Import Routines works on Mac but not PC!?!?!?!?

[Tomo1]

I have a Database that imports records from a text file. The text file is on a server run Solaris. It works fine from a PC but when I run it from it from a Mac Get I get a message 'There are no records for import'. I know this doesn't contain a lot of info so let me know if you require anything further?

[IdealData]

The subject and content of your request are completely opposed - which is the one that works ?

[Tomo1]

Doh. I've had a hard day. The subject should read: Import Routines works on PC but not on a Mac!?!?!?!

[Rigsby]

Please excuse me if I am well off the mark here, but I

[BobWeaver]

Just a guess here. Solaris is Unix which uses a linefeed to indicate end of line. PC's uses carriage return and linefeed to designate end of line. Mac uses a carriage return to designate end of line. I suspect you need to the the file through some translation utility that translates the linefeed into a carriage return.

[IdealData]

I think Bob is on the right lines here. Both platforms will expect to find the appropriate record and file terminator characters.

Also try running the script "by hand" on both platforms.

If you can see the text file and the "hand" event works then it's probably a filespec problem. This is certainly very different between PC and Mac and can cause scripts to fail because it can't find the file.

When you get a solution don't forget to let us know.

[CobaltSky]

I'd be guessing that Bob's suggestion is on the money, and that your problem lies with the record terminator characters used on the Solaris platform. I've not encountered this with Solaris, but have with other Unix sources.

If this is a problem you are likely to encounter with any frequency, you might wish to script a solution from within FileMaker, using the FreeWare plug-in from Protolight called File Toolbox:

Free download from: http://www.protolight.com/plPLUGs.html

This plug-in will allow you to work through the FileMaker external function interface to script a find and replace for designated ascii characters within a file at a specified path. You should be able to set it up to replace the line feed (ascii char# 010) characters with carriage returns (ascii char# 013), perform the import, then (if desired) change the 013s back to 010s in the remote file.