On Thu, Jul 4, 2013 at 5:51 PM, Bert Huijben <b...@qqmail.nl> wrote: >> -----Original Message----- >> From: phi...@apache.org [mailto:phi...@apache.org] >> Sent: donderdag 4 juli 2013 15:22 >> To: comm...@subversion.apache.org >> Subject: svn commit: r1499747 - >> /subversion/trunk/subversion/tests/libsvn_wc/op-depth-test.c >> >> Author: philip >> Date: Thu Jul 4 13:22:00 2013 >> New Revision: 1499747 >> >> URL: http://svn.apache.org/r1499747 >> Log: >> Add a regression test for issue 4388. >> >> * subversion/tests/libsvn_wc/op-depth-test.c >> (check_db_actual): Reset statement on error. >... >> @@ -1893,7 +1893,7 @@ check_db_actual(svn_test__sandbox_t* b, >> { >> const char *local_relpath = svn_sqlite__column_text(stmt, 0, b->pool); >> if (!apr_hash_get(path_hash, local_relpath, APR_HASH_KEY_STRING)) >> - return svn_error_createf(SVN_ERR_TEST_FAILED, >> svn_sqlite__close(sdb), >> + return svn_error_createf(SVN_ERR_TEST_FAILED, >> svn_sqlite__reset(stmt), >> "actual '%s' unexpected", local_relpath); >> apr_hash_set(path_hash, local_relpath, APR_HASH_KEY_STRING, NULL); >> SVN_ERR(svn_sqlite__step(&have_row, stmt)); >> @@ -1903,7 +1903,7 @@ check_db_actual(svn_test__sandbox_t* b, >> { >> const char *local_relpath >> = svn__apr_hash_index_key(apr_hash_first(b->pool, path_hash)); >> - return svn_error_createf(SVN_ERR_TEST_FAILED, >> svn_sqlite__close(sdb), >> + return svn_error_createf(SVN_ERR_TEST_FAILED, >> svn_sqlite__reset(stmt), >> "actual '%s' expected", local_relpath); > > Are these two changes related to the test? Or is this just fixing existing > bugs?
Yeah. The log message says *what*, but doesn't explain *why*. We like "why" in our log messages :-) Cheers, -g
