I can try and get back. Any documentation link on how to make web2py use psycopg2? I am very much a web2py/pyhon newbee.
On Tuesday, December 17, 2013 7:58:26 PM UTC+5:30, Massimo Di Pierro wrote: > > What if you use psycopg2? > > Massimo > > On Tuesday, 17 December 2013 08:25:01 UTC-6, Jayadevan M wrote: >> >> No. The number of fetches go up when the number of records go up. It is a >> very simple test case. First I inserted 100 records. >> insert into myt (nm, title) select generate_series(1,100), 'A'; >> >> There was just one fetch. Then I inserted many more >> insert into myt (nm, title) select generate_series(100,100000), 'A'; >> >> If I see the log, it is like this . A parse, a bind, an execute, then all >> 'execute fetch'es. >> 2013-12-17 19:07:57 IST [19893]: [23-1] user=myserrhpa,db=everest1 LOG: >> duration: 0.577 ms parse <unnamed>: select * from myt >> 2013-12-17 19:07:57 IST [19893]: [24-1] user=myserrhpa,db=everest1 LOG: >> duration: 0.420 ms bind <unnamed>/pg8000_portal_10: select * from myt >> 2013-12-17 19:07:57 IST [19893]: [25-1] user=myserrhpa,db=everest1 LOG: >> duration: 0.148 ms execute <unnamed>/pg8000_portal_10: select * from myt >> 2013-12-17 19:07:57 IST [19893]: [26-1] user=myserrhpa,db=everest1 LOG: >> duration: 0.151 ms execute fetch from <unnamed>/pg8000_portal_10: select * >> from myt >> 2013-12-17 19:07:58 IST [19893]: [27-1] user=myserrhpa,db=everest1 LOG: >> duration: 0.116 ms execute fetch from <unnamed>/pg8000_portal_10: select * >> from myt >> >> My test.py has just this >> def querytest(): >> sql="select * from myt " >> rows=db.executesql(sql) >> return dict(rows=rows) >> >> On Tuesday, December 17, 2013 7:46:45 PM UTC+5:30, Massimo Di Pierro >> wrote: >>> >>> You may have hidden recursive selects, for example in a represent >>> function. >>> >>> On Tuesday, 17 December 2013 07:41:47 UTC-6, Jayadevan M wrote: >>>> >>>> When I turned on logging for PostgreSQL during testing, I see that the >>>> number of fetches goes up dramatically for tables with a large number of >>>> rows. For example - >>>> 2013-12-17 19:08:03 IST [19893]: [791-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.093 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [792-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.137 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [793-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.093 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [794-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.096 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [795-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.063 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [796-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.091 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [797-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.064 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [798-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.094 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> 2013-12-17 19:08:03 IST [19893]: [799-1] user=myserrhpa,db=everest1 >>>> LOG: duration: 0.068 ms execute fetch from <unnamed>/pg8000_portal_10: >>>> select * from myt >>>> Is this normal? How can we control the number of fetches - some >>>> parameter to set like array_size? >>>> >>> -- Resources: - http://web2py.com - http://web2py.com/book (Documentation) - http://github.com/web2py/web2py (Source code) - https://code.google.com/p/web2py/issues/list (Report Issues) --- You received this message because you are subscribed to the Google Groups "web2py-users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/groups/opt_out.
