Communication and Computation Systems
Experimental Economics, Microeconomic Theory, Game Theory
7004
days inactive
7007
days old
5 comments
3 participants
participants (3)
-
Ernst Joachim Weniger -
Hrvoje Niksic -
Stephen J. Turnbull