Re: svn st .. regression on trunk

Eric Gillespie <epg@pretzelnet.org> writes:

> 0 obj% svn st ..
> ../subversion/libsvn_wc/lock.c:818: (apr_err=155022)
> svn: Path '..' ends in '..', which is unsupported for this operation
>
> This has always worked, up to 1.1.3. I saw some posts recently
> about handling of .. , maybe that has something to do with it?

The check in question, lock.c:818, always looked out of place in
svn_wc_adm_open_anchor and it's possible it could be removed, a quick
test shows that "svn st .." then appears to work. I intended for
svn_wc_adm_open_anchor to reproduce the constraints of the functions
it duplicated/optimised but the regression tests don't have good
coverage of odd paths like ".." so I may have got it wrong. The code
is common to to other functions like update and switch so it would
need a bit of testing.

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