Please pretest Emacs 21.3

Please pretest Emacs 21.3

[Richard Stallman]

For the next couple of weeks, would people please start using the RC
branch (which will be released as 21.3) rather than the trunk?  We
would like to make 21.3 as stable as possible.

Please send mail to the list if you use the RC branch.

Eli, can you send precise instructions for how to check out this
branch?

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On Sun, 26 May 2002, Richard Stallman wrote:

> Eli, can you send precise instructions for how to check out this
> branch?

Like this:

	cvs ... up -r EMACS_21_1_RC -d emacs

Instead of the ellipsis, put your normal CVS arguments (for me, it's
"-z3 -d eliz@subversions.gnu.org:/cvsroot/emacs").

Re: Please pretest Emacs 21.3

[Andreas Schwab]

Eli Zaretskii <eliz@is.elta.co.il> writes:

|> On Sun, 26 May 2002, Richard Stallman wrote:
|> 
|> > Eli, can you send precise instructions for how to check out this
|> > branch?
|> 
|> Like this:
|> 
|> 	cvs ... up -r EMACS_21_1_RC -d emacs

Please add a tag whenever you make a pretest.  I noticed that you didn't
tag the 21.2 release either, which is unfortunate.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Please pretest Emacs 21.3

[nferrier]

I've just updated my CVS to the branch and run "make". The build fails
with the following error:

make[1]: *** No rule to make target `/usr/local/src/emacs-cvs/src/../lisp/language/devanagari.elc', needed by `../etc/DOC'.  Stop.
make[1]: Leaving directory `/usr/local/src/emacs-cvs/src'


I'll try checking out the source again and rebuilding.


Nic Ferrier

Re: Please pretest Emacs 21.3

[Andreas Schwab]

nferrier@tapsellferrier.co.uk writes:

|> I've just updated my CVS to the branch and run "make". The build fails
|> with the following error:
|> 
|> make[1]: *** No rule to make target `/usr/local/src/emacs-cvs/src/../lisp/language/devanagari.elc', needed by `../etc/DOC'.  Stop.
|> make[1]: Leaving directory `/usr/local/src/emacs-cvs/src'

Use "make bootstrap".

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Please pretest Emacs 21.3

[nferrier]

Andreas Schwab <schwab@suse.de> writes:

> nferrier@tapsellferrier.co.uk writes: 
>  
> |> I've just updated my CVS to the branch and run "make". The build fails 
> |> with the following error: 
>  
> Use "make bootstrap". 

Sorry, I didn't realise that had to be used for all builds...

I've done that and I now get the following error during build of my
existing CVS tree (updated to the branch).

./temacs --batch --load loadup bootstrap
emacs: Invalid hash table rehash threshold: 134217728.0
make[1]: *** [bootstrap-emacs] Error 1
make[1]: Leaving directory `/usr/local/src/emacs-cvs/src'
make: *** [bootstrap-src] Error 2


I'll check out a new copy and have a go with that... but I guess
existing CVS working copies should really build. Shouldn't they?

One thing I haven't tried is a 'make clean' (or some other clean
target). Should I try that?



Nic Ferrier

Re: Please pretest Emacs 21.3

[nferrier]

nferrier@tapsellferrier.co.uk writes:

> Andreas Schwab <schwab@suse.de> writes: 
>  
> > nferrier@tapsellferrier.co.uk writes:  
> >   
> > |> I've just updated my CVS to the branch and run "make". The build fails  
> > |> with the following error:  
> >   
> > Use "make bootstrap".  
>  
> Sorry, I didn't realise that had to be used for all builds... 
>  
> I've done that and I now get the following error during build of my 
> existing CVS tree (updated to the branch). 
>  
> ./temacs --batch --load loadup bootstrap 
> emacs: Invalid hash table rehash threshold: 134217728.0 
> make[1]: *** [bootstrap-emacs] Error 1 
> make[1]: Leaving directory `/usr/local/src/emacs-cvs/src' 
> make: *** [bootstrap-src] Error 2 
>  
>  
> I'll check out a new copy and have a go with that... but I guess 
> existing CVS working copies should really build. Shouldn't they? 
>  

Fresh checkout does exactly the same thing as my updated copy:

./temacs --batch --load loadup bootstrap
emacs: Invalid hash table rehash threshold: 134217728.0
make[1]: *** [bootstrap-emacs] Error 1
make[1]: Leaving directory `/usr/local/src/emacs/src'
make: *** [bootstrap-src] Error 2


Nic

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On Mon, 27 May 2002, Andreas Schwab wrote:

> Please add a tag whenever you make a pretest.

I intended to tag the released version only.

Why is it important to tag a pretest, if we don't create a branch?  (Any 
additional CVS command I need to issue as part of the pretest preparation 
is a nuisance, something I could forget or mess up.  So I'd like to avoid 
anything that is not necessary.)

Re: Please pretest Emacs 21.3

[Andreas Schwab]

Eli Zaretskii <eliz@is.elta.co.il> writes:

|> On Mon, 27 May 2002, Andreas Schwab wrote:
|> 
|> > Please add a tag whenever you make a pretest.
|> 
|> I intended to tag the released version only.
|> 
|> Why is it important to tag a pretest, if we don't create a branch?

So that people can know what went into the pretest without having to
download the tarball.

|> (Any 
|> additional CVS command I need to issue as part of the pretest preparation 
|> is a nuisance, something I could forget or mess up.  So I'd like to avoid 
|> anything that is not necessary.)

That's what shell scripts are for.  It's always better to create too many
tags than to create too few, because retroactive tagging is a pain.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Please pretest Emacs 21.3

[D. Goel]

Greetings

if 21.3 has not been tested enough already, i shall try to install it
on a windoze 2000 (never instaled emacs on windows before..) and on a
sunsparc soon..


if anyone happens to have detailed instructions for a newbie, please
send them to me, else: no biggie :) 


Does one really need to have a shell-account with savannah in order to
be able to check out emacs21.3?  If so, does one really need a PGP or
GPG working to be able to do that? 

+----
| 
| Please send mail to the list if you use the RC branch.
+----

(PS: i don't know what RC or CVS is.. is that important to know? ) 

> On Sun, 26 May 2002, Richard Stallman wrote:
> 
> > Eli, can you send precise instructions for how to check out this
> > branch?
> 
> Like this:
> 
> 	cvs ... up -r EMACS_21_1_RC -d emacs
> 
> Instead of the ellipsis, put your normal CVS arguments (for me, it's
> "-z3 -d eliz@subversions.gnu.org:/cvsroot/emacs").
> 
> _______________________________________________
> Emacs-devel mailing list
> Emacs-devel@gnu.org
> http://mail.gnu.org/mailman/listinfo/emacs-devel


Have a good day,

DG				   <http://www.glue.umd.edu/~deego/>
-- 

Re: Please pretest Emacs 21.3

[Miles Bader]

"D. Goel" <deego@glue.umd.edu> writes:
> Does one really need to have a shell-account with savannah in order to
> be able to check out emacs21.3?  If so, does one really need a PGP or
> GPG working to be able to do that? 

No, you shouldn't need any of that.

See the web page

   http://savannah.gnu.org/cvs/?group_id=40

for more details on how to check emacs out of the CVS repository.
Anonymous access is probably fine for this usage.

When you give the `cvs co' command, you should add Eli's
`-r EMACS_21_1_RC' option just after the `co' part of the command line
[on the web page, there's other junk between `cvs' and `co', which you
still need].

> (PS: i don't know what RC or CVS is.. is that important to know? ) 

Um, a bit.  You at least have to make sure you have CVS installed!
If you can type `cvs --help' and get an answer, then that's probably
enough.

-Miles
-- 
Come now, if we were really planning to harm you, would we be waiting here, 
 beside the path, in the very darkest part of the forest?

Re: Please pretest Emacs 21.3

[D. Goel]

Re: my last post re: difficulty downloading..  please ignore it, i guess--->

i managed to check it out finally.. the README says 21.3.50..

(i placed it online to assist me in transferring it to various
places.. and if there's more cvs-phobic newbies like me out there,
they too can find it on: http://www.glue.umd.edu/~deego/emacspub
as a ./tar.gz. )

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 29 May 2002, D. Goel wrote:

> if 21.3 has not been tested enough already, i shall try to install it
> on a windoze 2000 (never instaled emacs on windows before..) and on a
> sunsparc soon..

Yes, please.  Thanks!

Re: Please pretest Emacs 21.3

[D. Goel]

Eli

i followed these instructions (that i read on savannah) from an
account on a dec-alpha:

+----
| 
| cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/emacs
| login 
|  
| cvs -z3 -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/emacs co
| emacs
+----

are these okay?   the README in the emacs/ now mentions version
21.3.50...  
(this was from a unix account)

[2] i was curious what the instructions are from a windows (2000)
box..  the command cvs does not work there..  does that mean i should
install cygwin first?  (i am trying to tar up the downloaded unix
version to ftp it to windows but keep exceeding my quota :)..

Re: Please pretest Emacs 21.3

[Miles Bader]

"D. Goel" <deego@glue.umd.edu> writes:
> +----
> | 
> | cvs -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/emacs
> | login 
> |  
> | cvs -z3 -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/emacs co
> | emacs
> +----
> 
> are these okay?   the README in the emacs/ now mentions version
> 21.3.50...  

That's the `HEAD' version.  To get the prestest, you should use this
instead for the second command:

   cvs -z3 -d:pserver:anoncvs@subversions.gnu.org:/cvsroot/emacs co -rEMACS_21_1_RC emacs
 
However, since you've already got an emacs directory checked out, you
can just cd there, and use `update' to do the work, like:

   cd emacs
   cvs update -rEMACS_21_1_RC -d -P

-Miles
-- 
Would you like fries with that?

Re: Please pretest Emacs 21.3

[D. Goel]

> 
>    cd emacs
>    cvs update -rEMACS_21_1_RC -d -P
> 

Miles

thanks.. i did this..  so i guess this dir emacs/ now contains what i
would have gotten if i followed the proper CVS instructions.. 

(BTW, the new README in emacs/ now mentions version 21.2...  this need
not worry me right?  this is just the 'tag' issue being discussed in
the list.. the actual version should still be 21.3 right?  )



> -Miles -- Would you like fries with that?
> 
> _______________________________________________
> Emacs-devel mailing list
> Emacs-devel@gnu.org
> http://mail.gnu.org/mailman/listinfo/emacs-devel


Have a good day,

DG				   <http://www.glue.umd.edu/~deego/>
-- 

Re: Please pretest Emacs 21.3

[Miles Bader]

"D. Goel" <deego@glue.umd.edu> writes:
> BTW, the new README in emacs/ now mentions version 21.2...  this need
> not worry me right?

It shouldn't worry you; I see the same thing.  I think the script which
is used to make the actual distribution automatically ensures that the
correct version is used.

> this is just the 'tag' issue being discussed in
> the list.. the actual version should still be 21.3 right?

You can check for sure doing `M-x emacs-version'.  The pre-release
should have version 21.2.90.x.

-Miles
-- 
Next to fried food, the South has suffered most from oratory.
  			-- Walter Hines Page

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 30 May 2002, Miles Bader wrote:

> "D. Goel" <deego@glue.umd.edu> writes:
> > BTW, the new README in emacs/ now mentions version 21.2...  this need
> > not worry me right?
> 
> It shouldn't worry you; I see the same thing.  I think the script which
> is used to make the actual distribution automatically ensures that the
> correct version is used.

I should have checked in the files that have the version stated in them, 
but it seems like I've forgotten.  Sorry.

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 29 May 2002, D. Goel wrote:

> (BTW, the new README in emacs/ now mentions version 21.2...

Ugh, sounds like I've forgotten to check-in another file after updating 
the version.  Should be fixed now.

Re: Please pretest Emacs 21.3

[Miles Bader]

Richard Stallman <rms@gnu.org> writes:
> Please send mail to the list if you use the RC branch.

I'm using it (though it claims to be version 21.2).

-Miles
-- 
.Numeric stability is probably not all that important when you're guessing.

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 27 May 2002, Miles Bader wrote:

> Richard Stallman <rms@gnu.org> writes:
> > Please send mail to the list if you use the RC branch.
> 
> I'm using it (though it claims to be version 21.2).

It should be version 21.2.90, unless I screwed up when making the 
tarball.

Re: Please pretest Emacs 21.3

[Miles Bader]

Eli Zaretskii <eliz@is.elta.co.il> writes:
> > I'm using it (though it claims to be version 21.2).
> 
> It should be version 21.2.90, unless I screwed up when making the 
> tarball.

I'm not using the tarball, I'm using the CVS RC branch with the tag you
gave (EMACS_21_1_RC).

M-x emacs-version says:

  GNU Emacs 21.2.1 (i686-pc-linux-gnu, X toolkit, Xaw3d scroll bars) of 2002-05-27 on mcspd15

-Miles
-- 
"I distrust a research person who is always obviously busy on a task."
   --Robert Frosch, VP, GM Research

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 27 May 2002, Miles Bader wrote:

> > It should be version 21.2.90, unless I screwed up when making the 
> > tarball.
> 
> I'm not using the tarball, I'm using the CVS RC branch with the tag you
> gave (EMACS_21_1_RC).

Sorry, I failed to commit the modified version.el.  Should be fixed now.

Re: Please pretest Emacs 21.3

[Miles Bader]

I notice that the RC for 21.3 contains the lisp reference manual (in the
lispref/ subdir), but neither Makefile.in nor configure.in contain any
rules for dealing with it (whereas the CVS head versions do).

Is this intentional?

-Miles
-- 
Somebody has to do something, and it's just incredibly pathetic that it
has to be us.  -- Jerry Garcia

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On 27 May 2002, Miles Bader wrote:

> I notice that the RC for 21.3 contains the lisp reference manual (in the
> lispref/ subdir), but neither Makefile.in nor configure.in contain any
> rules for dealing with it (whereas the CVS head versions do).
> 
> Is this intentional?

It's not intentional, but I don't know how to prevent this from 
happening.  The problems seems to be that, once a directory exists on 
the trunk, it will exist on the branches as well.

The branch version of make-dist makes sure the lispref directory is not 
present in the tarball, so the lack of Makefile's doesn't matter for the 
users.

If someone can tell how to persuade CVS not to make lispref visible on 
the branch, I'd be grateful.

Re: Please pretest Emacs 21.3

[Andreas Schwab]

Eli Zaretskii <eliz@is.elta.co.il> writes:

|> On 27 May 2002, Miles Bader wrote:
|> 
|> > I notice that the RC for 21.3 contains the lisp reference manual (in the
|> > lispref/ subdir), but neither Makefile.in nor configure.in contain any
|> > rules for dealing with it (whereas the CVS head versions do).
|> > 
|> > Is this intentional?
|> 
|> It's not intentional, but I don't know how to prevent this from 
|> happening.  The problems seems to be that, once a directory exists on 
|> the trunk, it will exist on the branches as well.

You just have to remove the branch tags from the files that are not
supposed to be on the branch.

|> If someone can tell how to persuade CVS not to make lispref visible on 
|> the branch, I'd be grateful.

The directory itself will always be visible, unless you use "cvs update
-P" to let cvs prune it.

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On Mon, 27 May 2002, Andreas Schwab wrote:

> You just have to remove the branch tags from the files that are not
> supposed to be on the branch.

There are no files on the branch that aren't supposed to be there.  My 
problem is only with directories.

> |> If someone can tell how to persuade CVS not to make lispref visible on 
> |> the branch, I'd be grateful.
> 
> The directory itself will always be visible, unless you use "cvs update
> -P" to let cvs prune it.

Thanks.  This is a user-level command; I thought there was some way of 
doing it for everyone, by issuing some CVS command once.

Re: Please pretest Emacs 21.3

[Andreas Schwab]

Eli Zaretskii <eliz@is.elta.co.il> writes:

|> On Mon, 27 May 2002, Andreas Schwab wrote:
|> 
|> > You just have to remove the branch tags from the files that are not
|> > supposed to be on the branch.
|> 
|> There are no files on the branch that aren't supposed to be there.  My 
|> problem is only with directories.

But all the files in lispref are not part of emacs (yet), are they?

Andreas.

-- 
Andreas Schwab, SuSE Labs, schwab@suse.de
SuSE GmbH, Deutschherrnstr. 15-19, D-90429 Nürnberg
Key fingerprint = 58CA 54C7 6D53 942B 1756  01D3 44D5 214B 8276 4ED5
"And now for something completely different."

Re: Please pretest Emacs 21.3

[Eli Zaretskii]

On Mon, 27 May 2002, Andreas Schwab wrote:

> |> There are no files on the branch that aren't supposed to be there.  My 
> |> problem is only with directories.
> 
> But all the files in lispref are not part of emacs (yet), are they?

Yes, they are.  They are there because the ELisp manual is maintained as 
part of the Emacs CVS tree, and versions 21.1 and 21.2 went into press 
with suitably updated releases of the ELisp manual.

The problem is that on the branch, the ELisp manual is not built as part 
of the Emacs build procedure (unlike Leim, for example).

Re: Please pretest Emacs 21.3

[Stefan Monnier]

> > |> If someone can tell how to persuade CVS not to make lispref visible on 
> > |> the branch, I'd be grateful.
> > 
> > The directory itself will always be visible, unless you use "cvs update
> > -P" to let cvs prune it.
> 
> Thanks.  This is a user-level command; I thought there was some way of 
> doing it for everyone, by issuing some CVS command once.

There isn't.  CVS (mis)handles directories a bit like early `tar'
commands: they are not versioned and are only used in order to place
files into them.
Most people should have "update -P" and "checkout -P" in their .cvsrc.


	Stefan

Re: Please pretest Emacs 21.3

[Kai Großjohann]

Richard Stallman <rms@gnu.org> writes:

> Please send mail to the list if you use the RC branch.

I use the RC branch personally and have installed the tarball for my
colleagues.  No problems (except for local configuration issues)
encountered.  Running on GNU/Linux x86, so nothing unusual, I'm afraid.

kai
-- 
Silence is foo!

Fonts and Latin-15

[Alex Schroeder]

When Kai sends mail these days, his name is encoded in Latin-15
instead of Latin-1 like previously.  On my current system this ß looks
"bad" because it is not rendered in the default font.  It is a bit
smaller and a bit more to the left than my Latin-1 ß.

Mine, looks good:

  character: ß (04337, 2271, 0x8df)
    charset: latin-iso8859-1 (Right-Hand Part of Latin Alphabet 1 (ISO/IEC 8859-1): ISO-IR-100)
 code point: 95
     syntax: word
   category: l:Latin  
buffer code: 0x81 0xDF
  file code: 0x81 0xDF (encoded by coding system emacs-mule)
       font: -Misc-Fixed-Medium-R-Normal--14-130-75-75-C-70-ISO8859-1

His, looks bad:

  character: ß (07537, 3935, 0xf5f)
    charset: latin-iso8859-15 (Right-Hand Part of Latin Alphabet 9 (ISO/IEC 8859-15): ISO-IR-203)
 code point: 95
     syntax: word
   category: l:Latin  
buffer code: 0x8E 0xDF
  file code: 0x8E 0xDF (encoded by coding system emacs-mule)
       font: -Misc-Fixed-Bold-R-Normal--13-120-75-75-C-70-ISO8859-15

Is this my problem (I don't have enough Latin-15 fonts installed) or
Emacs' problem (because it is choosing a different font with a
different size AFAICT)?

My locale is this:

~% locale 
LANG=de_CH.ISO-8859-1
LC_CTYPE="de_CH.ISO-8859-1"
LC_NUMERIC="de_CH.ISO-8859-1"
LC_TIME="de_CH.ISO-8859-1"
LC_COLLATE="de_CH.ISO-8859-1"
LC_MONETARY="de_CH.ISO-8859-1"
LC_MESSAGES="de_CH.ISO-8859-1"
LC_ALL=

And my .Xresources contains:

*.font:		7x14
Emacs.paneFont:	7x14

This is on the GNU Emacs 21.2.90.1 (i686-pc-linux-gnu, X toolkit,
Xaw3d scroll bars) of 2002-05-28 on confusibombus I built yesterday,
but I think this already appeared in Emacs 21.2.  But then again,
perhaps Kai used good ol' Latin-1 before that...  :)

Alex.
-- 
http://www.electronicintifada.net/diaries/index.html
http://www.us-israel.org/jsource/US-Israel/hr2506c.html

Re: Fonts and Latin-15

[Kai Großjohann]

Alex Schroeder <alex@emacswiki.org> writes:

> Is this my problem (I don't have enough Latin-15 fonts installed) or
> Emacs' problem (because it is choosing a different font with a
> different size AFAICT)?

I'm sure that Emacs _tries_ to choose similar fonts.  For me, I
installed the xfonts-foo-transcoded Debian packages, for foo in 75dpi,
100dpi, misc and possibly others.  Now the fonts look nice.  (Before I
had similar problems with Latin-2 characters sticking out like sore
thumbs.)

kai
-- 
Silence is foo!

Re: Fonts and Latin-15

[Stefan Monnier]

>   character: ß (04337, 2271, 0x8df)
>     charset: latin-iso8859-1 (Right-Hand Part of Latin Alphabet 1 (ISO/IEC 8859-1): ISO-IR-100)
>  code point: 95
>      syntax: word
>    category: l:Latin  
> buffer code: 0x81 0xDF
>   file code: 0x81 0xDF (encoded by coding system emacs-mule)
>        font: -Misc-Fixed-Medium-R-Normal--14-130-75-75-C-70-ISO8859-1
> 
> His, looks bad:
> 
>   character: ß (07537, 3935, 0xf5f)
>     charset: latin-iso8859-15 (Right-Hand Part of Latin Alphabet 9 (ISO/IEC 8859-15): ISO-IR-203)
>  code point: 95
>      syntax: word
>    category: l:Latin  
> buffer code: 0x8E 0xDF
>   file code: 0x8E 0xDF (encoded by coding system emacs-mule)
>        font: -Misc-Fixed-Bold-R-Normal--13-120-75-75-C-70-ISO8859-15

If you have a -Misc-Fixed-Medium-R-Normal--14-130-75-75-C-70-ISO8859-15 then
there's a bug in Emacs.  Otherwise, it's just that you don't have the
necessary font.  Use `xlsfonts' to find out.
You can also use unify-8859-on-decoding-mode to force all those latin-N ß
chars to be unified into the one and only latin-1 ß.


	Stefan

Re: Fonts and Latin-15

[Miles Bader]

"Stefan Monnier" <monnier+gnu/emacs@RUM.cs.yale.edu> writes:
> You can also use unify-8859-on-decoding-mode to force all those latin-N ß
> chars to be unified into the one and only latin-1 ß.

I used to do this, because otherwise the funny `i' in Pavel's name was
displayed in a ludicrously large font (like double the height of normal
characters!).

The problem was then whenever I edited a ChangeLog file, I'd end up
re-encoding the file in latin-1, which caused complaints.

I think what I _really_ want is something like `unify-8859-on-display-mode',
which wouldn't affect anything except display.

-Miles
-- 
Yo mama's so fat when she gets on an elevator it HAS to go down.

Re: Fonts and Latin-15

[Stefan Monnier]

> I think what I _really_ want is something like `unify-8859-on-display-mode',
> which wouldn't affect anything except display.

Maybe setting standard-display-table to ucs-mule-to-mule-unicode
would do ?


	Stefan "doubtful"

Re: Fonts and Latin-15

[Richard Stallman]

    I think what I _really_ want is something like `unify-8859-on-display-mode',
    which wouldn't affect anything except display.

That sounds like a good idea, if it is feasible.

Re: Please pretest Emacs 21.3

[Alex Schroeder]

Richard Stallman <rms@gnu.org> writes:

> Please send mail to the list if you use the RC branch.
> Eli, can you send precise instructions for how to check out this
> branch?

I am using the Emacs checked out as specified by Eli and have no
problems so far.  This is on a Debian stable, mostly using Gnus from
CVS (not Emacs CVS -- the real Gnus CVS).

Alex.
-- 
http://www.electronicintifada.net/diaries/index.html
http://www.us-israel.org/jsource/US-Israel/hr2506c.html

Re: Please pretest Emacs 21.3

[Juanma Barranquero]

On Sun, 26 May 2002 19:54:04 -0600 (MDT), Richard Stallman <rms@gnu.org> wrote:

> Please send mail to the list if you use the RC branch.

I'm using EMACS_21_1_RC on Windows 2K, and the pretest tarball on W2K
and RedHat Linux 7.2. Not a problem in sight, either compiling or using
them, except the bug/feature mentioned in another message about
w32-select-font selecting a -iso10646-1 font instead of a -iso8859-1 one.

And that's not a regression wrt 21.2 (it is wrt 21.1, though).


                                                           /L/e/k/t/u

Re: Please pretest Emacs 21.3

[Jason Rumney]

Juanma Barranquero <lektu@terra.es> writes:

> On Sun, 26 May 2002 19:54:04 -0600 (MDT), Richard Stallman <rms@gnu.org> wrote:
> 
> > Please send mail to the list if you use the RC branch.
> 
> I'm using EMACS_21_1_RC on Windows 2K, and the pretest tarball on W2K
> and RedHat Linux 7.2. Not a problem in sight, either compiling or using
> them, except the bug/feature mentioned in another message about
> w32-select-font selecting a -iso10646-1 font instead of a -iso8859-1 one.

What is the actual bug that this causes?  I didn't see any mention of
incorrect characters or fonts being displayed, or characters not
being displayed at all as a result of this change in behavior.
AFAIK the only change in behavior that this change causes is that
Emacs now displays latin-9 and mule-unicode-* characters out of the
box on Windows, whereas in 21.1 it was necessary to change
w32-charset-info-alist yourself (which also had the side effect of
changing the charset returned by w32-select-font for "Western" fonts).

Re: Please pretest Emacs 21.3

[Juanma Barranquero]

[-- Attachment #1: Type: text/plain, Size: 893 bytes --]

On 12 Jun 2002 10:32:11 +0100, Jason Rumney <jasonr@gnu.org> wrote:

> What is the actual bug that this causes?  I didn't see any mention of
> incorrect characters or fonts being displayed, or characters not
> being displayed at all as a result of this change in behavior.

On my machine, with -q --no-site-file and no display-specific resource
in HKLM\SOFTWARE\GNU\Emacs (I mean, no Emacs.Font, Emacs.Fontset-X,
Emacs.Geometry, etc.).

(insert (w32-select-font)) =>

-outline-Courier New-normal-r-normal-normal-13-97-96-96-c-*-iso10646-1

Copy&paste that into HKLM\S\G\E\Emacs.Font

emacs -q --no-site-file

The splash screen appears. If I do anything at all that causes a
redisplay (moving or resizing the frame, minimizing and restoring it,
whatever), the display is garbled.

I'm including a .jpg image of what I get.

                                                           /L/e/k/t/u

[-- Attachment #2: emacs-bug.jpg --]
[-- Type: image/jpeg, Size: 84578 bytes --]

Re: Please pretest Emacs 21.3

[Juanma Barranquero]

On Thu, 13 Jun 2002 08:51:05 +0200, Juanma Barranquero <lektu@terra.es> wrote:

> I'm including a .jpg image of what I get.

BTW, the .jpg is from a 21.3.50, but the same happens with 21.2 and 21.2.90.


                                                           /L/e/k/t/u