cmp (or echo) bug?

cmp (or echo) bug?

[David Balažic]

Hi!

In Cygwin terminal (bash) I typed:

cmp  <(echo echo1)  <(echo echo2)

This does not print anything.
Not even with -b.

On Linux (Ubuntu 12.04 in VMWare) it reports that the inputs are different.

Bug?
Or am I missing something?

I just updated to latest version of cygwin, to be sure.
32 bit version.
On Windows 8.1 Pro x64 (all updates)

Regards,
David

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

Re: cmp (or echo) bug?

[Marco Atzeri]

On 25/12/2015 22:32, David Balažic wrote:
> Hi!
>
> In Cygwin terminal (bash) I typed:
>
> cmp  <(echo echo1)  <(echo echo2)

only on 32bit, on 64 I see

$ cmp  <(echo echo1)  <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1

I suspect a timing issue, and the result is not well defined


> This does not print anything.
> Not even with -b.
>
> On Linux (Ubuntu 12.04 in VMWare) it reports that the inputs are different.
>
> Bug?
> Or am I missing something?
>
> I just updated to latest version of cygwin, to be sure.
> 32 bit version.
> On Windows 8.1 Pro x64 (all updates)
>
> Regards,
> David
>

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

Re: cmp (or echo) bug?

[Ismail Donmez]

Hi,

David Balažic <xerces9 <at> gmail.com> writes:

> In Cygwin terminal (bash) I typed:
> 
> cmp  <(echo echo1)  <(echo echo2)

I suspect its a bash bug since it works fine with zsh (tested 64bit only).

Re: cmp (or echo) bug?

[Helmut Karlowski]

---------------------------------------------------
> Hi,
> 
> David Balažic <xerces9 <at> gmail.com> writes:
> 
> > In Cygwin terminal (bash) I typed:
> > 
> > cmp  <(echo echo1)  <(echo echo2)
> 
> I suspect its a bash bug since it works fine with zsh (tested 64bit only).

zsh uses temp-files for process-substitution, while bash seems to use 
pipes which are not very good seekable. That might cause the 
difference. It only works in zsh and no other shell because of that.

Diff instead of cmp works also in bash btw.

-Helmut


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

Re: cmp (or echo) bug?

[David Balažic]

I tried it in zsh (32 bit cygwin) and there it works correctly:

$  cmp  <(echo echo1)  <(echo echo2)
/tmp/zshirbIJ1 /tmp/zshDsdZep differ: byte 5, line 1

So it seems the bug is in bash.

Regards,
David


On 26 December 2015 at 13:49, Ismail Donmez <ismail@i10z.com> wrote:
> Hi,
>
> David Balažic <xerces9 <at> gmail.com> writes:
>
>> In Cygwin terminal (bash) I typed:
>>
>> cmp  <(echo echo1)  <(echo echo2)
>
> I suspect its a bash bug since it works fine with zsh (tested 64bit only).

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

Re: cmp (or echo) bug?

[Michael Enright]

On Mon, Dec 28, 2015 at 9:08 AM, David Balažic wrote:
> I tried it in zsh (32 bit cygwin) and there it works correctly:
>
> $  cmp  <(echo echo1)  <(echo echo2)
> /tmp/zshirbIJ1 /tmp/zshDsdZep differ: byte 5, line 1
>
> So it seems the bug is in bash.
>
A different conclusion is also supportable: That the two pipe
mechanisms have different edge-case behavior, resulting in different
outcomes for the command depending on which pipe type is used..

I tried this on 32-bit cygwin, Windows 7/64:
$ cmp <(for i in 1 2 3 4 5; do echo echo$i; done) <(for i in 1 2 3 4
6; do echo echo$i; done)
/dev/fd/63 /dev/fd/62 differ: byte 29, line 5

Your output from zsh shows named pipes are in use. "man zshexpn" says
that zsh can use anonymous pipes, yet it chose not to in your case. In
my case bash chose to use anonymous pipes, even though "man bash" says
it may use named pipes.

The man pages for these shells describe essentially the same syntax
and mechanisms for this process substitution mechanism. What is not
defined:
1) How do commands such as my for loop command or your simple echo
command provide output properly so that EOF isn't detected spuriously
2) How do programs such as cmp or diff read from their input in such a
way that they are not fooled by a file status that might appear to be
EOF but isn't.
3) Crystalline clarity as to when the shell prefers one pipe type to
another, at least not to this reader, and not that it matters.

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

Re: cmp (or echo) bug?

[David Balažic]

It is irrelevant how it is implemented. The command means "compare the
output of those two commands" and the only correct result is "they
differ on byte 5".

It is clearly a bug.

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

Re: cmp (or echo) bug?

[Andrey Repin]

Greetings, David Balažic!

> It is irrelevant how it is implemented. The command means "compare the
> output of those two commands" and the only correct result is "they
> differ on byte 5".

> It is clearly a bug.

I wouldn't be so hastily jumping to conclusions, if I were you.


-- 
With best regards,
Andrey Repin
Tuesday, December 29, 2015 07:39:06

Sorry for my terrible english...

Re: cmp (or echo) bug?

[Houder]

On 2015-12-25 22:32, David Balažic wrote:
> Hi!
> 
> In Cygwin terminal (bash) I typed:
> 
> cmp  <(echo echo1)  <(echo echo2)
> 
> This does not print anything.
> Not even with -b.
> 
> On Linux (Ubuntu 12.04 in VMWare) it reports that the inputs are 
> different.
> 
> Bug?
> Or am I missing something?

@@ uname -a
CYGWIN_NT-6.1-WOW Seven 2.3.1(0.291/5/3) 2015-11-14 12:42 i686 Cygwin
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1
@@ ./cmp <(echo echo1) <(echo echo2)
/dev/fd/63 /dev/fd/62 differ: byte 5, line 1

etc.

But only after I had modified cmp.c (diffutils) as follows:

int
main (int argc, char **argv)
{
...

#if 0
       if (file_desc[f1] < 0 || fstat (file_desc[f1], stat_buf + f1) != 
0)
#else
       if (file_desc[f1] < 0 || f1 ? ( stat (file[1], stat_buf + f1) != 0 
)
                                   : ( stat (file[0], stat_buf + f1) != 0 
) ) // Henri: suspect fstat
#endif
         {
           if (file_desc[f1] < 0 && comparison_type == type_status)
             exit (EXIT_TROUBLE);
           else
             error (EXIT_TROUBLE, errno, "%s", file[f1]);
         }
     }

   /* If the files are links to the same inode and have the same file 
position,
      they are identical.  */

   if (0 < same_file (&stat_buf[0], &stat_buf[1])
       && same_file_attributes (&stat_buf[0], &stat_buf[1])
       && file_position (0) == file_position (1))
{ // Henri: diagnostics
#if 0
     printf("same_file = %d\n", same_file (&stat_buf[0], &stat_buf[1]) );
     printf("same_file_attributes = %d\n", same_file_attributes 
(&stat_buf[0], &stat_buf[1]) );
     printf("same file pos = %d\n", file_position (0) == file_position 
(1) );
#endif
     printf("file[0] = %s, file[1] = %s\n", file[0], file[1]);
     printf("file_desc[0] = %d, file_desc[1] = %d\n", file_desc[0], 
file_desc[1]);
     printf("bailing out: same file.\n");
     return EXIT_SUCCESS;
}

Regards,
Henri


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

Re: cmp (or echo) bug?

[David Balažic]

cat and diff work as expected:

$ cat  <(echo echo1)  <(echo echo2)
echo1
echo2


$ diff  <(echo echo1)  <(echo echo2)
1c1
< echo1
---
> echo2


So maybe the bug is really in cmp.

Regards,
David


On 29 December 2015 at 17:43, Houder <houder@xs4all.nl> wrote:
> On 2015-12-25 22:32, David Balažic wrote:
>>
>> Hi!
>>
>> In Cygwin terminal (bash) I typed:
>>
>> cmp  <(echo echo1)  <(echo echo2)
>>
>> This does not print anything.
>> Not even with -b.
>>
>> On Linux (Ubuntu 12.04 in VMWare) it reports that the inputs are
>> different.
>>
>> Bug?
>> Or am I missing something?
>
>
> @@ uname -a
> CYGWIN_NT-6.1-WOW Seven 2.3.1(0.291/5/3) 2015-11-14 12:42 i686 Cygwin
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
> @@ ./cmp <(echo echo1) <(echo echo2)
> /dev/fd/63 /dev/fd/62 differ: byte 5, line 1
>
> etc.
>
> But only after I had modified cmp.c (diffutils) as follows:
>
> int
> main (int argc, char **argv)
> {
> ...
>
> #if 0
>       if (file_desc[f1] < 0 || fstat (file_desc[f1], stat_buf + f1) != 0)
> #else
>       if (file_desc[f1] < 0 || f1 ? ( stat (file[1], stat_buf + f1) != 0 )
>                                   : ( stat (file[0], stat_buf + f1) != 0 ) )
> // Henri: suspect fstat
> #endif
>         {
>           if (file_desc[f1] < 0 && comparison_type == type_status)
>             exit (EXIT_TROUBLE);
>           else
>             error (EXIT_TROUBLE, errno, "%s", file[f1]);
>         }
>     }
>
>   /* If the files are links to the same inode and have the same file
> position,
>      they are identical.  */
>
>   if (0 < same_file (&stat_buf[0], &stat_buf[1])
>       && same_file_attributes (&stat_buf[0], &stat_buf[1])
>       && file_position (0) == file_position (1))
> { // Henri: diagnostics
> #if 0
>     printf("same_file = %d\n", same_file (&stat_buf[0], &stat_buf[1]) );
>     printf("same_file_attributes = %d\n", same_file_attributes
> (&stat_buf[0], &stat_buf[1]) );
>     printf("same file pos = %d\n", file_position (0) == file_position (1) );
> #endif
>     printf("file[0] = %s, file[1] = %s\n", file[0], file[1]);
>     printf("file_desc[0] = %d, file_desc[1] = %d\n", file_desc[0],
> file_desc[1]);
>     printf("bailing out: same file.\n");
>     return EXIT_SUCCESS;
> }
>
> Regards,
> Henri
>
>
>
> --
> Problem reports:       http://cygwin.com/problems.html
> FAQ:                   http://cygwin.com/faq/
> Documentation:         http://cygwin.com/docs.html
> Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
>

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