Re: Transfer table to fixed length text records

Tech-Archive recommends: Speed Up your PC by fixing your registry

1) Are you looking at the size of the file as reported in Windows Explorer
(which is approximate) or the exact size in bytes (e.g. in its properties
***)?

2) Check the encodings (code pages) involved. The one your computer uses for
the export may not be what the other computer expects, and this could cause
it to detect the wrong number of characters per line.


"Banterista" <Banterista@xxxxxxxxxxxxxxxxxxxxxxxxx> wrote in message
news:D83817D4-0D67-424B-9FFE-6749E2CB2E65@xxxxxxxxxxxxxxxx
Hi Phil,

I'm sorry I must have had a brainstorm! It's a fixed length length file
not
delimited. I've already tried your suggestion and in notepad the cursor
sits
on the 300th character and when you move the cursor one position to the
right, the line feed kicks in and moves you to the beginning of the next
line. What I don't understand is that the text file size stays the same
even
though I have reduced the record size to say 298 or 299 to allow for line
feed or carriage return characters. I've got over 900 records so there
should
be some variation in overall file size, I would have thought. There must
be
some hidden filler or control characters somewhere.

"Phil" wrote:

Look at it and see. Open the text file in a preferably stupid text
editor, like notepad. You don't want anything trying to "fix" the
results. See what you get.

A couple things pop to mind.
150 + 150 is going to be perhaps 307 bytes, two fields, two pairs of
quotes, a comma or tab, carriage return, and linefeed.

Are you sure this "well established" package is expecting a delimited
file, not fixed length, and are you sure that you are using the exact
same delimiters, for field, record and text, that this package expects?
They can be very picky sometimes.


Banterista wrote:

I have a table the consists of two fields that together make 300 bytes.
When
I export this delimited file to text there appear to be 300 characters.
However when this text file is imported to a well established package,
it is
validated for having records that are 300 bytes long. Needless to say
my text
file fails this check. I have reduced my table size to allow for a line
feed
or carriage return but it still fails the check. The strange thing is
that
even though I have reduced the record length the output text files are
still
the same overall size. Are there some hidden characters in the text
file to
take into consideration here?



.

Quantcast