The reason quotes are needed in CSV is because some brain surgeon long ago decided that commas aren't human characters and that they should become value delimiters. Thus, CSV elements containing commas need the surrounding quotes.
We MV people are thankful that our delimiters are far away from the human characters, 254 etc. Early MS basic allowed INPUT A,B and your answer contained a comma and their INPUT put it into the 2 variables. If your single answer to INPUT A was "Perth,SW", you got an 'extra ignored' and lost the SW.
It's a shame that we must live within the poor judgement of these early designs. But they're out there.
I have 2 utilities that i use extensively when building CSV's (rows & columns) from within Databasic when export/importing to a PC system. The first is called LINE.CSV and it converts LINE<1>, LINE<2> LINE<3> (a built variable representing one row) into a CSV row for appending to your eventual text file (OPENSEQ or emulator export). I have its opposite called CSV.LINE which will convert an invoming CSV row into LINE<1> etc paying attention to the commas. If anyone would like a copy email me off line. I'm probably going to publish them in Spectrum anyway.
BTW, to answer the second question, to remove the pennies from an ICONV'd value you would use MD03. That's Zero Three , not Oh. The Zero indicates the displayed decimals and the 3 is the converted. Thus, MD3 is the same as MD33.
Also, check to see if the comma contained in a number offends the numerical format of that cell. I don't think the comma contained is numeric, ie PRINT NUM("123,456") returns false and PRINT NUM(123,456) implies a second argument to the NUM function. you can't say X=123,456 and if you put quotes around it, it's not numeric.
my 1 cent.
_______________________________________________ u2-users mailing list [EMAIL PROTECTED] http://www.oliver.com/mailman/listinfo/u2-users