In article <1127986617.211523.221500 (AT) g43g2000cwa (DOT) googlegroups.com>,
nicolas.borde (AT) gmail (DOT) com says...
Quote:
Hi there,
I read a lot about Filemaker and ODBC. But is it possible to use SQL
queries in filemaker |
Sure you just setup a DSN for Filemaker as an ODBC source, make sure the
database is set to accept odbc, and then query that dsn through the SQL
execute script step, which pushes it through the odbc driver, and
finally imports the result set back into filemaker as new records.
Ah. No.
All that messing around the limited sql language support, translating
queries into something filemaker can process and then pushing all the
data through an external driver, then importing (and possibly indexxing
that import) tends to eat up the performance.
Quote:
That way I could make it really quicker and avoid the use of a lot of
statistics and calculations in my tables. |
Good database design, leveraging the strengths, and avoiding the
weaknesses within filemaker is the only way to boost performance.
Talking to it through a 3rd party driver in a foreign language is NOT
going to speed things up for you.
ODBC is useful as a standard, as it allows software developers to target
different databases with a common interface, but its rarely, if ever,
the fastest way of interfacing with a database.
Even MS SQL Server's performance suffers if you use ODBC, performance is
much faster using Microsofts various SQL Server specific drivers e.g.
the ADO.NET SQLServer providers over the ODBC providers.
ODBC for practically any database adds a layer of processing work. For
some databases like sql server that layer is pretty thin, for others,
like filemaker, that layer is pretty thick.
-cheers,
Dave