N
Nils Titley
Ok,
I see that there is a TextToColumns. Can you help me. Can I read the file
and write it back out over it self or do I have to create a new file?
I see that there is a TextToColumns. Can you help me. Can I read the file
and write it back out over it self or do I have to create a new file?
Dave Peterson said:If you're dates in the CSV file are in dmy order, then I would rename the
workbook to *.txt. Then I could import the data and specify that field as a
date in dmy order.
I _think_ that what's happening is that other users have a different data
order. The data that could pass for dates in that formatted will be converted
to a real date -- but probably not what you want.
For instance if the data shows: 12/02/2008
and you bring it in with your shortdate format of dmy, then you'll see a date of
Feb 12, 2008.
But if I bring it in using my mdy shortdate format, I'll see a date of Dec 2,
2008.
The real bad news is that with an ambiguous date format of mm/dd/yyyy, you can't
tell if the date is really want the CSV intended.
And those "dates" that do not look like dates that match the users short date
format (25/02/2008 for instance) will be a real date for you (dmy) = Feb 25,
2008.
For me with my mdy order, I'll get a text string of 25/02/2008. It won't be a
date.
So my recommendation is to not rely on the user's setting. Bring the data in a
different way--either by renaming the file as *.txt (so you see the wizard).
You could also try: Data|Import external data|get data to see the same text to
columns wizard.
====
My real point is just because the value looks like a date doesn't mean that it
is a date. And even if it is a date, it may not be what the original data
intended.