I initially came here having found the following thread:
thread701-1554516
This person was experiencing the same problem I am: Access 2007 does not appear to support importing/exporting text files using extended ASCII characters as delimiters. I am also trying to import Concordance DAT files using chr(254) and chr(020) as delimiters, which worked fine in Access 2003. However, now I get a vague XML-related error on importing which makes no sense.
Taking a suggestion from the previous thread, the chr(020) seems to be the main culprit causing the error message, since when I change it to something else (pipe) and keep the chr(254), the file will import. However the text delimiter chr(254) is not removed so the resulting table is not correct.
For instance, if I import this line:
þBEGDOCþ|þENDDOCþ|þNameþ|þFileNameþ
although the chr(254) characters are removed properly in the import wizrd preview, they are still imported, and not removed as text delmiters as they should be so that my first field comes out as:
þBEGDOCþ
instead of
BEGDOC
I realize that I can just change the delimiters, but the reason we use those characters is because these text files often hold OCR text, and the delimiters have to be characters which are rarely, if ever, found in OCR so that the fields can be split properly and so that any find-and-replace operation won't accidentally replace something meant to be part of the fielded data. Therefore, any delimiters that I pick will invariably need to be extended ASCII.
Plus, this is just frustrating that I can find hardly any documentation about Access 2007 supporting extended ASCII... especially since 2003 used to do it just fine.
thread701-1554516
This person was experiencing the same problem I am: Access 2007 does not appear to support importing/exporting text files using extended ASCII characters as delimiters. I am also trying to import Concordance DAT files using chr(254) and chr(020) as delimiters, which worked fine in Access 2003. However, now I get a vague XML-related error on importing which makes no sense.
Taking a suggestion from the previous thread, the chr(020) seems to be the main culprit causing the error message, since when I change it to something else (pipe) and keep the chr(254), the file will import. However the text delimiter chr(254) is not removed so the resulting table is not correct.
For instance, if I import this line:
þBEGDOCþ|þENDDOCþ|þNameþ|þFileNameþ
although the chr(254) characters are removed properly in the import wizrd preview, they are still imported, and not removed as text delmiters as they should be so that my first field comes out as:
þBEGDOCþ
instead of
BEGDOC
I realize that I can just change the delimiters, but the reason we use those characters is because these text files often hold OCR text, and the delimiters have to be characters which are rarely, if ever, found in OCR so that the fields can be split properly and so that any find-and-replace operation won't accidentally replace something meant to be part of the fielded data. Therefore, any delimiters that I pick will invariably need to be extended ASCII.
Plus, this is just frustrating that I can find hardly any documentation about Access 2007 supporting extended ASCII... especially since 2003 used to do it just fine.
