On 15 July 2010 20:37, Michael A. Peters <mpet...@shastaherps.org> wrote: > I'm currently working on a migration of my web site to postgresql so that > I can use PostGIS. > > I'm still working on the devel side but basically - > > mysqldump database > dump.sql > perl mysql2pgsql.perl dump.sql load.sql > psql database < load.sql > > does the trick, now it's cleaning up my sql in my code. > > I used MDB2 so once I finally figured out how to make a web user with > necessary permissions most of it "just worked" but - I had a table with a > column named habitat. On a query on that table, query failed. When > manually applying the query, it said habitat did not exist. > > I looked in the load file (there's no describe table ??) and habitat was > changed to habyteaat but I can find nothing about habitat being a special > word to postgresql. Is that a bug in the perl script I can just sed to fix > the output or do I really need to use a different column name? > >
Hmm funny: seems that the perl script is replacing every instance of 'bit' to 'bytea' (a PG datatype). definitely, 'habitat' is a valid column name. Nicolas _______________________________________________ postgis-users mailing list postgis-users@postgis.refractions.net http://postgis.refractions.net/mailman/listinfo/postgis-users
