I am trying to extract data from a CSV file which is TAB delimited, and I am using SUFFIX=TABT in the MAS file. For some reason, a space is being added between every character in the fields. Eventhough the actual field is 8 charcters long, I would have to change it to 17 to avoid the "data field larger than specified length" error.

Ex:
a DATE field in CSV file: "20090401"
DATE field outputed: " 2 0 0 9 0 4 0 1 "

Has anyone encountered this before?

This message has been edited. Last edited by: Kerry,

I don't get it. Is this a tab delimited or a CSV file? It can't be both, can it?

Can you post the master ?

I think that is the problem. I don't know how, but the CSV file is tab delimited. When I open it up in notepad, it really is delimited with tabs.

Here is part of the master:

FILENAME=STISADVPORT, SUFFIX=TABT, $
SEGNAME=STISADVPORT, $
FIELD=EXTRACTDATE ,EXTDATE , A17 , A17 , MISSING=ON ,$
FIELD=PORTFOLIO_ID,PORTID  , A13 , A13 , MISSING=OFF,$ 

Here is how the master should look according to the CSV file:

  
FILENAME=STISADVPORT, SUFFIX=TABT, $
SEGNAME=STISADVPORT, $
FIELD=EXTRACTDATE ,EXTDATE , YYMD, A8 , MISSING=ON ,$
FIELD=PORTFOLIO_ID,PORTID  , A6 , A6 , MISSING=OFF,$ 

Where does the data come from ?

Could it be unicode ?

Please read up on "SUFFIX" and the associated values associated with it in regards to CSV files. These masters (in the Tue Jun 02 2009 17:44 post) are for a tab delimited file.

quote:

I am trying to extract data from a CSV file which is TAB delimited

Doug, I think the file is Tab delimited.

dikanst, can you post some data ?

Yeah Waz, Therein lies the first dilemma: "I am trying to extract data from a CSV file which is TAB delimited" ... "but the CSV file is tab delimited". So, it's not a CSV file... Let's wait for some data.

Doug, I completely agree that the terminology is wrong, Tab delimited is Tab Delimited. Calling it CSV can be misleading.

Are you able to open the csv up in something other than notepad, say a hex editor, to see the actual contects of the file ?

The fact that the output contains a space between each character is very strange however what is also strange is that (if you wrote it accurately) the output shows double quotes as part of the output.

This is incorrect for a TABT file as data elements should not be wrapped within double quotes.

What you have in WebFOCUS terms is a DFIX format file.

Write your master as DFIX and you will probabley find that it works.
You can find out how to write it in the dev studio online help

I have seen this before. Reason in that particular case was that the text (or csv or tab delimited) file originated from a system that used a somewhat different text encoding scheme. In my case it was a unnicode enabled ms sql server 2005 system, that produced a file suited for unicode access. Which uses two bytes for every character. I had to use some file conversion program to get it right.

Thanks guys, I guess there's just something wrong with the format of the data. I converted it to a normal tab delimited text file, and everything works correctly.