> -----Original Message----- > From: i...@apache.org [mailto:i...@apache.org] > Sent: woensdag 3 augustus 2011 12:10 > To: comm...@subversion.apache.org > Subject: svn commit: r1153416 - > /subversion/trunk/subversion/libsvn_wc/status.c > > Author: ivan > Date: Wed Aug 3 10:10:22 2011 > New Revision: 1153416 > > URL: http://svn.apache.org/viewvc?rev=1153416&view=rev > Log: > Fix access to uninitialized variable. > > * subversion/libsvn_wc/status.c > (internal_status): Do not access NODE_STATUS if read_info() function > returned error. > > Modified: > subversion/trunk/subversion/libsvn_wc/status.c > > Modified: subversion/trunk/subversion/libsvn_wc/status.c > URL: > http://svn.apache.org/viewvc/subversion/trunk/subversion/libsvn_wc/stat > us.c?rev=1153416&r1=1153415&r2=1153416&view=diff > ========================================================== > ==================== > --- subversion/trunk/subversion/libsvn_wc/status.c (original) > +++ subversion/trunk/subversion/libsvn_wc/status.c Wed Aug 3 10:10:22 > 2011 > @@ -2464,21 +2464,24 @@ internal_status(svn_wc_status3_t **statu > db, local_abspath, > scratch_pool, scratch_pool); > > - if ((err && err->apr_err == SVN_ERR_WC_PATH_NOT_FOUND) > - || node_status == svn_wc__db_status_not_present > - || node_status == svn_wc__db_status_server_excluded > - || node_status == svn_wc__db_status_excluded) > + if (err && err->apr_err == SVN_ERR_WC_PATH_NOT_FOUND)
Did you get an error from this if somehow? (Or via a analysis tool) We use this pattern in a lot more places, in cases like r = (s != NULL) ? s->var : NULL, and that is safe in C and most other languages that use the same evaluation rules. And even if it wouldn't be safe for evaluation reasons this test wouldn't have a problem because any value of node_status would be ok to have the same result. Bert