I’ve run into occasional weird issues like this when trying to import a data set that contained non-alpha characters. In my case, the data — furnished by a client — imported correctly, but any attempts to search the resulting base or perform comparisons failed miserably. Even worse, when opened in any of the suite of word processors, text editors, and spreadsheets I typically keep on hand, the offending values were stripped out or rendered invisible. It wasn’t until I opened the file using a hexeditor — and at this point I don’t recall which one; irritatingly, the software download folder from that particular machine didn’t make the cross-country trip — that I could determine which lines from the CSV were corrupted.
If you haven’t been able to find a solution, I’d suggest finding a freeware hex editor and examining the source file. Look especially for embedded ASCII 0x0a, 0x0b, 0x0c, or 0x0d values, which might be signalling end-of-line to Airtable’s parser.
Edit: In looking back through old posts, I see from this the hex editor I used must have been Frhed.
