[RESOLVED] Access specification filetype for transfertext of CSVs

[DaveBo]

I'm trying to generate specifications files in Access automatically (in code) to handle different formats of CSV files.

i.e. I read in the 1st line containing the column names and then create a new spec in the MSysIMEXSpecs & MSysIMEXColumns tables to force all fields to come in as text.

It works fine except some files insist on a spec filetype in MSysIMEXSpecs of 1200 and others of 1252 and I can't figure out which wants what except by trial and error.
Apparently, 1200 is for unicode and the 1252 is Latin something.

Does anyone know if there's a way around this or if there's a way I can test the data file, via code, to see if it's unicode etc?

Thanks, DaveBo

[anhn]

The FileType field in MSysIMEXSpecs table with Jet 4.0 and later is the code page to use.

See full listing of codepages here: Character Set Recognition

Beside that (not sure):
0 = Windows (ANSI)
1 = DOS or OS/2 (PC-8)

[salvelinus]

I generally just manually import the .csv file first, and save the import spec there.
I realize you want to do it in code, so maybe that doesn't help you, but this way you won't have these errors.

[DaveBo]

Thanks, that's helpful. The CodePage field from that "Character Recognition" table seems to be what the specification FileType wants in it.

Somehow, when doing a manual import and creating a spec from that it knows what value to use here. I was hoping I could do the same when creating my specs from code.

The strange thing is, when I auto-generate the spec and import the file with my routine it looks like it worked, but if the FileType setting is wrong the new table is empty.

Thanks, DaveBo

[anhn]

If that helps you and your problem has been solved, please use thread tools to mark this thread as RESOLVED and you should rate the post that helps you.