DSXchange

I have a variable length file and I am using a parallel reader as an input stage for the variable length file. It is a must to specify the record delimiter within this stage.
Now I do not have a record delimiter as none of the records have a newline at the end of the record. What should my record delimiter be as I have tried null and none. I am out of ideas. Please help me.

MAY BE U SHOULD TRY FIXED WIDTH OPTION

What is the file type?

Do you mean it has a variable record lengths within the same file?

Shai

psriva wrote:Now I do not have a record delimiter as none of the records have a newline at the end of the record. What should my record delimiter be as I have tried null and none. I am out of ideas. Please help me.

What character(s) are used to distinguish the different records in your file? I.e. when you open your file with a text editor, how do you know when the first record ends and the second one starts?

Rob W

Pipe delimited.

Now when I specify the record delimiter as pipe (|), I get another error on one of the field saying " Delimiter for field 'name' not found".

Th following is my record schema:

record
{final_delim=none,delim='|',charset='ISO8859-1'}
(
account:string[max=10] {default='0'};
name:string[max=10] {default='a'};
location:string[max=10] ;
)

And below is the data file:

104246|abc|usa|104247|def|uk|104250|ghi|india|

Please help me write the right schema for this.

If you look at the data and if it is a complete record that you have provided, then you need to set your Final Delimiter as "|"(pipe) which is obvious from the record.

psriva wrote:Pipe delimited.

Now when I specify the record delimiter as pipe (|), I get another error on one of the field saying " Delimiter for field 'name' not found".

Th following is my record schema:

record
{final_delim=none,delim='|',charset='ISO8859-1'}
(
account:string[max=10] {default='0'};
name:string[max=10] {default='a'};
location:string[max=10] ;
)

And below is the data file:

104246|abc|usa|104246|abc|usa|104246|abc|usa|

Please help me write the right schema for this.

This data has '|' both as field delimiter and record delimiter. Can you try and correct the data? change record delimiter to a line termination character after location. something like:
104246|abc|usa
104246|abc|usa

You cannot read the data in the esixting format.
104246|abc|usa

I tried the same schema with sequential file stage and it runs fine.

What was the stage you were using earlier?

I am using the parallel reader stage currently.

There is no such thing as the "parallel reader" stage. I presume that you are using a Sequential File stage.

If your data have no line terminators then each row must be of the same length or of a known length described in the data. Otherwise there is no way that any utility can accurately read the data.

Have you tried setting the Final Delimiter property (not the same as the Record Delimiter property)?

I am using a custom stage developed for use within my company.

That was vital information omitted from your original post. I can only suggest close scrutiny of that stage's source code to determine how records of this type ought to be handled.