Re: new bite-sized task, issue #598

cmpilato@collab.net writes:

> First, a minor coding style nit. This file uses spaces between
> function calls and '('. That is, it should be 'printf ("file...' just
> as you've got it for the rest of your patch.

gotcha. will do.

>
> Secondly, we generally don't print error-ish conditions to the
> terminal. Perhaps we should either print nothing for files that don't
> exist, or return an appropriate "no such path" svn_error_t.
>
> Whatcha think?

ya, I considered that. but what if we're given a list of files to
stat, and only one is bogus? do we want to bail if any file doesn't
exist? CVS(*) doesn't do this. that is, it will print a message and
continue with the rest of the files given on the command line. the
whole point of the issue, as I construed it, was that svn was
currently printing nothing for files that don't exist, and that that
was wrong.

the error message is a bit arbitrary, though. could there be another
status character used for non-existant files? maybe '~'? the display
would at least be consistent for all output lines then. I suppose an
easy way of accomplishing that would be to change the status character
for svn_wc_status_none from ' ' to '~'. maybe that's a better
approach?

(*) is it OK to use CVS behavior as a baseline? is this considered
poor taste?

-- 
joe
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org