[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Failed to checkin an icon file (.ico) in RCS
From: |
Aaron S. Hawley |
Subject: |
Re: Failed to checkin an icon file (.ico) in RCS |
Date: |
Wed, 10 Nov 2004 10:48:09 -0500 (EST) |
Here's a possible answer to your question available on the Web. The URL
is below:
I'm hitting the binary save problem described [...]
$ diffutils-2.7/diff /bin/ls /bin/cat ; echo $?
Binary files /bin/ls and /bin/cat differ
1
$ diffutils-2.8.1/src/diff /bin/ls /bin/cat ; echo $?
Binary files /bin/ls and /bin/cat differ
2
[...]
Answer
Ok, I figured out the problem. If you compile RCS (using the default
Solaris diff), and then recompile to use the GNU diff, without running
"configure" again, the compile options from configure are set based on
Solaris diff, not GNU diff. In particular, the DIFFFLAGS are -n, not
-an; the -a "treat all files as text" flag is required for RCS to work with
binary files. (Maybe testenv could be enhanced to try doing ci on two
different binary files to make sure the rcs/diff code works correctly?)
-- KenShirriff - 08 Aug 2003
full:
<http://twiki.sourceforge.net/cgi-bin/view/Support/BinarySaveFailsRCSConfigError>
On Wed, 10 Nov 2004 address@hidden wrote:
> Hi,
>
> While working with RCS, I found a problem where by I was not able to check
> in an icon file (.ico).
> Please find the details below -
>
> ****************************
> Running -> ci lmicon.ico
>
> Using configuration label DEV
>
> /usr/local/bin/rcsdiff -r1.2 lmicon.ico
> ===================================================================
> RCS file: RCS/lmicon.ico,v
> retrieving revision 1.2
> diff -r1.2 lmicon.ico
> Binary files /tmp/T0VRaaU3 and lmicon.ico differ
> RCS/lmicon.ico,v <-- lmicon.ico
> new revision: 1.3; previous revision: 1.2
> enter log message, terminated with single '.' or end of file:
> >> SLM800 new icon.
> >> .
> ci: RCS/lmicon.ico,v: bad diff output line: Binary files /tmp/dQDaWU3 and
> /tmp/T
> 28RaOU3 differ
> ci aborted
>
> **********
> I think this is a bug. Any help on this will be highly appreciated. Please
> let me know if any other information/detail is required.
>
> Regards,
> Pratyush Kumar