Howdy folks, I've been trying to test some cross compatability of sqlite, and having a heck of a time with sqlites database .dumps and MiXeD CaSe based schemas. The problem stems from an odd behavior of .dump to not quote identifiers when outputting DDL, but to then quote the identifiers when outputting the DML. (Is this really the desired behavior? I can't think of any reason you would want it to work like this) This works ok for dumps going back into sqlite, however it doesn't work with other database systems, which attempt to casefold the schema, but then attempt to load data using the non-case-folded table names given in the insert statements. I think the ideal solution would probably be to have an option to switch between consistently quoteing or unquoteing (I can think of use cases for both methods), is there any chance something like that could be implemented? In the mean time, I'm wondering if anyone can think of another work around for this? TIA.
-- Robert Treat brighterlamp.org _______________________________________________ sqlite-users mailing list sqlite-users@sqlite.org http://sqlite.org:8080/cgi-bin/mailman/listinfo/sqlite-users
