For text files I like to use openseq and readseq. With embedded CR/LF's (which
are actually valid in a csv file) you would have to see if you needed to put
the lines back together. There were a couple of examples of reading csv files a
while back in CDP.
The other alternative is to use osopen and osbread. This will read a block of
text (size of your choosing and I've found you have to play a bit to get good
throughput). You then have to parse through the block literally character by
character.
hth
Colin Alfke
Calgary, Canada
> From: wphaskett
>
> I've received some import files (csv) that use either a CR and/or LF
> character in a test field (surrounded by "-marks). It seems to be
> somebody's idea of multi-value. However, I'm having a hard time
> extracting them (dynamic length lines) because UD considers it a
> field-mark. Therefore, the REMOVE() function, READSEQ statements don't
> work (a single line is considered two lines), and FORMLIST/READNEXT
> don't work because the key is too long for a select list.
>
> Anyone know how to extract a line of a simple csv file that contains a
> CR and/or LF within the line?
>
> Thanks,
>
> Bill Haskett
_______________________________________________
U2-Users mailing list
U2-Users@listserver.u2ug.org
http://listserver.u2ug.org/mailman/listinfo/u2-users
