Re: [PHP4BETA] cvs: /php4/ext/odbc php_odbc.h php_odbc.c
| From: | Leo Baschy | Date: | Sun, 12 Mar 2000 21:00:17 +0000 |
| Subject: | Re: [PHP4BETA] cvs: /php4/ext/odbc php_odbc.h php_odbc.c | ||
| References: | 1 | Groups: | php.version4 |
| Request: | Send a blank email to php-version4+get-11986@lists.php.net to get a copy of this message | ||
What we are discussing is almost like a chicken and egg problem, but not exactly.
We have to make up our mind:
- EITHER is the odbc_result dependent on a connection, dependent on a statement, is it our access
to ODBC,
- OR is the odbc_result an entity of its own that can carry data even when there is no ODBC, no
statement, no connection.
If odbc_result is our access to ODBC then it must disappear when the statement or the connection
disappear.
If odbc_result is an entity of its own then it must be in all access code fully protected, with
if-else for the cases when the connection, statement, ODBC exist or not.
One way or the other works. If neither, then crashes. If both, then stuck.
I vote for keeping it in the spirit of ODBC and see it as access to ODBC. Life and coding are
difficult enough as it is, we don't need another paradigm.
As a side effect in case EITHER it might work better without reference counting, in case OR it might
work better with reference counting.
If anything I write sounds not polite I apologize summarily. My main focus is to make a reliable
and useable package and I sincerely appreciate everyone's efforts.
- Leo Baschy <nirvana@got.net>