On Sun, 24 May 2009 21:53:55 -0700 Ryan Niebur <[email protected]> wrote:
> what needs to be done for this bug report? > Bottom line: we should consider checking if 'chdir $trunkdir' was really > executed before calling SDCommon::writeCfg but the user will always > remain in the directoy (s)he was when starting svn-whatever. So a we need to test that chdir returns true and die with a message if it fails. We also need to update the manpage to remind users that the working directory of the user can only be changed by the user and not by the script itself. Give the example of uupdate and other tools. Indeed, using the example of uupdate, we should probably consider a message in the standard output of the tool stating that "foo now exists in $dir - you need to change into that directory before proceeding.\n e.g. 'cd $dir'\n" or something along those lines. > afaict it was just Jeremiah > misunderstanding the output of svn-inject. Can we close it? Let's get that check into svn-inject first, improve the output of svn-inject for more clarity and get a note into the manpage. -- Neil Williams ============= http://www.data-freedom.org/ http://www.nosoftwarepatents.com/ http://www.linux.codehelp.co.uk/
pgpwFoQgsHfuh.pgp
Description: PGP signature
