Another NON trivial Cyrillic question

[Alexander Shopov]

Hi guys,
Another nontrivial question for cyrillic in Cygwin:
How do I make the command `rm' show cyrillic when doing
rm -v file_with_cyrillic_letters_in_filename
for example
rm -v кирил
.inputrc contains:
set meta-flag on
set convert-meta off
set output-meta on
Shell is bash

terminal is win32 rxvt (no X Window)

BTW - if you want your CP1251 windows to show cyrillic letters with 
cygwin less

export LESSCHARDEF=8bcccbcc18b119.b104.

in ~/.bash_profile

(I include this to be available in search engines).

Best regards:
al_shopov

Fwd: Building mysql 4.0.17 under latest Cygwin (1.5.7)

[Gerrit P. Haase]

Dies ist eine weitergeleitete Nachricht - This is a forwarded message
Von: Mikael Åsberg <[EMAIL PROTECTED]>
An : [EMAIL PROTECTED]
Datum  : Dienstag, 10. Februar 2004, 20:37
Betreff: Building mysql 4.0.17 under latest Cygwin (1.5.7)

===8<=== Original Nachrichtentext ===
Hello Mr Haase! I am trying to build mysql 4.0.17 under the latest
offical Cygwin (1.5.7). I am following your instructions found at
http://anfaenger.de/cygwin-1.5/mysql/.
However, I am getting a compilation error during make (an error that
occurs after it has been compiling for some time). I made a script
that looks like this:
#!/bin/bash
patch -p1 <../mysql-4.0.17.diff
export CFLAGS="-O3"
export CC=gcc
export CXX=gcc
export CXXFLAGS="-O3 -mcpu=pentiumpro -fno-omit-frame-pointer -felide-constructors 
-fno-exceptions -fno-rtti"
./configure \
--prefix=/usr \
--sysconfdir=/etc \
--localstatedir=/var/mysql/data \
--libexecdir=/usr/sbin \
--enable-thread-safe-client \
--with-unix-socket-path=/var/mysql/mysql.sock \
--with-mysqld-user=mysqld \
--with-berkeley-db \
--with-innodb \
--without-debug \
--without-openssl \
--with-mysqld-ldflags=-all-static \
--with-vio \
--with-embedded-server

As you can, all it does is applying the patch found at your site, exporting some 
variables and then running ./configure
with the options you suggested. The script completes without any errors.
Then I run $ make
I get:
mysqld.cc: In function `void* signal_hand(void*)':
mysqld.cc:1779: error: `sigwait' undeclared (first use this function)
mysqld.cc:1779: error: (Each undeclared identifier is reported only once for
   each function it appears in.)
make[4]: *** [mysqld.o] Error 1
make[4]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
make[3]: *** [all-recursive] Error 1
make[3]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
make[2]: *** [all] Error 2
make[2]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
make[1]: *** [all-recursive] Error 1
make[1]: Leaving directory `/home/Administrator/mysql-4.0.17'
make: *** [all] Error 2

Now I was wondering if you might have an idea what I did do wrong?
I haven't been using cygwin or the gnu tools very long so this is
all very new to me.

If you have any ideas, I would love to hear them!

Best regards

 

===8<== Ende des Original Nachrichtentextes =



Ciao,

Gerrit
-- 
=^..^=


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Building mysql 4.0.17 under latest Cygwin (1.5.7)

[Gerrit P. Haase]

Hallo Mikael,

Am Mittwoch, 11. Februar 2004 um 10:52 schriebst du:

> Hello Mr Haase! I am trying to build mysql 4.0.17 under the latest
> offical Cygwin (1.5.7). I am following your instructions found at
> http://anfaenger.de/cygwin-1.5/mysql/.
> However, I am getting a compilation error during make (an error that
> occurs after it has been compiling for some time). I made a script
> that looks like this:
> #!/bin/bash
> patch -p1 <../mysql-4.0.17.diff
> export CFLAGS="-O3"
> export CC=gcc
> export CXX=gcc
> export CXXFLAGS="-O3 -mcpu=pentiumpro -fno-omit-frame-pointer -felide-constructors 
> -fno-exceptions -fno-rtti"
> ./configure \
> --prefix=/usr \
> --sysconfdir=/etc \
> --localstatedir=/var/mysql/data \
> --libexecdir=/usr/sbin \
> --enable-thread-safe-client \
> --with-unix-socket-path=/var/mysql/mysql.sock \
> --with-mysqld-user=mysqld \
> --with-berkeley-db \
> --with-innodb \
> --without-debug \
> --without-openssl \
> --with-mysqld-ldflags=-all-static \
> --with-vio \
> --with-embedded-server

> As you can, all it does is applying the patch found at your site,
> exporting some variables and then running ./configure
> with the options you suggested. The script completes without any errors.
> Then I run $ make
> I get:
> mysqld.cc: In function `void* signal_hand(void*)':
> mysqld.cc:1779: error: `sigwait' undeclared (first use this function)
> mysqld.cc:1779: error: (Each undeclared identifier is reported only once for
>each function it appears in.)
> make[4]: *** [mysqld.o] Error 1
> make[4]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
> make[3]: *** [all-recursive] Error 1
> make[3]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
> make[2]: *** [all] Error 2
> make[2]: Leaving directory `/home/Administrator/mysql-4.0.17/sql'
> make[1]: *** [all-recursive] Error 1
> make[1]: Leaving directory `/home/Administrator/mysql-4.0.17'
> make: *** [all] Error 2

> Now I was wondering if you might have an idea what I did do wrong?
> I haven't been using cygwin or the gnu tools very long so this is
> all very new to me.

> If you have any ideas, I would love to hear them!

> Best regards

The command to compile mysqld.cc in my build was:

make[4]: Entering directory `/ftproot/pub/mysql/mysql-4.0.17/sql'
if gcc -DMYSQL_SERVER -DDEFAULT_MYSQL_HOME="\"/usr\"" -DDATADIR="\"/var/mysql/data\"" 
-DSHAREDIR="\"/usr/share/mysql\"" -DHAVE_CONFIG_H -I. -I. -I.. -I../bdb/build_unix 
-I../innobase/include -I./../include -I./../regex -I. -I../include -I. -O3 
-DDBUG_OFF -O3 -mcpu=pentiumpro -fno-omit-frame-pointer -felide-constructors 
-fno-exceptions -fno-rtti  -fno-implicit-templates -fno-exceptions -fno-rtti 
-DUSE_MYSYS_NEW -DDEFINE_CXA_PURE_VIRTUAL -MT mysqld.o -MD -MP -MF ".deps/mysqld.Tpo" \
  -c -o mysqld.o `test -f 'mysqld.cc' || echo './'`mysqld.cc; \
then mv -f ".deps/mysqld.Tpo" ".deps/mysqld.Po"; \
else rm -f ".deps/mysqld.Tpo"; exit 1; \
fi

This finishes without error under cygwin-1.5.5.

In my my_config.h is defined:
/* sigwait with one argument */
/* #undef HAVE_NONPOSIX_SIGWAIT */
[...]
/* POSIX sigwait */
/* #undef HAVE_SIGWAIT */

then this define in my_pthread.h is used:
#ifndef HAVE_NONPOSIX_SIGWAIT
#define my_sigwait(A,B) sigwait((A),(B))
#else
[...]

where sigwait is defined in .

Maybe there was something changed in later Cygwin releases?

Gerrit
-- 
=^..^=


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Another NON trivial Cyrillic question

[Alexander Shopov]

Just to clarify things a bit further:

My problem is that rm -v cyr_file_name
produces output:
removing `\344\356\344\356\blah-blah-bla'

instead of

removing `cyr_file_name'

I can ls and rm files with cyrillic names just fine.

Best regards:
al_shopov
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: definition of 'timezone'

[Corinna Vinschen]

On Feb 10 14:32, King Lung Chiu wrote:
> Is there a standard way to convert cygwin's 'timezone' to the number 
> expected by the Linux code?

Cygwin uses the timezone definition of BSD for a long time.
You get the posix timezone by one of two ways:

- Either use _timezone
- or #define timzonevar before including time.h.

Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Developermailto:[EMAIL PROTECTED]
Red Hat, Inc.

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Problem rolling back cygwin from 1.5.7.1 to 1.5.5.1

[Yaakov Selkowitz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1
Charles Plager wrote:
|
| I tried to do just that.  Part of the problem is that if you search for
| version numbers (say 1.5.7), the search engine isn't set up for that:
|
| Search results
| No matches were found for '1 and 5 and 7'
The mail archive search engine isn't that good, as you found out.
Usually we recommend googling for "site:cygwin.com 1.5.7 _fcntl64" (for
example).
| Rolling back 'ash' seems to have done the trick.  Thanks Yaakov.

Glad to hear everything's working again.  I do hope that you get your
problems with 1.5.7 figured out at some point so that you're able to use
it again, because as more programs are (re)built with it you may not be
able to use them with 1.5.5.
Yaakov
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org
iD8DBQFAKjRJpiWmPGlmQSMRAtUbAKClw3UY9J/5ZDkVmMmswh8fRjrCVgCghqE0
6HwtKMS9x1GDcQ0+2eB/AnM=
=I3Pu
-END PGP SIGNATURE-
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: XEmacs-21.4.14-2 consistently crashes

[Demmer, Thomas]

>
>> "Thomas" == Thomas Demmer writes:
>
>Thomas> Hi all,
>
>Hi Thomas, finally you send the message to the list :-)
>
>Thomas> I can reliably crash XEmacs by the following procedure:
>
>FWIW, me too. But since I will *never* use the Win32 version and have
>never used the Win32 version as already stated, somebody who is
>interested has to debug this or try sending a bug report to xemacs-beta.
>
>

Until last week I was using 21.4.11 or so from xemacs.org, built with
whatever cygwin they used and it ran fine, so I was assuming a cygwin bug
here. 
Do you know since when the FIXME comment below exists?

When I said the X version works fine, I actually lied. On exit it dies with
a fatal error(11) and the lisp backtrace

Lisp backtrace follows:

  kill-emacs()
  # bind (arg)
  save-buffers-kill-emacs(nil)
  # bind (command-debug-status)
  call-interactively(save-buffers-kill-emacs)
  # (condition-case ... . error)
  # (catch top-level ...)


But that is another story. 
The code fragment you mentioned only shows where the ill condition
is trapped, not where and why it occurs. Looks like I really have to
try to build a debug version myself, but that is pretty daunting :-)


Ciao
Tom

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

why can lstat fail? (was: Re: 1.5.7: nfs-server 2.2.47-2 troubles on XP)

[Baurjan Ismagulov]

Hello,

On Wed, Feb 04, 2004 at 04:41:28PM +0100, Baurjan Ismagulov wrote:
> I've installed cygwin snapshot 20030203 and run nfs-server-config. After
> starting the services, I cannot mount a directory with map_static:
> 
> linuxnfsclient:~# mount -o nolock 192.168.0.10:/etc /mnt
> mount: wrong fs type, bad option, bad superblock on 192.168.0.10:/etc,
>or too many mounted filesystems

1. I haven't received any answer within a week. I would be very grateful
   if anyone having nfs mounts with proper owners and permissions shared
   his setup. Sam?

2. Ethereal shows that mountd authorizes the MOUNT call. After that the
   client calls GETATTR for /etc, which fails with ENOENT. Running
   rpc.nfsd as SYSTEM with -d all -F options reveals that lstat("/etc",
   ...) fails with ENOENT (line 106 of nfsd-SYSTEM.log; confirmed under
   gdb).

   The question: What makes lstat fail?

   Playing with owners and permissions hasn't given any results
   (perm-original.txt and perm-modified.txt).

   I've also seen that nfsd changes the effective user and group ids.
   Trying to reproduce the problem, I've written a.c, which works
   without problems with both original and modified permissions (2113
   corresponds to root. FWIW, it is disabled, which doesn't seem to
   matter). What else in nfsd could make lstat fail?

   I've also tried to run nfsd as a normal user. If I have "Create a
   token object" privilege (I think this was meant with "impersonate
   logged on user" right, mentioned in nfs-server-2.2.47-2.README), the
   daemon does the same, i.e., fails on lstat. If, however, I don't have
   this privilege, the daemon fails to seteuid and allows the client to
   access the filesystem; naturally, one can't write anything due to
   permissions.

Any help would be greatly appreciated. Thank you in advance.

With kind regards,
Baurjan.
nfsd[1408] 02/12/104 14:35 auth.c 383 : auth_forward_lookup(h1) h1
nfsd[1408] 02/12/104 14:35 auth.c 383 : auth_forward_lookup(h1) h1
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 0 <-> loc 2113
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 0)
nfsd[1408] 02/12/104 14:35 ugid_map.c 366 : alloc ptr map 0xa045888 @ level 0 (id 
0-16777215)
nfsd[1408] 02/12/104 14:35 ugid_map.c 366 : alloc ptr map 0xa045c90 @ level 1 (id 
0-65535)
nfsd[1408] 02/12/104 14:35 ugid_map.c 381 : alloc id  map 0xa046098 @ level 2 (id 
0-255)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2113)
nfsd[1408] 02/12/104 14:35 ugid_map.c 366 : alloc ptr map 0xa0468a0 @ level 0 (id 
0-16777215)
nfsd[1408] 02/12/104 14:35 ugid_map.c 366 : alloc ptr map 0xa046ca8 @ level 1 (id 
0-65535)
nfsd[1408] 02/12/104 14:35 ugid_map.c 381 : alloc id  map 0xa0470b0 @ level 2 (id 
2048-2303)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 1 <-> loc 2114
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 1)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2114)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 2 <-> loc 2115
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 2)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2115)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 4 <-> loc 2116
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 4)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2116)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 8 <-> loc 2117
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 8)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2117)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 12 <-> loc 2118
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 12)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044c70, 2118)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa046ca8
nfsd[1408] 02/12/104 14:35 ugid_map.c 328 : h1:/etc map uid rem 30 <-> loc 2119
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ugid_get_entry(0xa044868, 30)
nfsd[1408] 02/12/104 14:35 ugid_map.c 397 : result = 0xa045c90
nfsd[1408] 02/12/104 14:35 ugid_map.c 356 : ug

Re: Another NON trivial Cyrillic question

[Igor Pechtchanski]

On Wed, 11 Feb 2004, Alexander Shopov wrote:

> Just to clarify things a bit further:
>
> My problem is that rm -v cyr_file_name
> produces output:
>
> removing `\344\356\344\356\blah-blah-bla'
>
> instead of
>
> removing `cyr_file_name'
>
> I can ls and rm files with cyrillic names just fine.
>
> Best regards:
> al_shopov

Al,

Take a look at fileutils sources, especially quotearg_buffer_restyled()
in lib/quotearg.c (which is what eventually gets called by "rm -v").
After a cursory glance, it looks like it is locale-aware, so if you have
your locale settings straight, cyrillic should print out just fine.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Another NON trivial Cyrillic question

[Roman Belenov]

Igor Pechtchanski <[EMAIL PROTECTED]> writes:

> After a cursory glance, it looks like it is locale-aware, so if you
> have your locale settings straight, cyrillic should print out just
> fine.

What does it mean to "have your locale settings straight" in cygwin ?
Is it possible to make cygwin respect Windows locale settings (it
seems that they don't affect cygwin out of the box) ? Or may be
one should install Unix-like locale database and use variables like
LANG to choose locale ?

-- 
With regards, Roman.


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Another NON trivial Cyrillic question

[Igor Pechtchanski]

On Wed, 11 Feb 2004, Roman Belenov wrote:

> Igor Pechtchanski <[EMAIL PROTECTED]> writes:
>
> > After a cursory glance, it looks like it is locale-aware, so if you
> > have your locale settings straight, cyrillic should print out just
> > fine.
>
> What does it mean to "have your locale settings straight" in cygwin ?
> Is it possible to make cygwin respect Windows locale settings (it
> seems that they don't affect cygwin out of the box) ? Or may be
> one should install Unix-like locale database and use variables like
> LANG to choose locale ?

The latter.
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

1.5.7: core sump: ls /proc/registry

[Chuck]

Forgive me if this gets posted twice. I posted to the gmane NG first with 
a munged address and then realized that gmane may not like that so I'm 
reposting with my real address.

I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
dump. 

$ ls /proc/registry
Segmentation fault (core dumped)

Listing a directory below /proc/registry seems to be fine.

$ ls /proc/registry/hkey_local_machine
./  ../  HARDWARE/  SAM/  SECURITY/  SOFTWARE/  SYSTEM/

Is this a known problem? I didn't see it in the FAQ or by searching the 
mailing list archives. I'm thinking it may have been introduced in 1.5.7.

MY O/S is Win2k 5.00.2195 Service Pack 4.
-- 
Chuck


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Another NON trivial Cyrillic question

[Alexander Shopov]

Well, this is not entirely true.
While programs do lookup the LANG envvar, they do not recode properly.
For example when I do:
export LANG=bg_BG.CP1251
or
export LANG=bg_BG.UTF-8
wget uses the bulgarian po-file translation, but it is not recoded 
properly and is just a Latin-1 dump of an UTF-8 file (the wget.mo is 
compiled from an properly encoded UTF-8 po file)
In both cases:
I get:

$ touch кирил
$ rm -v кирил
removing `\352\350\360\350\353'
I have installed
libgettextpo0 - 0.12.1-3
libiconv2 - 1.9.1-3
libintl - 0.10.38.-3
libintl1 - 0.10.40-1
libintl2 - 0.12.1-3
locale database is part of libc. Where do I look for it in Cygwin?
I have the sources for the Bulgarian locale. I could try to compile it.
While under GNU/Linux I prefer UTF-8 locales, I have to use a CP1251 one 
under windows.

Best regards:
al_shopov



What does it mean to "have your locale settings straight" in cygwin ?
Is it possible to make cygwin respect Windows locale settings (it
seems that they don't affect cygwin out of the box) ? Or may be
one should install Unix-like locale database and use variables like
LANG to choose locale ?


The latter.
Igor

Re: 1.5.7: core sump: ls /proc/registry

[Christopher Faylor]

On Wed, Feb 11, 2004 at 03:42:24PM +, Chuck wrote:
>I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
>dump. 
>
>$ ls /proc/registry
>Segmentation fault (core dumped)

"Works for me"

http://cygwin.com/problems.html will help you report a problem such that
we can track it down.
--
Please use the resources at cygwin.com rather than sending personal email.
Special for spam email harvesters: send email to [EMAIL PROTECTED]
and be permanently blocked from mailing lists at sources.redhat.com
[EMAIL PROTECTED]

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Another NON trivial Cyrillic question

[Igor Pechtchanski]

Umm, yes, character set/encoding support in Cygwin is minimal.  If you
look at the fileutils sources, though, there is some code that does
presumably locale-specific quoting.  I basically glanced at it, but if you
wanted to improve the behavior of Cygwin tools, that's the place I would
start from.
Igor
P.S. I believe there are precompiled locales as part of the gettext
package.

On Wed, 11 Feb 2004, Alexander Shopov wrote:

> Well, this is not entirely true.
> While programs do lookup the LANG envvar, they do not recode properly.
> For example when I do:
> export LANG=bg_BG.CP1251
> or
> export LANG=bg_BG.UTF-8
>
> wget uses the bulgarian po-file translation, but it is not recoded
> properly and is just a Latin-1 dump of an UTF-8 file (the wget.mo is
> compiled from an properly encoded UTF-8 po file)
> In both cases:
> I get:
>
> $ touch ?
> $ rm -v ?
> removing `\352\350\360\350\353'
>
> I have installed
> libgettextpo0 - 0.12.1-3
> libiconv2 - 1.9.1-3
> libintl - 0.10.38.-3
> libintl1 - 0.10.40-1
> libintl2 - 0.12.1-3
>
> locale database is part of libc. Where do I look for it in Cygwin?
> I have the sources for the Bulgarian locale. I could try to compile it.
> While under GNU/Linux I prefer UTF-8 locales, I have to use a CP1251 one
> under windows.
>
> Best regards:
> al_shopov
>
> >>What does it mean to "have your locale settings straight" in cygwin ?
> >>Is it possible to make cygwin respect Windows locale settings (it
> >>seems that they don't affect cygwin out of the box) ? Or may be
> >>one should install Unix-like locale database and use variables like
> >>LANG to choose locale ?
> >
> > The latter.
> > Igor

-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: 1.5.7: core sump: ls /proc/registry

[Pierre A. Humblet]

On Wed, Feb 11, 2004 at 11:26:59AM -0500, Christopher Faylor wrote:
> On Wed, Feb 11, 2004 at 03:42:24PM +, Chuck wrote:
> >I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
> >dump. 
> >
> >$ ls /proc/registry
> >Segmentation fault (core dumped)
> 
> "Works for me"

/bin/ls /proc/registry works for me too but
/bin/ls -F /proc/registry fails.

Pierre



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: 1.5.7: core sump: ls /proc/registry

[Corinna Vinschen]

On Feb 11 11:32, Pierre A. Humblet wrote:
> On Wed, Feb 11, 2004 at 11:26:59AM -0500, Christopher Faylor wrote:
> > On Wed, Feb 11, 2004 at 03:42:24PM +, Chuck wrote:
> > >I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
> > >dump. 
> > >
> > >$ ls /proc/registry
> > >Segmentation fault (core dumped)
> > 
> > "Works for me"
> 
> /bin/ls /proc/registry works for me too but
> /bin/ls -F /proc/registry fails.

Something's going wrong in get_nt_object_attribute.

Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Developermailto:[EMAIL PROTECTED]
Red Hat, Inc.

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

cygwin rebooting computer

[Anders Nyström]

I am a big fan of cygwin and has just learned the TAB completion thing.

Now, my harddisk is partitioned in 2 so sometimes (not always)
when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset

Am I doing anything wrong or what

rgds Anders Nyström


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: 1.5.7: core sump: ls /proc/registry

[Pierre A. Humblet]

On Wed, Feb 11, 2004 at 06:24:11PM +0100, Corinna Vinschen wrote:
> On Feb 11 11:32, Pierre A. Humblet wrote:
> > On Wed, Feb 11, 2004 at 11:26:59AM -0500, Christopher Faylor wrote:
> > > On Wed, Feb 11, 2004 at 03:42:24PM +, Chuck wrote:
> > > >I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
> > > >dump. 
> > > >
> > > >$ ls /proc/registry
> > > >Segmentation fault (core dumped)
> > > 
> > > "Works for me"
> > 
> > /bin/ls /proc/registry works for me too but
> > /bin/ls -F /proc/registry fails.
> 
> Something's going wrong in get_nt_object_attribute.

Right.
Problem happens with non-existant
/proc/registry/HKEY_PERFOMANCE_DATA

It's surely useless to call sd.malloc (len)
when the first call to RegGetKeySecurity has failed.

Pierre



--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: 1.5.7: core sump: ls /proc/registry

[Corinna Vinschen]

On Feb 11 18:24, Corinna Vinschen wrote:
> On Feb 11 11:32, Pierre A. Humblet wrote:
> > On Wed, Feb 11, 2004 at 11:26:59AM -0500, Christopher Faylor wrote:
> > > On Wed, Feb 11, 2004 at 03:42:24PM +, Chuck wrote:
> > > >I just upgraded to 1.5.7. When I do "ls /proc/registry" I get a core 
> > > >dump. 
> > > >
> > > >$ ls /proc/registry
> > > >Segmentation fault (core dumped)
> > > 
> > > "Works for me"
> > 
> > /bin/ls /proc/registry works for me too but
> > /bin/ls -F /proc/registry fails.
> 
> Something's going wrong in get_nt_object_attribute.

Should be fixed in CVS now.

Corinna

-- 
Corinna Vinschen  Please, send mails regarding Cygwin to
Cygwin Developermailto:[EMAIL PROTECTED]
Red Hat, Inc.

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

How to install *normal* WordNet after installing CygWin

[Lance Ramshaw]

I downloaded WordNet-2.0.exe from Princeton to install WordNet in its normal
Windows mode on a new XP machine.  However, I had installed Cygwin first,
and since WordNet2.0 is now also available within Cygwin, the normal
installation process seems to somehow trigger Cygwin's installer, and
neither version gets successfully installed.

Is there a way to disable Cygwin's installer enough to allow WordNet for
Windows to install normally?

Thanks,
Lance

BBN Technologies


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Different executables of the same source

[Shankar Unni]

Alex Vinokur wrote:

How can one get the creation time of object file foo.o?
Use "objdump -p".

But it looks like gcc doesn't stuff a timestamp into the .o, but does 
into the .exe.

Visual C++ puts a timestamp in both the .obj and .exe.

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Segfault in _cygwin_dll_entry

[peda]

> The difference, althought it really doesn't matter, is that
> libzsh-4.1.1.dll was rebased, while cygggi-2.dll isn't.  Something in 
the
> makeup of cygggi-2.dll causes the same condition as when 
libzsh-4.1.1.dll
> is rebased.

I found a couple of __declspec(dllexport) and __declspec(dllimport) from
a previous attempt at porting GGI to cygwin/mingw. When I removed these
the segfault disappeared.

Regards,
Peter Ekberg

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin rebooting computer

[Igor Pechtchanski]

On Wed, 11 Feb 2004, Anders Nyström wrote:

> I am a big fan of cygwin and has just learned the TAB completion thing.
>
> Now, my harddisk is partitioned in 2 so sometimes (not always)
> when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset
>
> Am I doing anything wrong or what
> rgds Anders Nystrom

Anders,

Cygwin by itself (i.e., without explicitly making the appropriate Windows
calls, as "shutdown" does) should not be able to reboot your machine.  I'd
suspect a faulty driver of some sort.  Does this problem happen on "ls
/cygdrive" as well?

My guess is you should be able to reproduce this problem from a pure
Windows program, by making the same call sequence as the one in
fhandler_cygdrive::set_drives() in src/winsup/cygwin/fhandler_disk_file.cc
(even just calling GetLogicalDriveStrings() might do it).
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin rebooting computer

[Shankar Unni]

Anders Nyström wrote:

Now, my harddisk is partitioned in 2 so sometimes (not always)
when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset
Am I doing anything wrong or what
I'm sure Chris and others are getting tired of saying this by now, so:

Follow the guidelines in http://cygwin.com/problems.html for reporting 
these problems. That'll be a good start. Please follow the directions 
*exactly*, especially the one about attaching your cygcheck output as an 
*attachment*.

No, cygwin shouldn't cause the OS to crash, so there's something else 
going on here..

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: cygwin rebooting computer

[Chris January]

> I am a big fan of cygwin and has just learned the TAB completion thing.
>
> Now, my harddisk is partitioned in 2 so sometimes (not always)
> when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset
>
> Am I doing anything wrong or what

Try looking in the System event log ('Control Panel'->'Administrative
Tools'->'Event Viewer' then click System) to see if that sheds any light on
the crashes.

Assuming you are using Windows XP:

Try opening 'Control Panel'->'System'.
Then 'Advanced' tab->'Settings'.
Uncheck 'Automatically restart'.

Next time you should see a blue screen instead of the computer rebooting.
See if this helps you solve the problem.

Chris


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

gcc 2

[Richard Greenwood]

I would like to compile some software (Grass) with gcc 2.x, ideally 
2.95. Is there a binary cygwin distribution of gcc2 available, and if 
so, where would I look for it?

Thanks,
--
Richard Greenwood
www.greenwoodmap.com
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

[ANNOUNCEMENT] Updated: naim-0.11.6.6-1

[Daniel Reed]

naim 0.11.6.6 is now available through the Cygwin Net Release.

naim is a console client for AOL Instant Messenger, ICQ, Internet Relay
Chat, and The lily CMC. Development work is done on Linux, but naim should
work on any system with an ANSI C compiler, BSD sockets, and ncurses. This
includes operating systems such as BeOS, FreeBSD, Mac OS 10, NetBSD, and
Microsoft Windows using Cygwin. naim has been in development since October,
1998. More information is available from:
http://naim.n.ml.org/

,--.
| To install or upgrade this package, run the Cygwin Net Release Setup |
| Program, available from [http://cygwin.com/setup.exe]. Answer all of the |
| questions to reach the Select Packages screen. You will find naim listed |
| in the "Net" category.   |
`--'


This is a mix of new features and bug fixes. All users are encouraged to
upgrade as soon as possible.


If there is any questionable behaviour at all, please contact me directly at
<[EMAIL PROTECTED]>, on AIM as "naim help," or on EFnet or Undernet IRC in #naim. If
you are interested in the use and development of naim, feel free to
subscribe to the naim-users mailing list. Just send a blank email to
<[EMAIL PROTECTED]> and reply to the confirmation message.

-- 
Daniel Reed <[EMAIL PROTECTED]> http://naim-users.org/nmlorg/   http://naim.n.ml.org/
Never be afraid to try something new. Remember: Amateurs built the
ark. Professionals built the Titanic.


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: Different executables of the same source

[Brian Ford]

On Wed, 11 Feb 2004, Shankar Unni wrote:

> Alex Vinokur wrote:
>
> > How can one get the creation time of object file foo.o?
> >
> Use "objdump -p".
>
> But it looks like gcc doesn't stuff a timestamp into the .o, but does
> into the .exe.
>
Just a simple correction.  gcc doesn't create .o or .exe files.  as/ld
do respectively :).

So, this would be an as "buglet", I guess.  Feel free to follow up in the
binutils mailing list if you feel the urge.

-- 
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
Phone: 314-551-8460
Fax:   314-551-8444

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: XEmacs-21.4.14-2 consistently crashes

[Dr. Volker Zell]

> "Thomas" == Thomas Demmer writes:

Thomas> Until last week I was using 21.4.11 or so from xemacs.org, built with
Thomas> whatever cygwin they used and it ran fine, so I was assuming a cygwin bug
Thomas> here.

In which mode does it run ?

The following is copied from a header file in the XEamcs source tree,
windowsnt.h:

1. Keep in mind that there are two possible OS environments we are dealing
   with -- Cygwin and Native Windows.  Cygwin provides a POSIX emulation
   layer on top of MS Windows -- in particular, providing the file-system,
   process, tty, and signal semantics that are part of a modern, standard
   Unix operating system.  MS Windows also provides these services, but
   through their own API, called Win32.  When compiling in a Cygwin
   environment, the Win32 API's are also available, and in fact are used
   to do native GUI programming.

2. There are two windowing environments we can target XEmacs for when
   running under MS Windows -- Windows native, and X. (It may seem strange
   to write an X application under Windows, but there are in fact many X
   servers out there running on Windows, and as far as I know there is no
   real (or at least, that works well) networking Window-system extension
   under MS Windows.  Furthermore, if you're porting a Unix application to
   Windows and use Cygwin to assist you, it might seem natural to use an
   X server to avoid having to port all the code to Windows.) For XEmacs,
   there are various reasons people could come up with for why we would
   want to keep maintaining X Windows under MS Windows support.

That gives us four possible build environments.  I (Ben) build
regularly on fully-native-everything, Andy builds on Cygwin + MS
Windows + X Windows for windowing.

The build flags used for these divisions are:

CYGWIN -- for Cygwin-only stuff.
WIN32_NATIVE -- Win32 native OS-level stuff (files, process, etc.).
HAVE_X_WINDOWS -- for X Windows (regardless of whether under MS Win)
HAVE_MS_WINDOWS -- MS Windows native windowing system (anything related to
   the appearance of the graphical screen).

Finally, there's also the MINGW build environment, which uses GCC
\(similar to Cygwin), but native MS Windows libraries rather than a
POSIX emulation layer (the Cygwin approach).  This environment defines
WIN32_NATIVE, but also defines MINGW, which is used mostly because
uses its own include files (related to Cygwin), which have a few
things messed up.


Thomas> Do you know since when the FIXME comment below exists?

Sorry, no.

Thomas> When I said the X version works fine, I actually lied. On exit it dies with
Thomas> a fatal error(11) and the lisp backtrace

This has been reported in the review of the XEmacs package by Charles
Wilson, too. He stated that it's happening for him when not running
under Administrator account, but as a normal user.

I can confirm that it's not happening under Admin account, that's why I
never recognized it. By the way, the latest 21.5.16 XEmacs also crashes
under Admin account on exit, but I usually don't care cause I start
Xemacs once and then exit only when I shutdown my system.

Thomas> The code fragment you mentioned only shows where the ill condition
Thomas> is trapped, not where and why it occurs. Looks like I really have to
Thomas> try to build a debug version myself, but that is pretty daunting :-)

If you find anything please report back or to the XEmacs-beta mailinglist.

Thomas> Ciao
Thomas> Tom

Ciao
  Volker


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: gcc 2

[Brian Ford]

On Wed, 11 Feb 2004, Richard Greenwood wrote:

> I would like to compile some software (Grass) with gcc 2.x, ideally
> 2.95. Is there a binary cygwin distribution of gcc2 available, and if
> so, where would I look for it?
>
Please try to search the archives before posting.

http://cygwin.com/ml/cygwin-announce/2003-10/msg00014.html
http://cygwin.com/ml/cygwin/2003-11/msg00472.html

etc., etc...

-- 
Brian Ford
Senior Realtime Software Engineer
VITAL - Visual Simulation Systems
FlightSafety International
Phone: 314-551-8460
Fax:   314-551-8444

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

[ANNOUNCEMENT] Updated: gtypist-2.7-2

[Yaakov Selkowitz]

-BEGIN PGP SIGNED MESSAGE-
Hash: SHA1

I have re-released GNU typist 2.7 with a few changes.

* gtypist is now dynamically linked with libiconv2 and libintl2, greatly
reducing the size of the binary.

* the scripts from the source tools directory (for creating new lessons)
are now available in /usr/share/gtypist/tools.  Some of these require
gawk and perl, so please make sure you have those installed before using
them!

Happy typing!

Yaakov


To update your installation, click on the "Install Cygwin now" link on
the http://cygwin.com/ web page.  This downloads setup.exe to your
system.  Then, run setup and answer all of the questions.
GNU typist can be found in the Text category.

If you have questions or comments, please send them to the Cygwin
mailing list at: cygwin at cygwin dot com. I would appreciate it if you
would use this mailing list rather than emailing me directly.

~  *** CYGWIN-ANNOUNCE UNSUBSCRIBE INFO ***

If you want to unsubscribe from the cygwin-announce mailing list, look
at the "List-Unsubscribe: " tag in the email header of this message.
Send email to the address specified there.  It will be in the format:

[EMAIL PROTECTED]

If you need more information on unsubscribing, start reading here:

http://sources.redhat.com/lists.html#unsubscribe-simple

Please read *all* of the information on unsubscribing that is available
starting at this URL.
-BEGIN PGP SIGNATURE-
Version: GnuPG v1.2.3 (MingW32)
Comment: Using GnuPG with Thunderbird - http://enigmail.mozdev.org

iD8DBQFAKnwTpiWmPGlmQSMRAuMBAKDT2wabJb+EcuTzMtEik3I3gRK/JwCg+k3Q
zn8+vtXK8fYzbFOg5cRLwVc=
=gvCD
-END PGP SIGNATURE-


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: How to install *normal* WordNet after installing CygWin

[Larry Hall]

At 01:03 PM 2/11/2004, Lance Ramshaw you wrote:
>I downloaded WordNet-2.0.exe from Princeton to install WordNet in its normal
>Windows mode on a new XP machine.  However, I had installed Cygwin first,
>and since WordNet2.0 is now also available within Cygwin, the normal
>installation process seems to somehow trigger Cygwin's installer, and
>neither version gets successfully installed.
>
>Is there a way to disable Cygwin's installer enough to allow WordNet for
>Windows to install normally?


Sounds like a question for the folks at Princeton.  Our recommendation to 
all third parties integrating Cygwin is to install their application and 
then point to or invoke setup.exe from the Cygwin site for the Cygwin piece.
That will keep existing Cygwin installations from breaking as well as 
subsequent.  It's quite possible that you now have two cygwin1.dlls on your
system which is a recipe for disaster.  There's also a hundred other things
that could be wrong but you'll need the Princeton folks to help you track
down what those could be.


--
Larry Hall  http://www.rfk.com
RFK Partners, Inc.  (508) 893-9779 - RFK Office
838 Washington Street   (508) 893-9889 - FAX
Holliston, MA 01746 


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: cygwin rebooting computer

[Larry Hall]

At 01:25 PM 2/11/2004, Shankar Unni you wrote:
>Anders Nyström wrote:
>
>>Now, my harddisk is partitioned in 2 so sometimes (not always)
>>when doing cd /cygdrive/d TAB, teh computer goes black = softboot = reset
>>Am I doing anything wrong or what
>
>I'm sure Chris and others are getting tired of saying this by now, so:
>
>Follow the guidelines in http://cygwin.com/problems.html for reporting these 
>problems. That'll be a good start. Please follow the directions *exactly*, especially 
>the one about attaching your cygcheck output as an *attachment*.
>
>No, cygwin shouldn't cause the OS to crash, so there's something else going on here..


Quite right.  And while any problem report to this list should follow the
guidelines of the link you pointed to, in this case, it's unlikely that 
additional information will help anyone here diagnose the problem.  These 
crashes are bugs in the O/S or drivers.  Much to the chagrin of some folks 
on this list, we're not a Windows O/S help desk. ;-)  So much discussion of
the issue is off-topic for this list.

Sorry.


--
Larry Hall  http://www.rfk.com
RFK Partners, Inc.  (508) 893-9779 - RFK Office
838 Washington Street   (508) 893-9889 - FAX
Holliston, MA 01746 


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Cygwin xinetd and in.talkd

[Brandon Bergren]

Hello.

I was trying to get talk working in my cygwin install using xinetd. I finally
figured out why it wasn't working.

on cygwin, the talk command tries to connect to the ntalk port, not the talk
port, as far as I can tell. (could just be my machine, tho.)

so, there should be a section in /etc/xinetd.d/talk for service ntalk.

seems to work ok for me after adding a section for ntalk.

Was this overlooked?

Thank you,
--Brandon Bergren

(Please CC me, I'm not subscribed. Thanks.)

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

xerces - need 2.3.0 src

[Steve Omand]

Hello,

I have been trying to get XML::xerces - Perl XML Parser - working with
my cygwin installation.
XML::xerces is currently at version 2.3.0_4 and requires xerces-C-2.3.0
- which is available 
in the cygwin distribution.  However, the XML::xerces needs the xerces
source files in addition to the binaries.

The current version of xerces-c in the cygwin distribution is 2.4.0.  
If I try to get xerces-c-2.3.0 source (via cygwin setup.exe installer),
what 
I in fact end up with is the source tree for xerces-c-2.4.0-1 

Can anyone help or advise?

thanks,

Steve.

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: xerces - need 2.3.0 src

[Igor Pechtchanski]

On Wed, 11 Feb 2004, Steve Omand wrote:

> Hello,
>
> I have been trying to get XML::xerces - Perl XML Parser - working with
> my cygwin installation.  XML::xerces is currently at version 2.3.0_4 and
> requires xerces-C-2.3.0 - which is available in the cygwin distribution.
> However, the XML::xerces needs the xerces source files in addition to
> the binaries.
>
> The current version of xerces-c in the cygwin distribution is 2.4.0.
> If I try to get xerces-c-2.3.0 source (via cygwin setup.exe installer),
> what I in fact end up with is the source tree for xerces-c-2.4.0-1
>
> Can anyone help or advise?
> thanks,
> Steve.

This is by design: setup will not let you install the sources for any
other version of a package if you have some version of the binaries
installed.  There are two ways to go about this: either downgrade to 2.3.0
temporarily, install the sources, and upgrade back, or just download the
sources directly from any mirror without using setup (e.g., via wget).
Igor
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

[ANNOUNCEMENT] CLISP 2.32-2 is available

[Sam Steingold]

CLISP 2.32-2 (marked `testing') is a CVS HEAD snapshot,
_not_ an official release.   beware.

User visible changes


* NO-APPLICABLE-METHOD, NO-PRIMARY-METHOD and NO-NEXT-METHOD now signal
  METHOD-CALL-ERROR or METHOD-CALL-TYPE-ERROR.

* New user variables CUSTOM:*APROPOS-MATCHER* and CUSTOM::*APROPOS-DO-MORE*
  make APROPOS more flexible in search and output.

* New function EXT:MOD-EXPT efficiently computes the composition of MOD
  and EXPT on integers.

* New function EXT:ARGV returns the command line arguments passed to runtime.

* DEFINE-METHOD-COMBINATION and friends are now implemented.
  Thanks to James Anderson <[EMAIL PROTECTED]>.


many bugs fixed, but that goes without saying...

-- 
Sam Steingold (http://www.podval.org/~sds) running w2k
  
 
((lambda (x) `(,x ',x)) '(lambda (x) `(,x ',x)))


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Problems w/make under Cygwin bash shell

[Brian Jorgage]

I have an older version of Cygwin (B20) under which "make" works fine.  I 
recently installed the latest version of Cygwin.  When I type "make" at the 
command prompt, I get the message:

bash: make: command not found

I have a Makefile in the local directory and ran a ./configure script which 
ran successfully.

So how do I configure my bash shell to recognize the make command ?  When I 
look at my environment variables, the only variable related to make is:

MAKE_MODE=unix

which I'm sure doesn't have any bearing on this.

I'm sure the answer is something simple.  It's been a while since I used 
makefiles.

Can anyone help me ?




--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: WML (perl-based program) segfaults

[Gerrit P. Haase]

Hallo Wolfgang,

Am Dienstag, 10. Februar 2004 um 16:26 schriebst du:

> Hello!

> I had used WML (offline HTML generator, Perl-based, version 2.0.9,
> ) happily under Cygwin with no problems whatsoever.
> Then along came some system-upgrade[1], and boom, eperl (a
> subcomponent of WML) just dies when trying to open a file. (That
> means, e.g. "eperl --help" produces some output, so the program
> doesn't *completely* just die) The other WML-components still work
> just fine.

> I'm no expert on debugging this kind of things, but I ./configure'd
> WML with --enable-debug and was able to get this from gdb:

> ###
> $ gdb eperl
> This GDB was configured as "i686-pc-cygwin"...
> Minimal symbols from eperl.exe...(no debugging symbols found)...
> [I think that's strange, as I *did* say --enable-debug]

Try also to define some debugging flags for the tools, hmmm, in the
eperl configure is already defined: CFLAGS='-g' && LDFLAGS='-g', maybe
it is neccessary to rebuild perl with symbols too?


> (gdb) run README
> [this just proves the point that the actual file doesn't matter]

> Starting program:
> /cygdrive/d/wosc/install/Cygwin/wml-2.0.9/wml_backend/p3_eperl/eperl.exe
> README

> Program received signal SIGSEGV, Segmentation fault.
> 0x1000ad60 in cygperl5_8_2!Perl_malloc () from /usr/bin/cygperl5_8_2.dll
> (gdb) bt
> #0  0x1000ad60 in cygperl5_8_2!Perl_malloc () from /usr/bin/cygperl5_8_2.dll
> #1  0x1007254c in cygperl5_8_2!Perl_sv_add_arena () from /usr/bin/cygperl5_8_2.dll
> #2  0x1007dc8b in cygperl5_8_2!Perl_newSVpvn () from /usr/bin/cygperl5_8_2.dll
> #3  0x10058488 in cygperl5_8_2!Perl_savesharedpv () from /usr/bin/cygperl5_8_2.dll
> #4  0x100586b7 in cygperl5_8_2!Perl_vmess () from /usr/bin/cygperl5_8_2.dll
> #5  0x10059171 in cygperl5_8_2!Perl_vcroak () from /usr/bin/cygperl5_8_2.dll
> #6  0x1005963f in cygperl5_8_2!Perl_croak_nocontext () from /usr/bin/cygperl5_8_2.dll
> [this sequence keeps repeating over and over]
> ###

> Does anybody have an idea what I might be looking at here, and what I
> could do about it?

> Thanks for your help,
> Wolfgang


> [1] from what I know, it might have been core cygwin.dll 1.5-3 to -4
> or perl-5.8.0 to -5.8.2, but that's just me guessing, because
> reverting to the old versions doesn't solve the problem.
> This just proves the saying "Never touch a running system"... :-(
> Current Versions are cygwin-1.5.7-1 and perl-5.8.2-1.

The first sentence at this website is:

"Support for Perl 5.8.0 added in WML 2.0.9"

Maybe it just won't work with perl-5.8.2 yet?

Also the configure script is kind of buggy as it doesn't find the
already installed modules:

> CHECK: Perl Language
> checking for Perl language... /bin/perl v5.008
> checking for Perl module Getopt::Long (2.16)... syntax error at -e line 1, near 
> "require Getopt::Long 2.16"
> Execution of -e aborted due to compilation errors.
> not found

$ perl -e 'use Getopt::Long; print $Getopt::Long::VERSION;'
2.34

[...]

> checking for Perl module IO::File (1.07)... syntax error at -e line 1, near "require 
> IO::File 1.07"
> Execution of -e aborted due to compilation errors.
> not found

$ perl -e 'use IO::File; print $IO::File::VERSION;'
1.10

> checking for Perl module Term::ReadKey (2.11)... syntax error at -e line 1, near 
> "require Term::ReadKey 2.11"
> Execution of -e aborted due to compilation errors.
> not found

$ perl -e 'use Term::ReadKey; print $Term::ReadKey::VERSION;'
2.21

> checking for MakeMaker's private install paths... ok

1 hour later:
I'm still trying to get it configured... it is a terrible package,
I wonder that it was possible to build it under Cygwin;)

When trying to install the modules used by WML I see this:

>   CPAN.pm: Going to build R/RB/RBS/File-PathConvert-0.9.tar.gz
> 
> 
> *** THIS MODULE (File::PathConvert) IS DEPRECATED. 
> 
> There are several known bugs, and it is not being actively
> maintained since all functionality is now available in
> modules (Cwd.pm and File::Spec) bundled in every Perl
> distribution of recent vintage.  This version is provided to
> fix a few bugs and to get the word out about the
> deprecation.
> 
> Please use Cwd::abs_path() instead of
> File::PathConvert::realpath() and File::Spec methods instead
> of all other path manipulation functions in this module.
> 
> If you use setfstype, you probably want to access a File::Spec::Foo
> module for the appropriate operating system.
> 
> There is also a version of File::Spec on CPAN for backwards
> compatability.
> 
> Thank you,
> 
> Barrie Slaymaker


Ok, now running:
./configure --with-openworld --enable-debug

and getting:
CHECK: Perl Language
checking for Perl language... /bin/perl v5.008
checking for Perl module Getopt::Long (2.16)... found
checking for Perl module Bit::Vector (5.2)... found
checking for Perl module File::PathConvert (ANY)... found
checking for Perl module Image::Size (2.6)... found
checking for Perl module IO::File (1.07)... found
checking for Perl module Term::Re

zsh-4.1.1-2 still seems broken under Cygwin-1.5.7-1 .

[Morris Siegel]

My PC is running under Windows XP Professional, and until recently with 
Cygwin-1.5.5-1 and zsh-4.1.1-1 .  I upgraded to Cygwin-1.5.6-1, installing 
everything available, in particular including zsh-4.1.1-2 .  zsh behaved in 
a buggy fashion.  I reported it; you kindly replied that similarly problems 
had been reported by others, and that Cygwin-1.5.7-1 should fix matters.  I 
upgraded to that, and the behavior is improved, but still buggy: (1) 
sometimes when I start zsh it hangs, sometimes it starts normally; (2) zsh 
command-line editing generally badly messes up the display when long command 
lines are being edited.  I seem to recall that problem (2) was a consistent 
nuisance a while back, but then some new Cygwin release (I don't remember 
which) fixed it.  (I generally run zsh under screen, by the way.) 

The only reason I upgraded from Cygwin-1.5.5 was to have the most up-to-date 
release; it behaved without problem for me.  Is there any way I can restore 
it? 

Thanks for your attention and reply. 

-- Morris M. Siegel
 [EMAIL PROTECTED]
--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: Problems w/make under Cygwin bash shell

[DePriest, Jason R.]

Since you upgraded from an ancient version of Cygwin, following the
directions in http://cygwin.com/problems.html is imperative.

Please attach what it asks for so the group can help you out.

It could be as simple as a package that did not get updated...

> -Original Message-
> From: cygwin-owner at cygwin dot com 
> [mail to:cygwin-owner at cygwin dot com] On Behalf Of Brian Jorgage
> Sent: Wednesday, February 11, 2004 1:21 PM
> To: cygwin at cygwin dot com
> Subject: Problems w/make under Cygwin bash shell
> 
> 
> I have an older version of Cygwin (B20) under which "make" 
> works fine.  I 
> recently installed the latest version of Cygwin.  When I type 
> "make" at the 
> command prompt, I get the message:
> 
> bash: make: command not found
> 
> I have a Makefile in the local directory and ran a 
> ./configure script which 
> ran successfully.
> 
> So how do I configure my bash shell to recognize the make 
> command ?  When I 
> look at my environment variables, the only variable related 
> to make is:
> 
> MAKE_MODE=unix
> 
> which I'm sure doesn't have any bearing on this.
> 
> I'm sure the answer is something simple.  It's been a while 
> since I used 
> makefiles.
> 
> Can anyone help me ?
> 


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

unzip problem after upgrading, was: cygwin/xfree86/emacs problems after upgrading

[Thomas L Roche]

Please excuse my delay in debugging this:

On Tue, 3 Feb 2004, Thomas L Roche wrote:
>> However I'm also, again only since upgrading [cygwin, from
>> 1.5.5-1], having problems with unzip, e.g.

>> >   inflating:
>> 
/d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class

>> > 735 [main] unzip 1820 cmalloc: cmalloc returned NULL
>> > error:  zipfile probably corrupt (segmentation violation)

>> which I get even unzip'ing a known-good file (i.e. one that I
>> unzip'ed successfully before upgrading from 1.5.5-1).

Igor Pechtchanski Wed, 4 Feb 2004 16:21:29 -0500 (EST)
> I'd suspect the unzip issue has to do with the path length to the
> file. 

Hmm: then why could I unzip the same file before upgrading, but not
after?

> You can try confirming it by creating a file in a deeply nested
> directory, putting that one file in a .zip, and then trying to unzip
> it into some place even deeper in the directory hierarchy.

OK, using the same path alleged to break above:

bash-2.05b$ pushd /d/temp

bash-2.05b$ mkdir -p qwer/asdf/zxcv

bash-2.05b$ mkdir -p 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation

bash-2.05b$ echo 'Foo! Bar! Baz!' > 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class

bash-2.05b$ cat 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
Foo! Bar! Baz!

bash-2.05b$ zip foo.zip 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
  adding: 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
 
(deflated 7%)

bash-2.05b$ unzip foo.zip -d qwer/asdf/zxcv/
Archive:  foo.zip
  inflating: 
qwer/asdf/zxcv/d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
 
 

bash-2.05b$ cat 
qwer/asdf/zxcv/d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
 

Foo! Bar! Baz!

as script:

pushd /d/temp
mkdir -p qwer/asdf/zxcv
mkdir -p 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation
echo 'Foo! Bar! Baz!' > 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
cat 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
zip foo.zip 
d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
unzip foo.zip -d qwer/asdf/zxcv/
cat 
qwer/asdf/zxcv/d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class
 


This works on my box, which is presently up-to-date. Are there other
hypotheses? If so, please let me know (and if possible, please reply
directly to me as well as to the list.)

If not: ISTM that something "down deep" in the base broke after 1.5.5-1:

0 both character-mode (e.g. unzip) and X (e.g. emacs) apps are broken

1 inputs that worked in 1.5.5-1 don't work now

2 failures are intermittent and apparently random

3 some error messages suggest malloc problems 

Am I missing something?


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

VIm packaging request/question

[Igor Pechtchanski]

Hi, Corinna,

Whenever you release the next version of VIm, would it be possible to
create a /usr/bin/vimless symlink pointing to $VIMRUNTIME/macros/less.sh?
Or is this an issue that should be taken up with the upstream people?
Igor
P.S. I just realized that this might require a manpage for 'vimless'...
-- 
http://cs.nyu.edu/~pechtcha/
  |\  _,,,---,,_[EMAIL PROTECTED]
ZZZzz /,`.-'`'-.  ;-;;,_[EMAIL PROTECTED]
 |,4-  ) )-,_. ,\ (  `'-'   Igor Pechtchanski, Ph.D.
'---''(_/--'  `-'\_) fL a.k.a JaguaR-R-R-r-r-r-.-.-.  Meow!

"I have since come to realize that being between your mentor and his route
to the bathroom is a major career booster."  -- Patrick Naughton

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: unzip problem after upgrading, was: cygwin/xfree86/emacs problems after upgrading

[Christopher Faylor]

On Wed, Feb 11, 2004 at 06:09:37PM -0500, Thomas L Roche wrote:
>Am I missing something?

Actual debugging of the problem, with a debugger?

cgf

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: zsh-4.1.1-2 still seems broken under Cygwin-1.5.7-1 .

[Larry Hall]

At 05:48 PM 2/11/2004, Morris Siegel you wrote:
>My PC is running under Windows XP Professional, and until recently with 
>Cygwin-1.5.5-1 and zsh-4.1.1-1 .  I upgraded to Cygwin-1.5.6-1, installing everything 
>available, in particular including zsh-4.1.1-2 .  zsh behaved in a buggy fashion.  I 
>reported it; you kindly replied that similarly problems had been reported by others, 
>and that Cygwin-1.5.7-1 should fix matters.  I upgraded to that, and the behavior is 
>improved, but still buggy: (1) sometimes when I start zsh it hangs, sometimes it 
>starts normally; (2) zsh command-line editing generally badly messes up the display 
>when long command lines are being edited.  I seem to recall that problem (2) was a 
>consistent nuisance a while back, but then some new Cygwin release (I don't remember 
>which) fixed it.  (I generally run zsh under screen, by the way.) 
>The only reason I upgraded from Cygwin-1.5.5 was to have the most up-to-date release; 
>it behaved without problem for me.  Is there any way I can restore it? 
>Thanks for your attention and reply. 
>-- Morris M. Siegel
> [EMAIL PROTECTED]


Sure.  Run setup again and downgrade all the packages you upgraded.  But
you're better off trying a snapshot to see if that fixes any problems
you're seeing.  Otherwise, the problems may persist beyond 1.5.7.  It's
good to be moving things forward rather than being stuck in the past. ;-)


--
Larry Hall  http://www.rfk.com
RFK Partners, Inc.  (508) 893-9779 - RFK Office
838 Washington Street   (508) 893-9889 - FAX
Holliston, MA 01746 


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: unzip problem after upgrading

[Thomas L Roche]

On Tue, 3 Feb 2004, Thomas L Roche wrote:
>>> However I'm also, again only since upgrading [cygwin, from
>>> 1.5.5-1], having problems with unzip, e.g.

 inflating:
>>> 
/d/eclipse/builds/20040121_1953-WB213-AD-V512D-00/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/presentation/EJBEditorOutlineContentProvider.class

>>> > 735 [main] unzip 1820 cmalloc: cmalloc returned NULL
>>> > error:  zipfile probably corrupt (segmentation violation)

>>> which I get even unzip'ing a known-good file (i.e. one that I
>>> unzip'ed successfully before upgrading from 1.5.5-1).

One more data point: just now I tried unzipping a new build, which a
colleague has unzipped using different tools. Unzip failed again,
slightly differently:

>   inflating: 
/d/eclipse/builds/20040210_1925-WB213-AD-V512D-W5/eclipse/plugins/com.ibm.etools.j2ee.ui/build/pluginBuild/commonbld/com/ibm/etools/ejb/ui/actions/CreateEJBRefAction.class
 
 
> 657 [main] unzip 2244 cmalloc: cmalloc returned NULL
> error:  zipfile probably corrupt (segmentation violation)
> Archive:  /d/eclipse/zips/wsa-jdk-20040210_1925-WB213-AD-V512D-W5.zip

Failed in the same plugin? Anyway:

Thomas L Roche 02/11/2004 06:08 PM
>> it seems pretty clear to me that something "down deep" in the base
>> broke after 1.5.5-1:

>> 0 both character-mode (e.g. unzip) and X (e.g. emacs) apps are broken

>> 1 inputs that worked in 1.5.5-1 don't work now

>> 2 failures are intermittent and apparently random

>> 3 error messages suggest malloc problems 

>> Am I missing something?

Christopher Faylor Wed, 11 Feb 2004 18:45:42 -0500
> Actual debugging of the problem, with a debugger?

Arrgghh--lousy timing, but pain threshold is met. 2Q:

* Is there a debugging howto more {complete, up-to-date} than

http://sources.redhat.com/cgi-bin/cvsweb.cgi/~checkout~/src/winsup/cygwin/how-to-debug-cygwin.txt?rev=1.9&content-type=text/plain&cvsroot=src

* Can someone provide directions (better yet, a script) for

http://sources.redhat.com/cgi-bin/cvsweb.cgi/~checkout~/src/winsup/cygwin/how-to-debug-cygwin.txt?rev=1.9&content-type=text/plain&cvsroot=src
> 1.  First things first

> The first thing you'll need to do is to build cygwin1.dll and
> your crashed application from sources. To debug them you'll need
> debug information, which is normally stripped from executables.
> You probably also want to build a version of the dll with more
> debugging capabilities by reconfiguring your build directory,
> specifying the --enable-debugging option to configure.


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

linking with static .lib

[Vikram Shrowty]

Hi,
   I have a windows static .lib that I want to use in a cygwin program.
   Now, since 
the .lib requires the microsoft c-runtime, I cant directly link it into
my cygwin program. So I thought I'd make a dll out of it that is linked
with the microsoft
c-runtime and then link with the import lib of that dll.
   So I did this to try and create the dll:
%gcc -mno-cygwin -shared -o mydll.dll -Wl,--export-all-symbols
-Wl,--whole-archive windows_static.lib Wl,--no-whole-archive

But im getting these errors:
fu01.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
fu02.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
fu03.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
fu04.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
fu05.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
fu06.o(.idata$3+0xc): more undefined references to
`_libmsvcrt_a_iname' follow
nmth00.o(.idata$4+0x0): undefined reference to `__nm___pctype'
nmth14.o(.idata$4+0x0): undefined reference to `__nmmb_cur_max'
nmth28.o(.idata$4+0x0): undefined reference to `__nm___iob'

And these warnings:
Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
Warning: .drectve `-defaultlib:LIBCMT ' unrecognized
Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
Warning: .drectve `-defaultlib:LIBCMT ' unrecognized
Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
.
.
many more of these.



First of all, am I going down the right path in linking to the lib? And
if so, what am
I doing wrong.

Thanks,
--Vikram

-- 
  Vikram Shrowty
  [EMAIL PROTECTED]

-- 
http://www.fastmail.fm - And now for something completely different…

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

debugging unzip, was: unzip problem after upgrading

[Thomas L Roche]

Thomas L Roche 02/11/2004 06:08 PM
>>> it seems pretty clear to me that something "down deep" in the base
>>> broke after 1.5.5-1:

>>> 0 both character-mode (e.g. unzip) and X (e.g. emacs) apps are
>>>   broken

>>> 1 inputs that worked in 1.5.5-1 don't work now

>>> 2 failures are intermittent and apparently random

>>> 3 error messages suggest malloc problems 

>>> Am I missing something?

Christopher Faylor Wed, 11 Feb 2004 18:45:42 -0500
>> Actual debugging of the problem, with a debugger?

Fortunately, useful help arrives:

Igor Pechtchanski 02/11/2004 09:20:22 PM:
> Make sure you have the "tar", "patch", "gcc", "gcc-g++", "gdb", and
> "make" packages installed (via Cygwin's setup).

> Download the unzip source (e.g., "wget ftp://mirrors.rcn.
> net/pub/sourceware/cygwin/release/unzip/unzip-5.50-5-src.tar.bz2").

> Untar the unzip source to /usr/src (e.g., "tar -C /usr/src -xjf
> unzip-5.50-5-src.tar.bz2").

> Enter that directory and run the prep, conf, build, and install
> steps of the script (e.g., "cd /usr/src && ./unzip-5.50-5.sh prep &&
> . /unzip-5.50-5.sh conf && ./unzip-5.50-5.sh build &&
> ./unzip-5.50-5.sh install").

> You should now have a debug build of unzip sitting in
> /usr/src/unzip-5.50/.inst (the maintainer graciously set it up to
> build with debugging by default)

more below

> Enter that directory and run gdb: "cd /usr/src/unzip-5.50/.inst && gdb
> --args ./unzip failing_zip_file"

almost: ./usr/bin/unzip.exe

I do

wget -O - 
ftp://mirrors.rcn.net/pub/sourceware/cygwin/release/unzip/unzip-5.50-5-src.tar.bz2 
| tar -C /usr/src -xj
cd /usr/src && ./unzip-5.50-5.sh prep && ./unzip-5.50-5.sh conf && 
./unzip-5.50-5.sh build && ./unzip-5.50-5.sh install
cd /usr/src/unzip-5.50/.inst && gdb --args ./usr/bin/unzip.exe 
/d/eclipse/zips/wsa-jdk-20040210_1925-WB213-AD-V512D-W5.zip

and get

> GNU gdb 2003-09-20-cvs (cygwin-special)
> Copyright 2003 Free Software Foundation, Inc.
> GDB is free software, covered by the GNU General Public License, and you 
are
> welcome to change it and/or distribute copies of it under certain 
conditions.
> Type "show copying" to see the conditions.
> There is absolutely no warranty for GDB.  Type "show warranty" for 
details.
> This GDB was configured as "i686-pc-cygwin"...(no debugging symbols 
found)...

Igor Pechtchanski 02/11/2004 09:20:22 PM:
> Set a breakpoint at the failing line (fileio.c, line 146) and start 
> debugging :-)

I do

(gdb) break fileio.c:146

> No symbol table is loaded.  Use the "file" command.

(gdb) file ./usr/bin/unzip.exe

> Reading symbols from ./usr/bin/unzip.exe...(no debugging symbols 
found)...done.

This sounds bad: if so, how to tweak the script above to include
debugging?


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: linking with static .lib

[Larry Hall]

At 09:27 PM 2/11/2004, Vikram Shrowty you wrote:
>Hi,
>   I have a windows static .lib that I want to use in a cygwin program.
>   Now, since 
>the .lib requires the microsoft c-runtime, I cant directly link it into
>my cygwin program. So I thought I'd make a dll out of it that is linked
>with the microsoft
>c-runtime and then link with the import lib of that dll.
>   So I did this to try and create the dll:
>%gcc -mno-cygwin -shared -o mydll.dll -Wl,--export-all-symbols
>-Wl,--whole-archive windows_static.lib Wl,--no-whole-archive
>
>But im getting these errors:
>fu01.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
>fu02.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
>fu03.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
>fu04.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
>fu05.o(.idata$3+0xc): undefined reference to `_libmsvcrt_a_iname'
>fu06.o(.idata$3+0xc): more undefined references to
>`_libmsvcrt_a_iname' follow
>nmth00.o(.idata$4+0x0): undefined reference to `__nm___pctype'
>nmth14.o(.idata$4+0x0): undefined reference to `__nmmb_cur_max'
>nmth28.o(.idata$4+0x0): undefined reference to `__nm___iob'
>
>And these warnings:
>Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
>Warning: .drectve `-defaultlib:LIBCMT ' unrecognized
>Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
>Warning: .drectve `-defaultlib:LIBCMT ' unrecognized
>Warning: .drectve `-defaultlib:OLDNAMES ' unrecognized
>.
>.
>many more of these.
>
>
>
>First of all, am I going down the right path in linking to the lib? 


No, not at all unfortunately.  Doing this doesn't get you around your 
initial stated concern, which is a valid one, of linking MSVCRT.  You'll
either need to remove the dependency in the library code itself or forgo
using the library.  There's a slim chance you'd get lucky and squeak by
with both MSVCRT and cygwin1.dll in the same code but if you don't know
what you're doing, the likelihood of success for this is minimal.  I 
can't recommend it.


>And
>if so, what am
>I doing wrong.


I guess, in a nutshell, you're starting from a faulty premise as I 
explained above.  The rest of what you see is just fallout from that
first mistep.

Sorry I don't have better news for you.


--
Larry Hall  http://www.rfk.com
RFK Partners, Inc.  (508) 893-9779 - RFK Office
838 Washington Street   (508) 893-9889 - FAX
Holliston, MA 01746 


--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

Re: zsh-4.1.1-2 still seems broken under Cygwin-1.5.7-1 .

[Peter A. Castro]

On Wed, 11 Feb 2004, Morris Siegel wrote:

> My PC is running under Windows XP Professional, and until recently with
> Cygwin-1.5.5-1 and zsh-4.1.1-1 .  I upgraded to Cygwin-1.5.6-1, installing
> everything available, in particular including zsh-4.1.1-2 .  zsh behaved in
> a buggy fashion.  I reported it; you kindly replied that similarly problems
> had been reported by others, and that Cygwin-1.5.7-1 should fix matters.  I
> upgraded to that, and the behavior is improved, but still buggy: (1)
> sometimes when I start zsh it hangs, sometimes it starts normally; (2) zsh
> command-line editing generally badly messes up the display when long command
> lines are being edited.  I seem to recall that problem (2) was a consistent
> nuisance a while back, but then some new Cygwin release (I don't remember
> which) fixed it.  (I generally run zsh under screen, by the way.)

cgf said in earlier email that this has been fixed in CVS and will be
pushed into the next release.  For the moment, if you really need zsh
(like me :), downgrade to 1.5.5 and wait for the update to appear.

> The only reason I upgraded from Cygwin-1.5.5 was to have the most up-to-date
> release; it behaved without problem for me.  Is there any way I can restore
> it?
>
> Thanks for your attention and reply.
>
>  -- Morris M. Siegel
>   [EMAIL PROTECTED]

-- 
Peter A. Castro <[EMAIL PROTECTED]> or <[EMAIL PROTECTED]>
"Cats are just autistic Dogs" -- Dr. Tony Attwood

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

RE: XEmacs-21.4.14-2 consistently crashes

[Demmer, Thomas]

>> "Thomas" == Thomas Demmer writes:
>
>Thomas> Until last week I was using 21.4.11 or so from xemacs.org,
built with
>Thomas> whatever cygwin they used and it ran fine, so I was assuming a
cygwin bug
>Thomas> here.
>
>In which mode does it run ?
>
>The following is copied from a header file in the XEamcs source tree,
>windowsnt.h:
>
>1. Keep in mind that there are two possible OS environments we are dealing
>   with -- Cygwin and Native Windows.  Cygwin provides a POSIX emulation
>   layer on top of MS Windows -- in particular, providing the file-system,
>   process, tty, and signal semantics that are part of a modern, standard
>   Unix operating system.  MS Windows also provides these services, but
>   through their own API, called Win32.  When compiling in a Cygwin
>   environment, the Win32 API's are also available, and in fact are used
>   to do native GUI programming.
>
>2. There are two windowing environments we can target XEmacs for when
>   running under MS Windows -- Windows native, and X. (It may seem strange
>   to write an X application under Windows, but there are in fact many X
>   servers out there running on Windows, and as far as I know there is no
>   real (or at least, that works well) networking Window-system extension
>   under MS Windows.  Furthermore, if you're porting a Unix application to
>   Windows and use Cygwin to assist you, it might seem natural to use an
>   X server to avoid having to port all the code to Windows.) For XEmacs,
>   there are various reasons people could come up with for why we would
>   want to keep maintaining X Windows under MS Windows support.
>
>That gives us four possible build environments.  I (Ben) build
>regularly on fully-native-everything, Andy builds on Cygwin + MS
>Windows + X Windows for windowing.

It used to run in Cygwin and MS Windows mode. It did not have X11
support built in, IIRC. Expressed with the flags you mentioned:
CYGWIN,HAVE_MS_WINDOWS

[...]

Ciao
Tom

--
Unsubscribe info:  http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/