Re: Who "fixed" md5sum??

Re: Who "fixed" md5sum??

[Christopher Faylor]

On Sat, Nov 17, 2001 at 03:53:53PM -0700, Mark Paulus wrote:
>I have a question about md5sum.  It seems like in one of the not
>too distant releases of textutils, the md5sum program altered it's
>behaviour.  Before the change, it would act like the debian/linux
>variety, in that if you would do an 
>md5sum < <somefile>
>you would get a result, and that would be all, e.g.:
>
>This is the result from my debian box.....
>
>mgpaulus@testbox1:/smbapps/cygwin.pkg$ md5sum < setup.exe
>d91a97dcd67b7cb9c6aaa408cf431b01
>
>Here's the same result, except from my cygwin environment:
>$ md5sum < setup.exe
>d91a97dcd67b7cb9c6aaa408cf431b01 *-
>
>My problem is the little '*-' at the end...  I have several unix ported scripts 
>that run md5sum, and expect only the significant md5sum digits, and the
>trailing stuff is causing all the scripts to break.  Would it be possible
>to get either 
>1) a rational for adding the trailing stuff, and making it act
>differently than it's un*x cousin; 
>
>and/or 
>
>2) an ability to turn that "feature" off,
>if everyone else absolutely requires it??

md5sum on my linux box returns a '-' in the second column if the md5sum
is being performed on standard input.  I assume that you are using an
older version on your linux system if yours doesn't do that.

Inspecting the source, I see that the '*' means that the file is binary.

As to why these changes were introduced, you'd have to ask the authors
of textutils.

cgf

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Re: Who "fixed" md5sum??

[Richard Hitt]

Hi.

On my incredibly old version of cygnus, md5sum outputs two tokens.  So I
guess you have to blame debian:  perhaps they've gone standard.

BASH.EXE-2.01$ type md5sum
md5sum is hashed (/cygnus/b19/H-i386-cygwin32/bin/md5sum)
BASH.EXE-2.01$ ls -l /cygnus/b19/H-i386-cygwin32/bin/md5sum
-rwxr-xr-x   1 rbh00    everyone    27648 Feb 26  1998
/cygnus/b19/H-i386-cygwin
32/bin/md5sum
BASH.EXE-2.01$ cat ACDS* | md5sum
9dd1a462e2a0dc110c19559ae5f85832  -
BASH.EXE-2.01$ md5sum < ACDS*
9dd1a462e2a0dc110c19559ae5f85832  -
BASH.EXE-2.01$

Richard


On Sat, 17 Nov 2001 15:53:53 -0700, you wrote:

>Hi,
>
>I have a question about md5sum.  It seems like in one of the not
>too distant releases of textutils, the md5sum program altered it's
>behaviour.  Before the change, it would act like the debian/linux
>variety, in that if you would do an 
>md5sum < <somefile>
>you would get a result, and that would be all, e.g.:
>
>This is the result from my debian box.....
>
>mgpaulus@testbox1:/smbapps/cygwin.pkg$ md5sum < setup.exe
>d91a97dcd67b7cb9c6aaa408cf431b01
>
>Here's the same result, except from my cygwin environment:
>$ md5sum < setup.exe
>d91a97dcd67b7cb9c6aaa408cf431b01 *-
>
>My problem is the little '*-' at the end...  I have several unix ported scripts 
>that run md5sum, and expect only the significant md5sum digits, and the
>trailing stuff is causing all the scripts to break.  Would it be possible
>to get either 
>1) a rational for adding the trailing stuff, and making it act
>differently than it's un*x cousin; 
>
>and/or 
>
>2) an ability to turn that "feature" off,
>if everyone else absolutely requires it??
>
>Thanks....


--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/

Who "fixed" md5sum??

[Mark Paulus]

Hi,

I have a question about md5sum.  It seems like in one of the not
too distant releases of textutils, the md5sum program altered it's
behaviour.  Before the change, it would act like the debian/linux
variety, in that if you would do an 
md5sum < <somefile>
you would get a result, and that would be all, e.g.:

This is the result from my debian box.....

mgpaulus@testbox1:/smbapps/cygwin.pkg$ md5sum < setup.exe
d91a97dcd67b7cb9c6aaa408cf431b01

Here's the same result, except from my cygwin environment:
$ md5sum < setup.exe
d91a97dcd67b7cb9c6aaa408cf431b01 *-

My problem is the little '*-' at the end...  I have several unix ported scripts 
that run md5sum, and expect only the significant md5sum digits, and the
trailing stuff is causing all the scripts to break.  Would it be possible
to get either 
1) a rational for adding the trailing stuff, and making it act
differently than it's un*x cousin; 

and/or 

2) an ability to turn that "feature" off,
if everyone else absolutely requires it??

Thanks....



--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Bug reporting:         http://cygwin.com/bugs.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/