On Feb 24, 2006, at 12:25 PM, Michael Bayer wrote:
yeah, if I just put a print statement on line 282 of schema.py,
which your test in the first case says was called once (which makes
no sense at all), in the other says was called 94 times (more
reasonable), I get the identical number of print's for a given
program. I think the profiler is getting screwed up by the very
deep recursion (or whatever it is) introduced by the bug.
Using a different profiler (hotshot this time). Profiled "import
model", only showing results for __getattr__ on line 49 of schema.py.
Before:
78488 function calls (13280 primitive calls) in 5.488 CPU
seconds
ncalls tottime percall cumtime percall filename:lineno(function)
64754/339 3.674 0.000 3.675 0.011 schema.py:49(__getattr__)
After:
14073 function calls (13280 primitive calls) in 0.903 CPU
seconds
ncalls tottime percall cumtime percall filename:lineno(function)
339 0.004 0.000 0.006 0.000 schema.py:49(__getattr__)
Wow, very different results--you're right, the number of primitive
calls is the same in both cases. Anyway, it looks like you fixed a
major bug here.
~ Daniel
-------------------------------------------------------
This SF.Net email is sponsored by xPML, a groundbreaking scripting language
that extends applications into web and mobile media. Attend the live webcast
and join the prime developer group breaking into this new coding territory!
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=110944&bid=241720&dat=121642
_______________________________________________
Sqlalchemy-users mailing list
Sqlalchemy-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/sqlalchemy-users
