[postgis-users] bugs in postgis-1.0RC2 with postgres 7.2.1

[alexbodn at 012.net.il]

hi markus,

although i am mainly concerned with debian, which currently uses postgresql 7.2.1 in woody and 7.4.7 in sarge (older debian version, potato, uses postgresql 6.5.x, not supported by postgis at all), i also wish to contribute to best versioning management in postgis itself. the only reference to postgis on postgresql 7.1.x was about a red hat installation.

postgis is one of the few software packages that support such a wide range of another package, here postgresql. of course, not all postgis subpackages have to support all postgresql / pgjdbc versions, but we just have to put it clearly in the makefiles.

> I think that we cannot use the version information that was extracted 
> for compiling PostGIS itsself (the C code) because we have no guarantee 
> that the pgjdbc.jar is the same version. In fact, I currently use the 
> 8.0 jar file because of some huge improvements, but have a V7.4 server.
> I'm not really shure how we can do it. It may be necessary to use java 
> code for this version extraction, or do some "probe compiling".

to compile postgis jdbc, we always need the javac, so what we need might be a small java program that would use the postgresql.jar from the classpath, and print the pgjdbc version on stdout. the output might serve for code preprocessing (postgis sql scripts contain embedded #ifdef's for cpp).

also, for packaging purposes, there should be known what pgjdbc served to compile postgis jdbc, and the jar depends on it.

> As far as I know, PostGIS itsself only supports 7.2 and upper.

the postgis documentation and sources show support for 7.1 and later.

> Is there any real need to use a pgjdbc 7.1 jar against a 7.2 server running PostGIS?

nice joke. i would rather think pgjdbc 7.1 with pgsql 7.1. from your message, the pgjdbc interface has changed with 7.3, so there might be only a matter of test, that i might be able to do.

best regards,

alex