A PROBLEM

[Anubhav Agrawal]

SIR,
   There are 2 VC++ projects. one of them is using MFC 
APPLICATION WIZARD and the other is a WIN32 CONSOLE program. Now 
what i require is to somehow use the MFC program in the WIN32 
console program.
 Is there any way of doing it directly.
 waitng for ur answer ( can u please send the answer on my email 
id also)

thankin you in anticipation

ANUBHAV AGRAWAL
___
Get www. mycompany .com and 5 matching email ids.
Just Rs. 1499/ year.
Click here http://www.rediffmailpro.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/

Re: exim 4.20.x patch to handle sendmail -O switch

[Philip Hazel]

On Tue, 10 Jun 2003, Jason Pyeron wrote:

> Dear Sirs/Madams,
>
> Please find enclosed patches for your respective distributions. These
> patches allow exim to not fail on sendmail like switch -O.

Thanks, Jason. I've added it to my list of things to look at.

Regards,
Philip

-- 
Philip HazelUniversity of Cambridge Computing Service,
[EMAIL PROTECTED]  Cambridge, England. Phone: +44 1223 334714.


--
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: A PROBLEM

[Elfyn McBratney]

On Wed, 11 Jun 2003, Anubhav Agrawal wrote:

> SIR,
> There are 2 VC++ projects. one of them is using MFC
> APPLICATION WIZARD and the other is a WIN32 CONSOLE program. Now
> what i require is to somehow use the MFC program in the WIN32
> console program.
>   Is there any way of doing it directly.
>   waitng for ur answer ( can u please send the answer on my email
> id also)
>
> thankin you in anticipation

Sorry, does this have anything to do with Cygwin? If not, then please take your
questions to a Windows forum or newsgroup.

Elfyn

-- 
Elfyn McBratney (mailto:[EMAIL PROTECTED])
Systems Administrator
ABCtales.com / Ubertales.co.uk


--
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: AC_CYGWIN?

[Ronald Landheer-Cieslak]

Note, though, that there is absolutely no problem with the filename 
".mumbleinit" on Cygwin (I just did a `touch .mumbleinit` to make sure)

In fact, there are less and less reasons to distinguish Cygwin from *NIX 
platforms. (Thanks to the core developers and various contributers for 
that!)

rlc

On Tue, 10 Jun 2003, Alexander Enchevich wrote:

> Thanks for the tip Max, I looked in 'info autoconf' and found this 
> 
> =info autoconf===
> If you want to base a decision on the system where your program 
> will be run, make sure you use the `$host' variable, as in the 
> following excerpt:
> 
>  case $host in
>*-*-msdos* | *-*-go32* | *-*-mingw32* | *-*-cygwin* | *-*-windows*)
>  MUMBLE_INIT="mumble.ini"
>  ;;
>*)
>  MUMBLE_INIT=".mumbleinit"
>  ;;
>  esac
>  AC_SUBST([MUMBLE_INIT])
> =info autoconf===
> 
> So I guess that's all I need...
> 
> >Of course, checking for a platform kind of goes against the whole
> >feature-not-platform tests ideal of autoconf.
> 
> I have to! :/ The code is already there, full of defines (#ifdef WIN32 and
> #ifdef APPLE) and it was not written with Linux or cygwin in mind and now it
> has to be converted, so... 
> 
> -Original Message-
> From: Max Bowsher [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, June 10, 2003 5:16 PM
> To: Alexander Enchevich; [EMAIL PROTECTED]
> Subject: Re: AC_CYGWIN? 
> 
> 
> Alexander Enchevich wrote:
> > Hi
> >
> > What's the proper way to check if I am compiling on a cygwin system from
> > within an autoconf configure.in script?
> 
> >From the subject, you clearly already have an idea. If you check the
> autoconf docs, they will tell you that that macro is obsolete, and tell you
> how you should be doing it.
> 
> Of course, checking for a platform kind of goes against the whole
> feature-not-platform tests ideal of autoconf.
> 
> 
> Max.
> 
> 
> --
> 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/
> 
> --
> 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/
> 


--
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: [ANNOUNCEMENT] Updated: libxslt-1.0.30-2

[Patrick Eisenacher]

Elfyn,

you rock! :o)

The new version runs smooth like butter.

Thank you so much for all your effort and energy and time that you put 
into making this new version happen. You did a tremendous job!

Patrick

I've made a new version of LibXSLT (1.0.30-2) available for downloading.

This is a *maintainance release* only. The only addition is a one-line
fix from current CVS, which fixes fixes the SIGSEGV mentioned in this
 thread.


--
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 w/ c++,threads,static initializers

[Rasmus Hahn]

Hi Thomas,

Problem solved: it was not the mutex but the but the pthread_mutexattr_t
that caused the error. cygwin's pthread_mutexattr_init () function checks
if its argument points to a valid object. If the pthread_mutexattr_t's 
value still points to a valid object (from a previous call)
pthread_mutexattr_init () fails, and so do later pthread_mutex_init()s with
this attribute. I think this behaviour is not correct since an uninitialized
pthread_mutexattr_t can point to anything (as a previously created attribute),
so this is probably a bug in cygwin (or what does the International Pthread
Attributes Committee say?). A quick look in winsup/cygwin/thread.cc shows
that this kind of check is done for more of the ..._init() functions, so there
may be similar effects (but i did not test that yet). Demo-Program below does
not work with (my) cygwin but does in my linux box:

#include 
int main () {
  pthread_mutexattr_t attr;
  pthread_mutexattr_init (&attr);
  if (pthread_mutexattr_init (&attr)) {
printf ("error should not happen\n");
  }
}

Greetings - Rasmus


> Rasmus Hahn wrote:
> > On Tue, Jun 10, 2003 at 09:25:43AM +0200, Thomas Pfaff wrote:
> > 
> >>Rasmus Hahn wrote:
> >>
> >>
> >>>Hello,
> >>>
> >>>i am using cygwin on my Windows2000 and my c++ program segfaults
> >>>when using pthread-mutexes. In particular i am using a lot of
> >>>mutexes from within constructors of static objects. These constructors
> >>>are called at a very early time of execution of the program and do
> >>>calls of pthread_mutex_init, pthread_mutex_lock and such. Taken a glance
> >>>at the cygwin source i noticed that the static initializers (constructors
> >>>of static objects) are run _before_ some user_data->threadinterface is
> >>>initialized (look at dll_crt0_1 () in winsup/cygwin/dcrt0.cc). Isnt it 
> >>>possible to use thread-functions from static initializers?
> >>>BTW i am using cygwin snapshot 20030602 compiled with gcc-3.3 since the
> >>>out-of-the-box gcc from the cygwin-package is a prerelease and definitely
> >>>contains bugs that do not allow it to compile my code.
> >>>
> >>
> >>Could you provide a test case ?
> > 
> > 
> >   No, i cannot. As it turned out the initializers are run at least but i
> >   dont know if before or after cygwin-initialization. Only when run with
> >   gdb the program behaves differently (crash in constructors). That makes
> >   reproduction & location of errors a little harder.
> > 
> 
> If your program only segfaults in the debugger than the problem is a
> harmless IsBadWritePtr which is used to check for an already initialized
> mutex.
> You can ignore this one. Set a breakpoint at main and continue.
> To avoid it you can set the mutex to PTHREAD_MUTEX_INITIALIZER before
> you call pthread_mutex_init.
> 
> Thomas
> 
> 
> 

-- 


--
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 w/ c++,threads,static initializers

[Thomas Pfaff]

Rasmus Hahn wrote:
Hi Thomas,

Problem solved: it was not the mutex but the but the pthread_mutexattr_t
that caused the error. cygwin's pthread_mutexattr_init () function checks
if its argument points to a valid object. If the pthread_mutexattr_t's 
value still points to a valid object (from a previous call)
pthread_mutexattr_init () fails, and so do later pthread_mutex_init()s with
this attribute. I think this behaviour is not correct since an uninitialized
pthread_mutexattr_t can point to anything (as a previously created attribute),
so this is probably a bug in cygwin (or what does the International Pthread
Attributes Committee say?). A quick look in winsup/cygwin/thread.cc shows
that this kind of check is done for more of the ..._init() functions, so there
may be similar effects (but i did not test that yet). Demo-Program below does
not work with (my) cygwin but does in my linux box:

#include 
int main () {
  pthread_mutexattr_t attr;
  pthread_mutexattr_init (&attr);
  if (pthread_mutexattr_init (&attr)) {
printf ("error should not happen\n");
  }
}
Have a look at
http://www.opengroup.org/onlinepubs/007904975/functions/pthread_mutexattr_init.html 

"Results are undefined if pthread_mutexattr_init() is called specifying 
an already initialized attr attributes object."

Unfortunately the spec does not specify EBUSY (which as returned at the 
moment) as a valid return code for pthread_mutexattr_init, therefore i 
will change all attr_init functions to simply return 0 if they are 
called with an already initialized attribute.

Thomas

--
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 w/ c++,threads,static initializers

[Robert Collins]

Because the results are undefined, you are permitted to return EBUSY...

Rob


--
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 w/ c++,threads,static initializers

[Thomas Pfaff]

Robert Collins wrote:
Because the results are undefined, you are permitted to return EBUSY...

Linux and Solaris do not return an error when the attr is already 
initialized, therefore i will do the change to return 0.
The code must be changed anyway, since some attr_init functions 
erroneously return EAGAIN instead of ENOMEM.

Thomas

--
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: non network install

[Max Bowsher]

B Thomas wrote:
> In-Reply-To: <[EMAIL PROTECTED]>
>
> Thank you very much for all your replies. Having the distribution in
> folder below the top one does indeed solve the problem.
>
> I wrote a bunch of tiny bash scripts to make the check dependencies and
> make the iso image for two CD's. I shall put up the scripts and detailed
> instructions on my web page this weekend. I will be gladly help in
> making the patch to the users manual too. I am familiar with TeX, LaTeX
> and HTML but not Texinfo. I shall try learning it , but that will take
> me time. If there is a posibility for converting LateX/Tex with Texinfo
> that would be great. However I would like to develop the scripts into
> a simple but effective program for making the images first.
> Here are some of my concerns with writing a program for making CD
> images, please do make suggetions as appropriate . Right now the way I
> have it set is that the entire distribution fits onto 2 CD's , except
> 1997 and 1998 mail-archives. I have tried to make these CD's so that
> all packages on CD 1 do not depend on any package on CD 2 . However
> packages in CD 2 may depend on CD 1. I did this as I was not sure how
> setup.exe would behave if it found unmet dependencies. I guess I could
> have deleted the older versions and unnecessary things like
> mail-archives so that it all fit on one CD. However personally I was
> disinclined to do this as I don't see why we must have such a
> restriction that the CD ROM distribution has to be one CD. Moreover
> Cywin is bound to grow. Also having older version may be a good idea
> since if there is a bug in a later version that is unacceptable to the
> user he may choose an older one.

That is true, but it is far easier to install from a single CD.
Cygwin does currently fit on a single CD.

Michael Chase's clean_setup.pl
(http://home.ix.netcom.com/~mchase/zip/) may help here:
It can parse setup.ini and produce a list of urls to download, or move files
around, subject to certain filters, including "not previous".

> The way I am doing the dependency checking right now is not very
> satisfactory and would appreciate suggestions . I greped through
> the setup.hints files in each package directory , picked out the
> "requires" lines , appropriately edited it using sed , makeing a table
> which I saved in a file. I also have another file that is a table with
> two columns - one the package name (directory) and the other the CD to
> which it belong. A script then uses these two files to check
> dependencies of CD 1 are fullfilled within it .
>
> I am acutely aware that this is not the way to go so would like to build
> a portable and robust script that would able to do all of the above
> using just the ftp archive . I would like to be insensitive to changes
> in the ftp archive and be portable to atleast windows and linux .
> I am familiar with C, C++, Java , BASH , Fortran and would be willing
> tor try another language if appropriate.  What do you think ?

Perl.

> Also both dpkg, and rpm have very powerfull dependency checking engines
> so this seems like reinventing the wheel. Alternatively is it better to
> hack the setup.exe code to do all this. I haven't read the code so am
> not very sure if it is portable .

No, setup.exe isn't portable. However, if you want to try and make it so, go
ahead :-) .


Max.


--
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: Some useful permission setting scripts

[Max Bowsher]

Max Polk wrote:
> WARNING:  If you omit "administrators" then if you reinstall Windows,
>   you can NEVER, NEVER, EVER get to the file again, even as
>   administrator!  (SOLUTION: use partition magic, convert
>   your drive to "FAT32" drops the bad permission, then
>   convert your drive back to "NTFS" which sets default
>   permissions letting you get to your files again).

Not true.
Let me introduce you to an easier way...

As an Administrator, simply "Take ownership" ( Properties > Security >
Advanced > Owner, and set it to yourself or Administrators ).
You can now edit the permissions.


--
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: Some useful permission setting scripts

[Corinna Vinschen]

On Wed, Jun 11, 2003 at 12:59:59PM +0100, Max Bowsher wrote:
> Max Polk wrote:
> > WARNING:  If you omit "administrators" then if you reinstall Windows,
> >   you can NEVER, NEVER, EVER get to the file again, even as
> >   administrator!  (SOLUTION: use partition magic, convert
> >   your drive to "FAT32" drops the bad permission, then
> >   convert your drive back to "NTFS" which sets default
> >   permissions letting you get to your files again).
> 
> Not true.
> Let me introduce you to an easier way...
> 
> As an Administrator, simply "Take ownership" ( Properties > Security >
> Advanced > Owner, and set it to yourself or Administrators ).
> You can now edit the permissions.

Or even use `chown' in Cygwin as administrator.
Btw., you can also use Cygwin's setfacl instead of cacls.

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_NT-5.0 ZPC207050 1.3.22(0.78/3/2) bug

[Hornsby, Kelly]

I just discovered that an environment variable OSTYPE is not longer exported by 
default to a subprocess under this version of Cygwin.  Why did this occur?  Was it 
intentional, or inadvertent?  I use this often, and although I can perform an export 
OSTYPE explicitly,  I now have lots of broken makefiles that won't work anymore.  Do 
you have any suggestions?  Otherwise, can this guy be fixed in the next minor release?


Kelly Hornsby
Software Engineer
Northrop Grumman Space Technology
293 Highway 247 South
Warner Robins GA 31088
1-478-329-2153 (voice)
1-478-929-7335 (fax)
mailto:[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/

Problem in shell scripting - possible memory leak / violation?

[Tomasz Rojek]

Hi,

After long hours (for example 7 days with hibernation / dehibernation only) of running
cygwin without restart it often happens that I get similar error to those below. I 
tried
to execute some commands in loop (it works perfectly after machine restart!):

$ for i in `seq 0 9`; do for j in `seq 0 9`; do echo $i$j; ls $i$j/; done; done
00
user_info.sah*  work_unit.sah*
01
user_info.sah*  work_unit.sah*
02
user_info.sah*  work_unit.sah*
03
user_info.sah*  work_unit.sah*
04
user_info.sah*  work_unit.sah*
05
user_info.sah*  work_unit.sah*
06
user_info.sah*  work_unit.sah*
07
user_info.sah*  work_unit.sah*
08
user_info.sah*  work_unit.sah*
09
user_info.sah*  work_unit.sah*
10
user_info.sah*  w144*
11
user_info.sah*  w145*
12
user_info.sah*  w143*
13
user_info.sah*  w142*
14
user_info.sah*  w141*
15
user_info.sah*  w140*
16
user_info.sah*  w139*
17
user_info.sah*  w138*
18
user_info.sah*  w137*
19
user_info.sah*  w136*
20
user_info.sah*  w135*
21
user_info.sah*  w134*
22
user_info.sah*  w133*
23
user_info.sah*  w132*
24
bash: /usr/bin/ls: Permission denied
25
bash: /usr/bin/ls: Permission denied
26
bash: /usr/bin/ls: Permission denied
27
bash: /usr/bin/ls: Permission denied
28
bash: /usr/bin/ls: Permission denied
29
bash: /usr/bin/ls: Permission denied
bash: /usr/bin/seq: Permission denied
bash: /usr/bin/seq: Permission denied
bash: /usr/bin/seq: Permission denied
bash: fork: Permission denied

So I have closed cygwin and any processes it had created. The I tried to run it once 
again
from CLI:

C:\>FOLDER\cygwin\cygwin_shell.bat
The system cannot execute the specified program.

Also - any CLI app I tried to run caused this error message:

Insufficient system resources exist to complete requested service.

Did anybody encounter such problem?
Greetings

-- 
Tomasz Rojek




--
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_NT-5.0 ZPC207050 1.3.22(0.78/3/2) bug

[Corinna Vinschen]

On Wed, Jun 11, 2003 at 06:04:40AM -0700, Hornsby, Kelly wrote:
> I just discovered that an environment variable OSTYPE is not longer exported by 
> default to a subprocess under this version of Cygwin.  Why did this occur?  Was it 
> intentional, or inadvertent?  I use this often, and although I can perform an export 
> OSTYPE explicitly,  I now have lots of broken makefiles that won't work anymore.  Do 
> you have any suggestions?  Otherwise, can this guy be fixed in the next minor 
> release?

Quote from the bash 2.05 CHANGES file:

Bash no longer auto-exports HOSTNAME, HOSTTYPE, MACHTYPE, or OSTYPE,
even if it assigns them default values.

export OSTYPE in your profile will help.

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/

command window/console with cols > 132, vim restore screen (t_ti, t_te)

[Ljubomir Josifovski]

When I use a command window/cygwin console with large number of
columns (e.g. 160, 180) vim does not restore the screen upon
exit/suspend (^Z) correctly (or maybe does not save it correctly
upon entry). The restored region is just a black area (black fg
on black bg?). The "console" is created via shortcut targeting
"C:\cygwin\bin\bash.exe --login -i", and I set the window size in
Properties/Layout/Window size/Width. Has anyone experienced the
same problem (I don't think it is new, nor it appeared recently
- I just got fed up enough to ask if anyone knows better), even
how to solve it?

My TERM=cygwin, does not happen with xterm, and the vim t_ti and
t_ti vars are as the example in the "help
restorescreen",

set t_ti=^[7^[[r^[[?47h t_te=^[[?47l^[8

and I have the current cygwin, vim, terminfo/termcap.

thanks in advance,

-- 
Ljubomir Josifovski

--
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_NT-5.0 ZPC207050 1.3.22(0.78/3/2) bug

[Max Bowsher]

Hornsby, Kelly wrote:
> I just discovered that an environment variable OSTYPE is not longer
exported by default to a
> subprocess under this version of Cygwin.  Why did this occur?  Was it
intentional, or
> inadvertent?  I use this often, and although I can perform an export
OSTYPE explicitly,  I now
> have lots of broken makefiles that won't work anymore.  Do you have any
suggestions?  Otherwise,
> can this guy be fixed in the next minor release?

Nothing to do with Cygwin.

This was a change in the upstream version of bash, in version 2.05a.

This version was released as a Cygwin package in NOVEMBER 2001 !!!

Your makefiles are making an invalid assumption that OSTYPE will always be
defined. Fix them to not assume that.



Max.


--
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/

colons with rsync

[Ben Smith]

I am trying to rsync from a SCO box to a Windows 2000 server as a
(temporary) backup while I build a more suitable Linux backup server.
Rsync on SCO reports errors such as the following:

rsync: recv_generator: mkdir
"var/opt/K/SCO/MMDF/2.43.3b/spool/mail/:saved": No such file or
directory (2)
stat var/opt/K/SCO/MMDF/2.43.3b/spool/mail/:saved : No such file or
directory

I know this has to do with the colons in the directory/file names.  All
of the files that have this error have colons in the path somewhere.

They are both a recent version, 2.5.6  protocol version 26.  I guess
another snafu is that my cygwin install is very basic, cygwin-lite with
ssh and rsync.

Does rsync convert between Windows/Unix filenames automatically in
instances such as this?  Is there a way to use cygpath to get around
this limitation?  Thanks in advance for your help.

-Ben Smith
[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: w32api/windef.h

[Igor Pechtchanski]

On Wed, 11 Jun 2003, Lionel Zhou wrote:

> - Original Message -
> From: Igor Pechtchanski <[EMAIL PROTECTED]>
> Date: Mon, 9 Jun 2003 09:17:21 -0400 (EDT)
> To: Lionel Zhou <[EMAIL PROTECTED]>
> Subject: Re: w32api/windef.h
>
> > On Sun, 8 Jun 2003, Lionel Zhou wrote:
> >
> > > - Original Message -
> > > From: Christopher Faylor <[EMAIL PROTECTED]>
> > > Date: Sun, 8 Jun 2003 11:21:43 -0400
> > > To: [EMAIL PROTECTED]
> > > Subject: Re: w32api/windef.h
> > >
> > > > On Sun, Jun 08, 2003 at 07:19:30AM -0500, Lionel Zhou wrote:
> > > > >In file included from utils.c:4:
> > > > >/usr/include/w32api/windef.h:273: parse error before "void"
> > > > >
> > > > >has anyone successfully used the windef.h header and could you tell me
> > > > >how to compile program like this?
> > > > >
> > > > >my windef.h is attached here.
> > > >
> > > > #include 
> > > >
> > > > Not windef.h.  That might help.
> > > >
> > > > cgf
> > >
> > > thank you christopher and elfyn,
> > >
> > > i tried your advice but it doesn't solve the problem.
> > >
> > > grep windef /usr/local/w32api/windows.h
> > > #include 
> > >
> > > so that windows.h is just a superset of header definitions which
> > > includes windef.h
> > >
> > > program compile with the same error:
> > >
> > > In file included from windows.h:48:
> > >  from utils.c:4:
> > > /usr/include/w32api/windef.h:273: parse error before "void"
> > >
> > > grep void /usr/include/w32api/windef.h
> > > #define NULL ((void*)0)
> > > typedef CONST void *PCVOID,*LPCVOID;
> > > typedef void *HGDIOBJ;
> > >
> > > i also tried the -mwin32 compile flag which lead to more errors because
> > > of conflicting definitions. the program is wrote for linux and is being
> > > ported to windows. this is about the only function that uses win32 api
> > > in this program.
> > >
> > > lionel
> >
> > Lionel,
> >
> > Check the part of your program *before* the '#include "windows.h"' (either
> > the program itself, or the header files).  This could be a common C
> > problem with a missing semicolon after a struct definition (e.g.,
> > something like
> >
> > struct AAA {
> >   int blah;
> > }
> >
> > void main(void) { ... }
> >
> > will give you a similar error).
> >   Igor
>
> thank you prof. igor,
>
> this is not the problem here. i have made a test program like this.
>
> utils.c
> #include "ffss.h"
> #include "utils.h"
> #include 
>
> char *FFSS_GetOS(void)
> {
>   return FFSS_SERVER_OS;
> }
> end here
>
> this short program won't compile, no win32 api is used.
>
> In file included from utils.c:4:
> /usr/include/w32api/windef.h:273: parse error before "void"
>
> how ever, if i change the order of the headers, like this:
>
> utils2.c
> #include 
> #include "ffss.h"
> #include "utils.h"
>
> char *FFSS_GetOS(void)
> {
>   return FFSS_SERVER_OS;
> }
> end here
>
> this still don't compile and a lot of errors crop out.
>
>
> In file included from ffss.h:58,
>  from utils.c:3:
> /usr/include/bzlib.h:163: two or more data types in declaration of `type name'
> /usr/include/bzlib.h:186: two or more data types in declaration of `type name'
> /usr/include/bzlib.h:262: two or more data types in declaration of `type name'
> In file included from utils.c:3:
> ffss.h:64:1: warning: "LOG_INFO" redefined
> In file included from /usr/include/syslog.h:14,
>  from /usr/local/include/skyutils.h:256,
>  from ffss.h:49,
>  from utils.c:3:
> /usr/include/sys/syslog.h:21:1: warning: this is the location of the previous 
> definition
> In file included from utils.c:3:
> ffss.h:65:1: warning: "LOG_WARNING" redefined
> In file included from /usr/include/syslog.h:14,
>  from /usr/local/include/skyutils.h:256,
>  from ffss.h:49,
>  from utils.c:3:
> /usr/include/sys/syslog.h:19:1: warning: this is the location of the previous 
> definition
> In file included from utils.c:3:
> ffss.h:66:1: warning: "LOG_ERR" redefined
> In file included from /usr/include/syslog.h:14,
>  from /usr/local/include/skyutils.h:256,
>  from ffss.h:49,
>  from utils.c:3:
> /usr/include/sys/syslog.h:18:1: warning: this is the location of the previous 
> definition
> In file included from /usr/include/sys/socket.h:15,
>  from /usr/include/cygwin/if.h:19,
>  from /usr/include/net/if.h:14,
>  from utils.h:6,
>  from utils.c:4:
> /usr/include/cygwin/socket.h:18: redefinition of `struct sockaddr'
> In file included from /usr/include/sys/socket.h:15,
>  from /usr/include/cygwin/if.h:19,
>  from /usr/include/net/if.h:14,
>  from utils.h:6,
>  from utils.c:4:
> /usr/include/cygwin/socket.h:28: redefinition of `struct linger'
> In file included from /usr/include/cygwin/if.h:19,
>  from /usr/include/net/if.h:14,

Re: colons with rsync

[Igor Pechtchanski]

On Wed, 11 Jun 2003, Ben Smith wrote:

> I am trying to rsync from a SCO box to a Windows 2000 server as a
> (temporary) backup while I build a more suitable Linux backup server.
> Rsync on SCO reports errors such as the following:
>
> rsync: recv_generator: mkdir
> "var/opt/K/SCO/MMDF/2.43.3b/spool/mail/:saved": No such file or
> directory (2)
> stat var/opt/K/SCO/MMDF/2.43.3b/spool/mail/:saved : No such file or
> directory
>
> I know this has to do with the colons in the directory/file names.  All
> of the files that have this error have colons in the path somewhere.
>
> They are both a recent version, 2.5.6  protocol version 26.  I guess
> another snafu is that my cygwin install is very basic, cygwin-lite with
> ssh and rsync.
>
> Does rsync convert between Windows/Unix filenames automatically in
> instances such as this?  Is there a way to use cygpath to get around
> this limitation?  Thanks in advance for your help.
>
> -Ben Smith

Ben,

Windows filenames cannot contain ':', so there is no way around this
limitation in Cygwin, since it uses Windows filenames.

I did manage to create a file called ':saved' (on NTFS).  It didn't show
up in directory listings (simple 'ls'), but did in an explicit 'ls
:saved'.  I also could write to it and read from it, so I'm not sure
exactly where or how the file was created ('mv :saved ./Zsaved' said "mv:
cannot move `:saved' to a subdirectory of itself, `./Zsaved'", so I'm
guessing Win2k got pretty confused).  'rm :saved' worked, though.
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: colons with rsync

[Max Bowsher]

Igor Pechtchanski wrote:
> I did manage to create a file called ':saved' (on NTFS).  It didn't show
> up in directory listings (simple 'ls'), but did in an explicit 'ls
> :saved'.  I also could write to it and read from it, so I'm not sure
> exactly where or how the file was created ('mv :saved ./Zsaved' said "mv:
> cannot move `:saved' to a subdirectory of itself, `./Zsaved'", so I'm
> guessing Win2k got pretty confused).  'rm :saved' worked, though.

I think the data went into a named stream attached to the directory you did
these tests in. That would be consistent with the mv error.


Max.


--
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: colons with rsync

[Igor Pechtchanski]

On Wed, 11 Jun 2003, Max Bowsher wrote:

> Igor Pechtchanski wrote:
> > I did manage to create a file called ':saved' (on NTFS).  It didn't show
> > up in directory listings (simple 'ls'), but did in an explicit 'ls
> > :saved'.  I also could write to it and read from it, so I'm not sure
> > exactly where or how the file was created ('mv :saved ./Zsaved' said "mv:
> > cannot move `:saved' to a subdirectory of itself, `./Zsaved'", so I'm
> > guessing Win2k got pretty confused).  'rm :saved' worked, though.
>
> I think the data went into a named stream attached to the directory you did
> these tests in. That would be consistent with the mv error.
>
> Max.

Yeah, thanks.  I remember there was a discussion of this a while ago, but
couldn't find it (I did browse the user's guide too, with similar
results).  In any case, this is not at all what the OP wanted.
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/

Error message when compling simple program, help please :)

[Max Palatnik]

Hi guys,

First time emailing a community like this, hopefully it works.  I just 
installed cygwin using the setup program available on the website and I'm 
trying to copile a program I have from the Sams publishing "teach yourself C 
in 24 hours book"

The program is the following:

#include 

main()
{
  printf ("Howdy, neighbor! This is my first C program.\n");
  return 0;
}
I am using the following command:

$ gcc 02l01.c
/usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin/bin/ld: 
cannot find -luser32
collect2: ld returned 1 exit status

I have no idea what to do now. I searched google for help but i didn't find 
anything referring specifically to the -luser32 error.  If someone could 
help me out I'd really appreciate it :)

Thanks to all who reply

Max

_
Tired of spam? Get advanced junk mail protection with MSN 8. 
http://join.msn.com/?page=features/junkmail

--
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: Error message when compling simple program, help please :)

[Robb, Sam]

> $ gcc 02l01.c
> /usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin
> /bin/ld: 
> cannot find -luser32
> collect2: ld returned 1 exit status
> 
> I have no idea what to do now. I searched google for help but 
> i didn't find anything referring specifically to the -luser32
> error.  If someone could help me out I'd really appreciate it :)
> 
> Thanks to all who reply

Max,

  You can use http://cygwin.com/packages/ to find out which
Cygwin package provides a particular file.

  In your case, the "-luser32" switch to GCC is telling it to
link to a library named "libuser32".  Using the package search
page, you can find out that libuser32 is provided by:

  w32api/w32api-2.2-1   Win32 API header and library import files

  So - did you install the w32api package?

  BTW - general development questions are generally not appropriate
fot this list, unless they're specific to Cygwin.

-Samrobb

--
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: Processes are unable to fully discard their controlling terminal

[Matt]

I would like ssh-agent to function like you say it is
for you... but i can't understand what you did besides
obtain the latest versions which i should have..  what
else do i need to do?
-
Christopher Faylor wrote:
> On Sun, Mar 02, 2003 at 12:43:28PM -0500,
Christopher Faylor wrote:
>> On Sun, Mar 02, 2003 at 12:33:04PM -0500,
Christopher Faylor wrote:
>>> On Sun, Mar 02, 2003 at 03:33:56PM -, Max
Bowsher wrote:
 Processes that attempt to shed their controlling
terminal do so
 well enough for the tty to show as ? in ps
output, but they still
 keep the console window they were launched from
open, after all
 other processes using it have exited.

 This is particularly annoying with ssh-agent - if
I exit the shell
 in the terminal that launched it, that terminal
becomes a zombie,
 hanging around for no purpose. If I close that
window using the
 Windows close box, the ssh-agent is killed.

 I'm not an expert in the ways of Unix ttys. Can
anyone help me
 understand where the problem lies?
>>>
>>> setsid()
>>
>> I think I see the problem.  It will be tricky to
fix but I'll see
>> what I can do.
>
> I've checked in some changes that seem to fix the
reported behavior.
>
> Does the current snapshot rectify this behavior?

It does. Thankyou *very* much! A same-day bugfix is
absolutely amazing!

I can now successfully share an ssh-agent between all
my shells, with it
starting with the first one, and ending with the last,
and no zombie windows
if I end shells in the wrong order.


Max.


__
Do you Yahoo!?
Yahoo! Calendar - Free online calendar with sync to Outlook(TM).
http://calendar.yahoo.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/

Re: Perl 5.8.0-3 requires binary-mounted .cpan

[Gerrit P. Haase]

Gary schrieb:

> -1 worked with text mounts, never tried binary ones but I have to assume they
> worked as well.  This is with cygwin1.dll "1.5.0s(0.87/3/2) 20030608".

Hmmm, this is not better as before, with the default (unpatched)
perl-5.8.0 we get CR/LF lineendings on every perl generated output,
regardless which mode the mount is, unless PERLIO=raw or PERLIO=perlio
is used as environment setting, now, after I patched it, everything
breaks when perl is used on textmode mounts.


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: AC_CYGWIN?

[Alexander Enchevich]

> Note, though, that there is absolutely no problem with the filename 
> ".mumbleinit" on Cygwin (I just did a `touch .mumbleinit` to make sure)

On Cygwin there isn't but on "Win" there is - try copying/renaming an
existing file to something like .filename in Explorer (I tried it on Win2K).
You will get an err. message informing you that "You must type a filename"
:)

-Original Message-
From: Ronald Landheer-Cieslak [mailto:[EMAIL PROTECTED]
Sent: Wednesday, June 11, 2003 1:17 AM
To: Alexander Enchevich
Cc: [EMAIL PROTECTED]
Subject: RE: AC_CYGWIN? 


Note, though, that there is absolutely no problem with the filename 
".mumbleinit" on Cygwin (I just did a `touch .mumbleinit` to make sure)

In fact, there are less and less reasons to distinguish Cygwin from *NIX 
platforms. (Thanks to the core developers and various contributers for 
that!)

rlc

On Tue, 10 Jun 2003, Alexander Enchevich wrote:

> Thanks for the tip Max, I looked in 'info autoconf' and found this 
> 
> =info autoconf===
> If you want to base a decision on the system where your program 
> will be run, make sure you use the `$host' variable, as in the 
> following excerpt:
> 
>  case $host in
>*-*-msdos* | *-*-go32* | *-*-mingw32* | *-*-cygwin* | *-*-windows*)
>  MUMBLE_INIT="mumble.ini"
>  ;;
>*)
>  MUMBLE_INIT=".mumbleinit"
>  ;;
>  esac
>  AC_SUBST([MUMBLE_INIT])
> =info autoconf===
> 
> So I guess that's all I need...
> 
> >Of course, checking for a platform kind of goes against the whole
> >feature-not-platform tests ideal of autoconf.
> 
> I have to! :/ The code is already there, full of defines (#ifdef WIN32 and
> #ifdef APPLE) and it was not written with Linux or cygwin in mind and now
it
> has to be converted, so... 
> 
> -Original Message-
> From: Max Bowsher [mailto:[EMAIL PROTECTED]
> Sent: Tuesday, June 10, 2003 5:16 PM
> To: Alexander Enchevich; [EMAIL PROTECTED]
> Subject: Re: AC_CYGWIN? 
> 
> 
> Alexander Enchevich wrote:
> > Hi
> >
> > What's the proper way to check if I am compiling on a cygwin system from
> > within an autoconf configure.in script?
> 
> >From the subject, you clearly already have an idea. If you check the
> autoconf docs, they will tell you that that macro is obsolete, and tell
you
> how you should be doing it.
> 
> Of course, checking for a platform kind of goes against the whole
> feature-not-platform tests ideal of autoconf.
> 
> 
> Max.
> 
> 
> --
> 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/
> 
> --
> 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/
> 

--
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: Processes are unable to fully discard their controlling terminal

[Max Bowsher]

Some time ago I (Max Bowsher) wrote:
> I can now successfully share an ssh-agent between all
> my shells, with it
> starting with the first one, and ending with the last,
> and no zombie windows
> if I end shells in the wrong order.

Matt wrote:
> I would like ssh-agent to function like you say it is
> for you... but i can't understand what you did besides
> obtain the latest versions which i should have..  what
> else do i need to do?

Some shell scripting.

Attached are the shell scripts I *source* from my .bash_profile and
.bash_logout.

It's not a perfect system (if a shell is killed in such a way that
.bash_logout doesn't execute, an agent process can remain), but it seems to
work well for the most part.



Max.


agent_refcount_start
Description: Binary data


agent_refcount_stop
Description: Binary data
--
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/

port query: any "fping"? & cygwin headers

[Soren Andersen]

Hey!

Hello all. Has anyone ever ported the network tool "fping" to cygwin? I
came across some Google results that seemed to be viable but it appears
not (I don't have the complete URL I checked, but it was a site that
showed a dir listing under a software project named "mon". maybe some of
you know what "mon" is). Anyway, the version of fping found was very
old (2.2b1).

The source rpm from the above referenced (kind of) site was an rpm and
right there it doesn't seem to be a cygwin port. But the url did seem to
include "cygwin". Anyway I extracted the sources from the rpm file and
tried to configure and build. I get a bunch of undefined symbols, like

  ICMP_

where a lot of things follow "ICMP_". I checked headers and looked for
any such defines in any header mentioned, found none. I did find that
there's a header

  /usr/include/cygwin/icmp.h

that is an empty file. Is this a TODO for Cygwin? Another header under
netinet/, "/usr/include/netinet/ip_icmp.h", just #include's this empty
file. The fact that another header does this with an *empty* file seems
decidedly bizarre to me.

I'd like to have fping ported to cygwin. I have written a tool that
finds the best CPAN mirrors for Perl users/admins to connect to,
systematically, and that script calls the external program fping. it's
working great on Debian but cannot on cygwin, where fping isn't
available.

Regards.

-- 
See my OpenPGP key at https://savannah.gnu.org/people/viewgpg.php?user_id=6050
GnuPG public key fingerprint  | "Only when efforts to reform society have as
 BD26 A5D8 D781 C96B 9936 |  their point of departure the reformation of
 310F 0573 A3D9 4E24 4EA6 |  the inner life -- human revolution -- will
they lead us with certainty to a world of lasting peace and true human security."
-- Daisaku Ikeda

--
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: colons with rsync

[Ben Smith]

Igor Pechtchanski wrote:

> Windows filenames cannot contain ':', so there is no way around this
> limitation in Cygwin, since it uses Windows filenames.

Igor,

So does this mean that Cygwin doesn't do any file munging?  Is this
something that applications are normally left to deal with?

I found this post that mentions how samba deals with similar problems:
http://www.mail-archive.com/[EMAIL PROTECTED]/msg05701.html

I know that cygwin can't just write a file called c:\foo:bar.txt but
shouldn't there be some sort of translation to be able to deal with this
situation, or is it contrary to the design philosophy of cygwin in some
way?

-Ben
[EMAIL PROTECTED]
[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: colons with rsync

[Igor Pechtchanski]

On 11 Jun 2003, Ben Smith wrote:

> Igor Pechtchanski wrote:
>
> > Windows filenames cannot contain ':', so there is no way around this
> > limitation in Cygwin, since it uses Windows filenames.
>
> Igor,
>
> So does this mean that Cygwin doesn't do any file munging?  Is this
> something that applications are normally left to deal with?
>
> I found this post that mentions how samba deals with similar problems:
> http://www.mail-archive.com/[EMAIL PROTECTED]/msg05701.html
>
> I know that cygwin can't just write a file called c:\foo:bar.txt but
> shouldn't there be some sort of translation to be able to deal with this
> situation, or is it contrary to the design philosophy of cygwin in some
> way?
>
> -Ben

Ben,

Please make sure your mailer doesn't ignore Reply-To: (which I did set).

As for having a filename encoding for Cygwin, this issue has been
discussed on this list in the past (last year, IIRC).  Many variants have
been suggested, including RFC 1738 encoding.  People were generally
opposed to this sort of change, mostly because these files would then
become non-transparent with respect to Windows tools (e.g. explorer).  You
might wish to review those threads.  In fact,
 might be relevant to
your problem...
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: latest version of vi messes up bash for me

[Stephen Biggs]

On 10 Jun 2003 at 16:56, Larry Hall wrote:

> Stephen Biggs wrote:
> > On 9 Jun 2003 at 21:16, Pierre A. Humblet wrote:
> > 
> > 
> >>On Tue, Jun 10, 2003 at 03:57:32AM +0200, Stephen Biggs wrote:
> >>
> >>
> >> 
> >>
> >>>Maybe it is relevant to say that I am invoking cygwin by the shortcut of:
> >>>C:\WINDOWS\COMMAND.COM /E:4096 /c C:\cygwin\bin\login steve
> >>>with the /etc/passwd file entries:
> >>>user::400:401:user:/home/user:/bin/bash
> >>>steve::502:100:steve:/home/steve:/bin/bash
> >>>sshd::545:545:sshd::/bin/false
> >>>
> >>>and /etc/group:
> >>>all::544:
> >>>users::100:
> >>>user::401:
> >>>
> >>>... instead of the normal batch file since this is the only way I could 
> >>>think of to log myself on as another user into Cygwin on Windows 98.
> >>>
> >>>Perhaps this is what is messing me up, but can anybody tell me a better 
> >>>way to 
> >>>do it?  I need the user steve so I can use ssh to a different machine.  
> >>
> >>Your passwd file is screwed up. Didn't cygwin generate one for you?
> >>
> >>If you login into Windows, use the Users control panel, create a user
> >>Steve. Log into Windows. Start Cygwin. Cygwin will use the entry for Steve.
> >>
> >>If you don't log into Windows, Cygwin looks for uid 500. So just put
> >>Steve:*:500:544:steve:/home/steve:/bin/bash
> > 
> > 
> > It didn't work with the '*', so I took it out...
> > 
> > 
> >>Don't use uid 400, gid 401, they are reserved for non-existent entries.
> > 
> > 
> > Got it...
> > 
> > 
> >>>Again, 
> >>>this all worked before I updated my VI package.
> >>
> >>That's another story. No idea. Use emacs? :)
> > 
> > 
> > Heh...
> > 
> > Still no change with the blind terminal after VI.
> 
> It really would be helpful to see the output of cygcheck -r -s -v
> (_attached_).  Can't you just start the bash prompt from cygwin.bat
> at least to get this information?  Also, it would be useful to know
> if vim works for you if you login to Windows and use cygwin.bat
> instead.
> 
> 
-- signature snipped --

After I created the user ID 'steve' in Windows 98 and then logged on to 
it, Vim works flawlessly now.  It seems to be some sort of ID screw-up 
(proved by the lockup with cygcheck) if your Windows user name is not the 
same as your cygwin name.  I can see how this can be a problem on a 
system like NT/2K/XP, but Windows 98 is inherently insecure, anyway... oh 
well... Windows 98 is going away soon, anyway, but there are still a lot 
of people using it, especially out of the US.

I do have this problem that Windows 98 now shows a menu at start-up 
asking for which user id to logon with, even though there is only one id, 
that with no password. Anybody know how to get the old version of Windows 
98 (NOT the SE version!) to logon automatically?  It doesn't show the 
login dialog like later versions of Win98.

--
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: colons with rsync

[Ben Smith]

Thanks for the link.  I can see now that this is a hot topic.

http://cygwin.com/ml/cygwin/2002-09/msg00858.html

It seems to me that, with respect to the colon issue at least, it is
being actively ignored in order to maintain the "File Streams"
functionality of NTFS in Cygwin.  The question is "Why?"  By all
accounts, "Alternate File Streams" is little known, little used,
possibly problematic due to unreported filespace usage, and unexpected
within Cygwin.  All this seems to achieve is to break lots of normal
Unix applications under Cygwin.

In this reply, you said:

http://sources.redhat.com/ml/cygwin/2003-06/msg00317.html

"In all other details (including restricted characters in filenames),
Cygwin uses the underlying filesystem's conventions.  If we go out of
our way to be compatible with Linux in this aspect, why not also support
"aux" as the filename, or support '\' in filenames?  The argument for
not doing the latter was that we don't want to disassociate ourselves
from the underlying filesystem, IIRC, so why go back on it now?"

I would ask (because I really want to know):

What's the difference?  

Scenario 1:  Cygwin follows all Windows filesystem conventions and
ignores Unix features that are not present on Windows.  Cygwin is
Windows on Unix on Windows, which is completely useless since NO Unix
apps use functionality that is specific to Windows and most Unix apps
require functionality that is not supported in Windows.  Applications
have to be rewritten to work around limitations of Cygwin/Windows.

Scenario 2:  Cygwin tries to follow all Unix filesystem conventions and
map them to Windows.  Nothing is lost since the files that were
accessible under Windows through Scenario 1 are still accessible. 
Functionality is gained since Cygwin can work around a lot of Windows
limitations and act as an emulation layer for Unix apps.

Also, since no one seems to be interested in making this change to
Cygwin itself, is there really a technical reason that rsync can't be
adjusted to work around it?  I don't think the current behavior achieves
anything other than strife.

-Ben


--
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: Error message when compling simple program, help please :)

[Tomasz Rojek]

> I am using the following command:
>
> $ gcc 02l01.c
> /usr/lib/gcc-lib/i686-pc-cygwin/3.2/../../../../i686-pc-cygwin/bin/ld:
> cannot find -luser32
> collect2: ld returned 1 exit status
Hi Max,

I would advise you to go thru the archive. There you can find a simplest syntax how to
compile a C program using gcc on Cygwin. In shorter way - below syntax creates desired
binary (at least for me :)

gcc 02l01.c -o 02l01.exe

so you have to specify output binary file and that's all ;-)
Greetings

-- 
Tomasz Rojek




--
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: colons with rsync

[Christopher Faylor]

On Wed, Jun 11, 2003 at 03:36:03PM -0500, Ben Smith wrote:
>Also, since no one seems to be interested in making this change to
>Cygwin itself, is there really a technical reason that rsync can't be
>adjusted to work around it?  I don't think the current behavior achieves
>anything other than strife.

I think you missed two key points: 1) The best argument is code.  If you
think that having cygwin support colons is a great idea, demonstrate
your idea with working code.  2) We're a mean bunch.  We thrive on
strife.

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: Problem in shell scripting - possible memory leak / violation?

[Tomasz Rojek]

I should have included this info in my initial post:

$ uname -a
CYGWIN_NT-5.1 waw0968b 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin

Greetings

-- 
Tomasz Rojek


begin 666 cygcheck output.txt
M)"!C>6=C:&5C:R M8PI#>6=W:[EMAIL PROTECTED]&%C:V%G92!);F9O7!T(" @(" @(" @(" @(" @(" @(#$N,"TQ"F-T86=S(" @(" @(" @(" @
M(" @(" @(#4N-2TT"F-Y9W)U;G-R=B @(" @(" @(" @(" @(# N.38M,0IC
M>6=U=&EL" @(" @(" @(" @(" @(" @(" @,BXU+C0M,@IF=G=M(" @(" @(" @(" @
M(" @(" @(" R+C0N-RTR"F=A=VL@(" @(" @(" @(" @(" @(" @(#,N,2XR
M+3(*9V-C(" @(" @(" @(" @(" @(" @(" @,RXR+3,*9V-C+6UI;F=W(" @
M(" @(" @(" @(" @,C P,C X,3FEP(" @(" @(" @(" @(" @
M(" @(" Q+C,N,RTT"FEN971U=&EL&UL,B @(" @(" @(" @(" @(" @,BXU
M+C" @(" @(" @(" @(" @(" @(" @
M,BXX+C0M-0IM-" @(" @(" @(" @(" @(" @(" @(" Q+C0M,0IM86ME(" @
M(" @(" @(" @(" @(" @(" S+C7-V:6YI=" @(" @(" @(" @(" @(" [EMAIL PROTECTED],*
M=&%R(" @(" @(" @(" @(" @(" @(" @,2XQ,RXR-2TQ"G1C;'1K(" @(" @
M(" @(" @(" @(" @(#(P,#,P,C$T+3$*=&-P7W=R87!P97)S(" @(" @(" @
M(" @-RXV+3$*=&5R;6-A<" @(" @(" @(" @(" @(" @,C P,C Y,S M,0IT
M97)M:6YF;R @(" @(" @(" @(" @(" U+C,M,@IT97AI;F9O(" @(" @(" @
M(" @(" @(" T+C(M- IT97AT=71I;',@(" @(" @(" @(" @(" R+C N,C$M
M,0IT:69F(" @(" @(" @(" @(" @(" @(" S+C8N,"TQ"G1I;64@(" @(" @
M(" @(" @(" @(" @(#$N-RTQ"G5N:71S(" @(" @(" @(" @(" @(" @(#$N
M-SFEP(" @
M(" @(" @(" @(" @(" @(" @,BXS+3(*>FQI8B @(" @(" @(" @(" @(" @
M(" @,2XQ+C0M,0H*57-E("UH('1O('-E92!H96QP(&%B;W5T(&5A8V@@http://cygwin.com/ml/#unsubscribe-simple
Problem reports:   http://cygwin.com/problems.html
Documentation: http://cygwin.com/docs.html
FAQ:   http://cygwin.com/faq/

':' - was rsync..

[Robert Collins]

Sorry for not replying in thread - I deleted the email a little fast :p.

Anyway, ':' is -not- in the POSIX portable filename character set:
http://www.opengroup.org/onlinepubs/007908799/xbd/glossary.html#tag_004_000_
207

':' as a special behaviour is not a Windows invention, it harks (in windows'
case) from the VMS roots, and perhaps further back still.

Secondly, solutions to handle ':' -will- involve not writing the ':' to
disk, which means parsing and interpreting the filename for validity on
every open, and during opendir etc. It's overhead - do we need it...

Rob




--
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: ':' - was rsync..

[Christopher Faylor]

On Thu, Jun 12, 2003 at 07:18:12AM +1000, Robert Collins wrote:
>Sorry for not replying in thread - I deleted the email a little fast
>:p.
>
>Anyway, ':' is -not- in the POSIX portable filename character set:
>http://www.opengroup.org/onlinepubs/007908799/xbd/glossary.html#tag_004_000_
>207
>
>':' as a special behaviour is not a Windows invention, it harks (in
>windows' case) from the VMS roots, and perhaps further back still.
>
>Secondly, solutions to handle ':' -will- involve not writing the ':' to
>disk, which means parsing and interpreting the filename for validity on
>every open, and during opendir etc.  It's overhead - do we need it...

I'm working on a per-mount option that would allow UNIX-like flexibility
in filename choice.  It's something tha Red Hat thinks it vaguely needs
so I thought I'd do it in my spare time when I'm not being driven crazy
by management issues and politics.

For the record, I am nowhere near a working implementation right now and
I don't have any ETA.

cgf
--
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

--
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: Processes are unable to fully discard their controlling terminal

[Matt]

works like a champ.. thanks for the pointer.  you've
never tried to get cygwin to use pageant have you (or
anybody)?? pageant = putty's ssh agent
--- Max Bowsher <[EMAIL PROTECTED]> wrote:
> Some time ago I (Max Bowsher) wrote:
> > I can now successfully share an ssh-agent between
> all
> > my shells, with it
> > starting with the first one, and ending with the
> last,
> > and no zombie windows
> > if I end shells in the wrong order.
> 
> Matt wrote:
> > I would like ssh-agent to function like you say it
> is
> > for you... but i can't understand what you did
> besides
> > obtain the latest versions which i should have.. 
> what
> > else do i need to do?
> 
> Some shell scripting.
> 
> Attached are the shell scripts I *source* from my
> .bash_profile and
> .bash_logout.
> 
> It's not a perfect system (if a shell is killed in
> such a way that
> .bash_logout doesn't execute, an agent process can
> remain), but it seems to
> work well for the most part.
> 
> 
> 
> Max.
> 

> ATTACHMENT part 2 application/octet-stream
name=agent_refcount_start


> ATTACHMENT part 3 application/octet-stream
name=agent_refcount_stop
> --
> 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/


__
Do you Yahoo!?
Yahoo! Calendar - Free online calendar with sync to Outlook(TM).
http://calendar.yahoo.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/

Re: Processes are unable to fully discard their controlling terminal

[Max Bowsher]

Matt wrote:
> works like a champ.. thanks for the pointer.  you've
> never tried to get cygwin to use pageant have you (or
> anybody)?? pageant = putty's ssh agent

No, and I'm pretty sure that it is impossible.


Max.


--
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: latest version of vi messes up bash for me

[Larry Hall]

Stephen Biggs wrote:
After I created the user ID 'steve' in Windows 98 and then logged on to 
it, Vim works flawlessly now.  It seems to be some sort of ID screw-up 
(proved by the lockup with cygcheck) if your Windows user name is not the 
same as your cygwin name.  I can see how this can be a problem on a 
system like NT/2K/XP, but Windows 98 is inherently insecure, anyway... oh 
well... Windows 98 is going away soon, anyway, but there are still a lot 
of people using it, especially out of the US.


Right.  But Cygwin isn't going away and it's utilities come from
environments where security isn't disabled.  Now I'm not saying
that the behavior you saw is optimal or even desirable.  Might even
be a bug. ;-)  But the way you're using it now is the way it's designed
to work regardless of the underlying O/S capabilities.  So working this
way is the path of least resistance.  Of course, you can always debug
it if you'd like to work another way and it doesn't work for you. ;-)

I do have this problem that Windows 98 now shows a menu at start-up 
asking for which user id to logon with, even though there is only one id, 
that with no password. Anybody know how to get the old version of Windows 
98 (NOT the SE version!) to logon automatically?  It doesn't show the 
login dialog like later versions of Win98.


Sorry, I don't use 9x here.  It offends my sensabilities (more than
other Windows O/Ss).  But I seem to recall that there were options to
log in by default as a particular user.  I know NT/W2K has this.
Check out tweakui or X-Setup.  I believe they both have the ability
to set a particular automatic login.
BTW, if you respond to this message, note that the reply-to has been
set to this list.  You don't need to include "my" email address as well.
--
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/

patch in cygutil/lpr

[gilles civario]

Hello.

I've found that the new version of lpr provided with cygutils-1.1.4 no longer
works with my PostScript network printer. With the previous version of cygutils
(1.1.3), I had to change the datatype provided to StartDocPrinter from "raw"
to "RAW". This time, the trouble I've encounted wath this one :
$ ./lpr.exe lpr.1
lpr: printer error: StartDocPrinter error: Invalide descriptor
$ uname -a
CYGWIN_NT-5.0 hepatique 1.3.22(0.78/3/2) 2003-03-18 09:20 i686 unknown unknown Cygwin
After some searches, I've found a solution by changing the DesidedAccess of the
PRINTER_DEFAULTS argument in OpenPrinter, from PRINTER_ALL_ACCESS to 
PRINTER_ACCESS_USE.
$ diff -urN Printer.cc.orig Printer.cc
--- Printer.cc.orig 2003-06-11 05:01:58.0 +0200
+++ Printer.cc  2003-06-12 07:10:06.0 +0200
@@ -281,7 +281,7 @@
   prDef.pDatatype = "RAW";
   prDef.pDevMode = m_devMode;
-  prDef.DesiredAccess = PRINTER_ALL_ACCESS;
+  prDef.DesiredAccess = PRINTER_ACCESS_USE;
   if (m_devHandle == INVALID_HANDLE_VALUE &&
   !OpenPrinter(tstr(m_devName), &m_devHandle, &prDef))
And now, it works perfectly for me.

Regards.

Gilles Civario.



--
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: port query: any "fping"? & cygwin headers

[Gerrit P. Haase]

Hallo Soren,

>   /usr/include/cygwin/icmp.h

> that is an empty file. Is this a TODO for Cygwin? Another header under
> netinet/, "/usr/include/netinet/ip_icmp.h", just #include's this empty
> file. The fact that another header does this with an *empty* file seems
> decidedly bizarre to me.


There is an inetutils release 1.4.2 which includes libicmp, probably
you'll need to port this first:
http://fresh.t-systems-sfr.com/unix/src/misc/.warix/inetutils-1.4.2.tar.gz.html


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/