Re: different behaviour of bash and sh

Re: different behaviour of bash and sh

[Chet Ramey]

> Machine Type: i686-pc-cygwin
> 
> Bash Version: 2.04
> Patch Level: 7
> Release Status: release
> 
> Description:
> The following script behaves differently in bash and sh:
> 
>     % cat script
>     #!/bin/sh --
>     checkstat() {
> 	echo $?
>     }
>     false
>     checkstat
>     # End of file
> 
> Repeat-By:
> 
>     $ type bash
>     bash is /usr/bin/bash
>     $ bash script
>     1
>     $ type sh
>     sh is /usr/bin/sh
>     $ sh script
>     0
> 
> The 0 is wrong IMHO, it should be 1.  (Even HP with their BD-/bin/sh
> returns 1).

You're reporting a bug in the cygwin sh, not bash.

-- 
``The lyf so short, the craft so long to lerne.'' - Chaucer
( ``Discere est Dolere'' -- chet)

Chet Ramey, CWRU    chet@po.CWRU.Edu    http://cnswww.cns.cwru.edu/~chet/

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Ralf Fassel]

* Andrej Borsenkow
| So what? The question is, if the exibited behaviour of *sh* is
| correct; the question was not "if sh is bash".

Earnie wanted to point out that my report was inappropriate at the
bash-bug mailinglist (which I had included since on my Linux systems
/bin/sh is a link to /bin/bash, so I just assumed the two being
identical on Cygwin too, which is not the case end of long sentence)

| IMHO that is a bug in sh (or ash if you insist). The 'echo $?' is
| the next command executed after 'false' - so , there is nothing that
| can change the value of '$?' in between.

For completeness sake: On my SUSE 7.1 System, `ash', `sh' and `bash'
all echo `1', so I think this is really a problem in the cygwin port
of `ash' which gets installed as /bin/sh.

I just had a quick glance at the ash source, but I'm in for the
weekend now, so no patches :-)

R'

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Andrej Borsenkow]

On Fri, 9 Mar 2001, Earnie Boyd wrote:

> In the Cygwin release
>
> sh != bash
> sh == ash
>


So what? The question is, if the exibited behaviour of *sh* is correct;
the question was not "if sh is bash".

IMHO that is a bug in sh (or ash if you insist). The 'echo $?' is the next
command executed after 'false' - so , there is nothing that can change the
value of '$?' in between.

-andrej

> Earnie.
>
> Ralf Fassel wrote:
> >
> > Configuration Information [Automatically generated, do not change]:
> > Machine: i686
> > OS: cygwin
> > Compiler: gcc
> > Compilation CFLAGS:  -DPROGRAM='bash.exe' -DCONF_HOSTTYPE='i686' -DCONF_OSTYPE='cygwin' -DCONF_MACHTYPE='i686-pc-cygwin' -DCONF_VENDOR='pc' -DSHELL -DHAVE_CONFIG_H   -I.  -I../src -I../src/include -I../src/lib -I/src/bash/distro/usr/include -g -O2
> > uname output: CYGWIN_NT-4.0 MARS 1.1.8(0.34/3/2) 2001-01-31 10:08 i686 unknown
> > Machine Type: i686-pc-cygwin
> >
> > Bash Version: 2.04
> > Patch Level: 7
> > Release Status: release
> >
> > Description:
> > The following script behaves differently in bash and sh:
> >
> >     % cat script
> >     #!/bin/sh --
> >     checkstat() {
> >         echo $?
> >     }
> >     false
> >     checkstat
> >     # End of file
> >
> > Repeat-By:
> >
> >     $ type bash
> >     bash is /usr/bin/bash
> >     $ bash script
> >     1
> >     $ type sh
> >     sh is /usr/bin/sh
> >     $ sh script
> >     0
> >
> > The 0 is wrong IMHO, it should be 1.  (Even HP with their BD-/bin/sh
> > returns 1).
> >
> > R'
> >
> > --
> > Want to unsubscribe from this list?
> > Check out: http://cygwin.com/ml/#unsubscribe-simple
>
> _________________________________________________________
> Do You Yahoo!?
> Get your free @yahoo.com address at http://mail.yahoo.com
>
>
> --
> Want to unsubscribe from this list?
> Check out: http://cygwin.com/ml/#unsubscribe-simple
>
>

-- 
------------------------------------------------------------------------
Andrej Borsenkow 		Fax:   +7 (095) 737 2747
Siemens ICP ITS Moscow		Tel:   +7 (095) 737 2723

E-Mail: Andrej.Borsenkow@mow.siemens.ru
-------------------------------------------------------------------------


--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Ralf Fassel]

* Earnie Boyd
| I was pointing your fallacy especially to the bash-bugs list.

Ok, point taken.  True, I wasn't aware that sh != bash in cygwin.

| If you feel that ash is incorrect then please, download the source,
| fix it, submit a patch to cygwin-apps@cygwin.com for consideration.

Business as usual :-)

R'

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Earnie Boyd]

Ralf Fassel wrote:
> 
> * Earnie Boyd
> | In the Cygwin release
> |
> | sh != bash
> | sh == ash
> 
> And `ash' is supposed to return 0 for $? of `false'?
> 

I wasn't saying ash does or doesn't have a bug; I was pointing your
fallacy especially to the bash-bugs list.  If you feel that ash is
incorrect then please, download the source, fix it, submit a patch to
cygwin-apps@cygwin.com for consideration.

Earnie.

_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Ralf Fassel]

* Earnie Boyd
| In the Cygwin release
| 
| sh != bash
| sh == ash

And `ash' is supposed to return 0 for $? of `false'?

R'

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

Re: different behaviour of bash and sh

[Earnie Boyd]

In the Cygwin release

sh != bash
sh == ash

Earnie.

Ralf Fassel wrote:
> 
> Configuration Information [Automatically generated, do not change]:
> Machine: i686
> OS: cygwin
> Compiler: gcc
> Compilation CFLAGS:  -DPROGRAM='bash.exe' -DCONF_HOSTTYPE='i686' -DCONF_OSTYPE='cygwin' -DCONF_MACHTYPE='i686-pc-cygwin' -DCONF_VENDOR='pc' -DSHELL -DHAVE_CONFIG_H   -I.  -I../src -I../src/include -I../src/lib -I/src/bash/distro/usr/include -g -O2
> uname output: CYGWIN_NT-4.0 MARS 1.1.8(0.34/3/2) 2001-01-31 10:08 i686 unknown
> Machine Type: i686-pc-cygwin
> 
> Bash Version: 2.04
> Patch Level: 7
> Release Status: release
> 
> Description:
> The following script behaves differently in bash and sh:
> 
>     % cat script
>     #!/bin/sh --
>     checkstat() {
>         echo $?
>     }
>     false
>     checkstat
>     # End of file
> 
> Repeat-By:
> 
>     $ type bash
>     bash is /usr/bin/bash
>     $ bash script
>     1
>     $ type sh
>     sh is /usr/bin/sh
>     $ sh script
>     0
> 
> The 0 is wrong IMHO, it should be 1.  (Even HP with their BD-/bin/sh
> returns 1).
> 
> R'
> 
> --
> Want to unsubscribe from this list?
> Check out: http://cygwin.com/ml/#unsubscribe-simple

_________________________________________________________
Do You Yahoo!?
Get your free @yahoo.com address at http://mail.yahoo.com


--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple

different behaviour of bash and sh

[Ralf Fassel]

Configuration Information [Automatically generated, do not change]:
Machine: i686
OS: cygwin
Compiler: gcc
Compilation CFLAGS:  -DPROGRAM='bash.exe' -DCONF_HOSTTYPE='i686' -DCONF_OSTYPE='cygwin' -DCONF_MACHTYPE='i686-pc-cygwin' -DCONF_VENDOR='pc' -DSHELL -DHAVE_CONFIG_H   -I.  -I../src -I../src/include -I../src/lib -I/src/bash/distro/usr/include -g -O2
uname output: CYGWIN_NT-4.0 MARS 1.1.8(0.34/3/2) 2001-01-31 10:08 i686 unknown
Machine Type: i686-pc-cygwin

Bash Version: 2.04
Patch Level: 7
Release Status: release

Description:
The following script behaves differently in bash and sh:

    % cat script
    #!/bin/sh --
    checkstat() {
	echo $?
    }
    false
    checkstat
    # End of file

Repeat-By:

    $ type bash
    bash is /usr/bin/bash
    $ bash script
    1
    $ type sh
    sh is /usr/bin/sh
    $ sh script
    0

The 0 is wrong IMHO, it should be 1.  (Even HP with their BD-/bin/sh
returns 1).

R'

--
Want to unsubscribe from this list?
Check out: http://cygwin.com/ml/#unsubscribe-simple