current build issues

Hey Sebastian!

Just tested with your p-86.  configure still dies I'm afraid.  It's
the same. (missing a `case')

--

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|       SXEmacs - The only _______ you'll ever need.       |
|         Fill in the blank, yes, it's THAT good!          |
|------------------------------------<steve <at> sxemacs.org>---|

configure broken (was: [Success] hroptatyr <at> sxemacs.org--sxemacs/sxemacs--hrop--22.1.6--patch-85 on i686-pc-linux-gnu)

[Re-directed to -devel (MFT set)]

* Sebastian Freundt <hroptatyr <at> sxemacs.org> writes:

  > Hm, since yacc, bison, flex and lex are not utterly necessary at the
  > moment, I suggest you kick out the yacc test in the configure.ac:

  > - MYAC_YACC_BISON
  > - AC_PROG_LEX
  > + dnl MYAC_YACC_BISON
  > + dnl AC_PROG_LEX

  > and try again. Oh, and apply p86 if you want no steenkin compiler warnings
  > :)

configure: WARNING: : accepted by the compiler, rejected by the preprocessor!
configure: WARNING: : proceeding with the compiler's result
../configure: line 17342: syntax error near unexpected token `;;'
../configure: line 17342: `    ;;'

Compilation exited abnormally with code 2 at Wed Nov  1 21:35:49

--

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|       SXEmacs - The only _______ you'll ever need.       |
|         Fill in the blank, yes, it's THAT good!          |
|------------------------------------<steve <at> sxemacs.org>---|

Re: [Success] steve <at> sxemacs.org--2006/sxemacs--main--22.1.6--patch-87 on i686-pc-linux-gnu

Steve Youngs <steve <at> sxemacs.org> writes:

> This is actually the same build from my last build-report, this time
> after installation.
>
> First up, `make DESTDIR=/some/pkging/dir install' works!!  Very happy
> about that. 
>
> OK, after installation, it finds more things... I now have toolbar
> glyphs for example.  And the load-path looks fine.
>
> Good news always has bad news just around the corner... the scratch
> buffer doesn't get the initial-scratch-message inserted.  And
> worse... 
>
>   C-x V initial-scratch-message RET
>   => The library \
>  
"/home/steve/programming/SXEmacs/core/sxemacs--main--22.1.6/=bld-FHS-automake/lisp/startup" \
>    is not in the path.
>
> But C-x 4 l (find-lib-other-win) startup RET works.  It finds the
> installed startup.el in /usr/local/share/sxemacs-22.1.6/lisp/

Indeed. Afaik, it still uses ${prefix} internally and make DESTDIR= just
redirects to some other dir _without_ adapting the prefix.
make prefix=/other/dir/i/want/to/be/able/to/start/from/ install should
work though.

Re: [Success] steve <at> sxemacs.org--2006/sxemacs--main--22.1.6--patch-87 on i686-pc-linux-gnu

* Sebastian Freundt <hroptatyr <at> gna.org> writes:

  > Steve Youngs <steve <at> sxemacs.org> writes:
  >> First up, `make DESTDIR=/some/pkging/dir install' works!!  Very happy
  >> about that. 

[...]

  >> C-x V initial-scratch-message RET
  >> => The library \
  >>
"/home/steve/programming/SXEmacs/core/sxemacs--main--22.1.6/=bld-FHS-automake/lisp/startup" \
  >> is not in the path.
  >> 
  >> But C-x 4 l (find-lib-other-win) startup RET works.  It finds the
  >> installed startup.el in /usr/local/share/sxemacs-22.1.6/lisp/

  > Indeed. Afaik, it still uses ${prefix} internally and make DESTDIR= just
  > redirects to some other dir _without_ adapting the prefix.

Exactly how it should be.

  > make prefix=/other/dir/i/want/to/be/able/to/start/from/ install should
  > work though.

It doesn't matter if that works because you don't run from
$DESTDIR/$prefix/bin, you create a package (distro packages like .deb,
.rpm etc) in $DESTDIR, then install that package into $prefix.

  > However, the glyph/toolbar thingie is strange indeed (and reproducible

sxemacs_version.m4

Nelson mentioned in sxemacs-builds...

  > I contend that we should _always_ generate the sxemacs_version.m4 on
  > autogen.sh, because this is _definitely not_ the most expensive
  > operation there, and you will probably only notice the failure further
  > down the road...

Yes, I agree.  Good idea.  I'll fix it.  Unfortunately
arch.sxemacs.org seems to be down, so it'll have to wait till it comes
back up.

--

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|       SXEmacs - The only _______ you'll ever need.       |
|         Fill in the blank, yes, it's THAT good!          |
|------------------------------------<steve <at> sxemacs.org>---|

22.1.6--version-0 is just around the corner

Hi Folks!

Righto, as of now, we are in feature freeze.  That means that I will
only merge bug fixes and doc updates.  And no, you can't sneak in
anything new disguised as a bug fix.  Sebastian, I'm looking at
_you_. 

Oki, what needs to be done before v-0: [1]

  o The open bugs at issues.sxemacs.org have to be dealt with.  There
    is currently 2... #13 and #19.  Go check em out, if you think you
    can help, go for it!  I'm certain the assignees of those bugs
    won't mind if you fix them. 

  o The test suite is currently failing in the weak-tests.  This is a
    known issue and a fix is available.  I'll take care of this.

  o INSTALL needs a little update to bring it inline with the current
    build chain.  There are probably lots of other doc updates needed
    because of the automake migration, but INSTALL is all I'm
    concerned about for this release.  Of course, if you wanna go
    ahead and make those other doc updates, who am I to stop you? 

ETTR (Estimated Time To Release):

I'm thinking, about 2 - 3 weeks.  Definitely before December.

Sometime tomorrow I'll put a snapshot tarball on ftp.sxemacs.org for
the non-tla-enabled folks.

Snapshot sxemacs-22.1.6-20061107.tar.bz2 available

Hi Folks!

I've just uploaded a snapshot tarball to ftp.sxemacs.org.  You'll find
`sxemacs-22.1.6-20061107.tar.bz2' in...

  <ftp://ftp.sxemacs.org/pub/sxemacs/snapshot/≥

EFS compat link...

  /ftp.sxemacs.org:/pub/sxemacs/snapshot/

SPECIAL NOTE:
============
The FTP server that hosts us seems to be having a little difficulty,
you will have to turn passive mode _off_.

Snapshot Revision:
-----------------
This snapshot corresponds to...

  steve <at> sxemacs.org--2006/sxemacs--main--22.1.6--patch-95

Building Snapshots:
------------------
Goats, virgins, and buckets of lard are generally not required, but
you never know...

Firstly, you _DON'T_ need to do anything with autogen.sh.  Just do the
very familiar...

Re: [Success] steve <at> sxemacs.org--2006/sxemacs--main--22.1.6--patch-101 on i686-pc-linux-gnu

* Nelson Ferreira <njsf <at> sxemacs.org> writes:

  > "SY" == Steve Youngs <steve <at> sxemacs.org> writes:
  SY>  Sebastian, this is inplace with your p-95
  SY>  erm, this new "* means requested but not enabled" stuff I find _very_
  SY>  confusing.  IMO the summary is better without it.

  > I say put this only once at the start or end of the Summary...

  > Like after Compilation Environment and Installation Defaults, or just
  > before 'No bogus options' for a footnote like style.

I think Sebastian was trying to make things clearer for those mutually
exclusive things like GMP <-> MP.  And I'm sure we had this sorted out
pre-automake migration.  We just combined them like...

  (G)MP disabled

--

-- 
|---<Steve Youngs>---------------<GnuPG KeyID: A94B3003>---|
|       SXEmacs - The only _______ you'll ever need.       |
|         Fill in the blank, yes, it's THAT good!          |
|------------------------------------<steve <at> sxemacs.org>---|

[Apple Developer Bug Reporting] Re: Bug ID 4832488: C library does not handle properly long double values

For your information :)

From: Apple Developer Bug Reporting <devbugs <at> apple.com>
Subject: Re: Bug ID 4832488: C library does not handle properly long double values
Date: 2006-11-16 19:38:54 GMT

Hello Nelson,

This is a follow-up to Bug ID# 4832488.  We have received the following update regarding your report:

<GMT15-Nov-2006 22:42:03GMT> Vanaja Pasumarthi:
Engineering has determined that this issue behaves as intended based on the following information:

http://developer.apple.com/documentation/developertools/Conceptual/LowLevelABI/Articles/32bitPowerPC.html

"(*) In Mac OS X v10.4 and later and GCC 4.0 and later, the size of the long double extended precision data type
is 16 bytes (it’s made up of two 8-byte doubles). In earlier versions of Mac OS X and GCC, long double is
equivalent to double. You should not use the long double type when you use GCC 4.0 or later to develop or in
programs targeted at Mac OS X versions earlier than 10.4."

PowerPC uses head-tail long double format, which is really two consecutive (regular) doubles whose sum it
the full value.  The first double is equal to the long double value rounded to a double, while the second

SXEmacs to debian

Hi

Where I can find deb-packages for debian?

--

-- 

 .''`.   Kirill A. Kórinskiy <catap <at> catap.ru>
: :'  :  proud maniac programmer
`. `'`   http://catap.ru/ - +7 (916) 3-604-704 - xmpp:catap <at> catap.ru
  `-  Debian - when you have better things to do than fixing systems (c)madduck