Is there a MacOS ODBC driver that reads SQL-command text files?

I've been searching without luck for a MacOS iODBC driver that can read saved .SQL files exported in Microsoft SQL Server format. Does one exist?

We've got a large pile of research data stored in one app that can export as Excel spreadsheets or SQL files (eg, a text file full of SQL CREATE TABLE and INSERT statements). We need to import this data into another app (Stata 开发者_Go百科9) that runs under MacOS and can import Excel files, its own format, or from an ODBC source. So, I need an ODBC driver that can read plain SQL files as its source. We don't need a driver that actually talks to an MSSQL database, because there is no actual database here; just a plain .SQL file with MSSQL-style commands in it.

Unfortunately, the default MacOS install seems to come with no ODBC drivers whatsoever, not even one for reading flat files or SQLite databases.

The current workflow for moving this data — exporting it from DatStat as an Excel spreadsheet, opening that spreadsheet and fixing it by hand to conform to Stata's need, then saving and reimporting into Stata — is ridiculously labor-intensive and also loses a lot of important metadata like variable descriptions and annotations.

I think that best thing to do here is load the data from DatStat to a database and then load it back into Stata. First, export your data from DatStat to a .sql file. I'm not familiar with DatStat, but if you can do this in bulk or via the command line it would be best. You can access your OS's terminal in Stata by using the -shell- command. After you have a .sql file, say foo.sql, you can use the following Stata code to send it to a database and then import into Stata.

odbc sqlfile("foo.sql"), dsn("DataSourceName")
odbc load, exec("SELECT * FROM CustomerTable") dsn("DataSourceName")

You could even issue a final command to cleanup the tables in the database if you don't think you'll use this database again and you don't want it taking up space. Use something like:

odbc exec("DROP TABLE CustomerTable")

Yes, this will probably be slow if your dataset is large, but it could be nice once your data is in the database because you can query parts of it at a time instead of importing the whole thing.

Lastly, you mentioned that no ODBC driver for Mac exists for MS SQL Server. If that is the case, you may want to install one of the open-source database systems like MySQL or PostgreSQL. I'm not a Mac user but drivers for these must exist for mac.

Good luck!