DSXchange

Hi all;

i am getting error like this

job1..Sequential_File_0.DSLink2: read_delimited() - row 1, too many columns in record.

what might be the problem. please suggests me


Initially i had 47 columns in source (tab delimited file). when i view the data using sequential file (source), i can able to view the data. but my 'project 'architect' changed the input file, to 49 columns. so when i view wthe data for this i am getting above error.

Thanks for ur help

You will need to include the 2 columns in your column defn tab.

U just need to reimport the metadata or add 2 more columns in your input stage and the problem would be solved

Hi sainath ;

Sainath.Srinivasan wrote:You will need to include the 2 columns in your column defn tab.

before i imported table defintion again with 49 fields. any way i got it.

the problem is with line termination

Thnaks for ur help

The problem is not with the line terminator..its with column delimiter. View your data and see if you have this delimiter anywhere in the record as part of the data...

Check the number of columns in your source file. Maybe your source file did not change to 49 cols.

H isainath;

Sainath.Srinivasan wrote:Check the number of columns in your source file. Maybe your source file did not change to 49 cols.

I got it when i changed the "line termination from uinx style to DOS style"

thanks for ur reply;

thanks

All,

I got this message when trying to read a text file as just one long field:

RemoveSpecChars_BaseProduct..PIXP81_FTPSourceFile.clean_out: read_delimited() - row 94, too many columns in record

Now row 94 contained commas (which were not meant to be delimiters), hence the "extra" columns.

The cure was to change the following in the Format Tab:

Delimiter 000
Quote Char 000

I believe this means that there is no delimiter processing (although it is a "delimited" file). Anyway, the above fixed the problem.

Regards,

Brian