| From: | Jaime Casanova <systemguards(at)gmail(dot)com> | 
|---|---|
| To: | PG Hackers <pgsql-hackers(at)postgresql(dot)org> | 
| Subject: | someone working to add merge? | 
| Date: | 2005-11-11 14:03:19 | 
| Message-ID: | c2d9e70e0511110603q1799d811u6e4564be516b10ce@mail.gmail.com | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-hackers | 
Hi,
there is someone working in add the MERGE statement?
i don't find much about what a good implementation of merge must have...
i think what it needs to do is something like:
- try to lock the rows for update
    - if the lock cannot be immediatly acquire ask why
    - if the rows are already locked, wait and try again?
    - if no rows were found try de insert part
    - if there was any other error, abort
    - else update
so i suppose we can reuse many of the code breaking the merge in 3
pieces... for now they are just thougths, i will think more in this
and try to implement it...
comments? ideas? suggestions?
--
regards,
Jaime Casanova
(DBA: DataBase Aniquilator ;)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Martijn van Oosterhout | 2005-11-11 14:04:06 | Re: Getting table name/tuple from OID | 
| Previous Message | Merlin Moncure | 2005-11-11 13:50:58 | Re: 8.0 -> 8.1 dump duplicate key problem? |