> Hartmut Holzgraefe wrote: > > > i've started looking into the patch yesterday, there are still some > > small issues with it though ... i'll provide you with more detailed > > comments later today or early tomorrow ... > > open questions: > > - should the functions return an error if using the specified oid > isn't possible or just throw a warning and return the oid that > was actually assigned and used by the server?
I think it should return errors in the case. The intention for the PostgreSQL API is, avoiding automatic oid assigning by PostgreSQL. This is necessary for query based replication tools such as pgpool. Plus, if PostgreSQL fails to assign the specified oid, the transaction will be in abort state anyway. > changes: Thanks. I will look into these. -- Tatsuo Ishii SRA OSS, Inc. Japan > - there was a mismatch between autoconf macro names and what > was tested against in the source > > - ZEND_ARG_INFO entries for the new parameter > > - pg_lo_create() > > - option parsing is more lightweight and only > one #ifdef is needed > > - added a test case file > > - updated documentation > > - pg_lo_import() > > - new parameter is now placed after file name, not before, > as we usually do with new parameters > > - only one #ifdef block needed > > - updated documentation > > - still need to add a test case, function compiles but > i haven't tested it yet > > - possible coding style issues, still breaking in a new > emacs installation ... > > -- > Hartmut Holzgraefe, MySQL Regional Support Manager EMEA > > Sun Microsystems GmbH, Sonnenallee 1, 85551 Kirchheim-Heimstetten > Amtsgericht Muenchen: HRB161028 > Geschaeftsfuehrer: Thomas Schroeder, Wolfgang Engels, Dr. Roland Boemer > Vorsitzender des Aufsichtsrates: Martin Haering -- PHP Internals - PHP Runtime Development Mailing List To unsubscribe, visit: http://www.php.net/unsub.php
