Log Message:
-----------
Stop setting probin to the compiled bytecode.
Ultimately the savings yielded by caching the bytecode are normally slight, and
using probin broke pg_dump's, so this ended up not being a win. However, if
compilation of large procedures does prove to be taxing, this feature can be
reincarnated by creating a PL/Py bytecode cache table. Until that time comes,
just go through the regular compilation route when loading a Python procedure.
Modified Files:
--------------
be/src:
function.c (r1.22 -> r1.23)
(http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/python/be/src/function.c.diff?r1=1.22&r2=1.23)
module.c (r1.29 -> r1.30)
(http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/python/be/src/module.c.diff?r1=1.29&r2=1.30)
pl.c (r1.39 -> r1.40)
(http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/python/be/src/pl.c.diff?r1=1.39&r2=1.40)
be/include/pypg:
cis.h (r1.12 -> r1.13)
(http://cvs.pgfoundry.org/cgi-bin/cvsweb.cgi/python/be/include/pypg/cis.h.diff?r1=1.12&r2=1.13)
---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match
