Jeff Thies wrote:
> 
>   After trying for a week to get the DBD::mysql driver loaded in FreeBSD
> (couldn't find DBI during install) virtual server I thought I'd try
> DBD::AnyData.
> 
>   Much to my surprise, I could do this:
> my $sql='create table table1(field1 varchar(12),field2 varchar(12))';
> 
> Are any other data types supported by any of the formats?

Data types are currently "supported" in the sense that you can use them
in a CREATE statement as you did.  SQL::Statement (which underlies
AnyData) will check that the types are valid SQL92 types and give an
error message for the CREATE if not.  It then promptly forgets what you
told it about types.  Eventually it won't do that (expect data type
support in the next month or two).  SQL::Statement won't let you do
things like try to do math operations on strings, but it does that by
looking at the actual values in the columns, not by refering to how the
data was typed in the CREATE statement.  

> Any reason to prefer Pipe over CSV over Tab? Or is it just the way I
> prepare the data to be inserted?

These have to do with the storage of the data and nothing else.  Once
you declare how you have the data stored (or how you want it stored if
you are creating the table for the first time), it really makes no
difference, they all operate the same.  If you have data stored in one
format and you want it stored in another, see the import/export/convert
methods.

> Any way to hack an id field or auto increment?

That is also on my to-do list.  In the meantime, you can create your own
(e.g. by doing a MAX(id) and incrementing that and using the result as
the id for the next insert) or use one of the CPAN modules that creates
counters and/or unique ids.

> Any cool reasons to use the XML format if I'm not going to be exporting
> XML.

Nope.  If you don't have or want the data in XML, there's no reason to
prefer it over other formats.

-- 
Jeff

Reply via email to