Data360 Analyze

Hi 

I have a file that contains multiple records which are of a differing recordtype. This means that certain record types have more columns than another.

The delimiter is a pipe.

I tried to use the delimited node which reads the first record but errors the remaining ones.

I cant name headers as the first record doesnt have the same as the other records.

Ultimately I wish to load the whole file and then split on record type which is the first column in the file.

Any advice appreciated.

Thanks

Martin

Hi Martin,

You can configure the CSV/Delimited node to assume each record in the file is a single field (e.g. named 'Data'). You can specify the name for the dummy field in the 'FieldNames' property.

The node would be configured with a Delimiter character that will not exist in the actual file data e.g. '¬' or a control character '\07'.

If the file data has a header record this can be skipped by setting the 'SkipHeaderRows' property to a value of '1'.

Once the data has been imported you can then split the data on the leading characters of the 'Data' field and then use further CSV/Delimited nodes that are configured to source their (pipe-delimited) data from an input field.

You may also be interested in this article:

https://support.infogix.com/hc/en-us/community/posts/360028941493-Expanding-a-List-into-Multiple-Columns

Attached files

Read_Records_w_Different_Num_Fields - 2 Sept 2021.lna

Example Data file used with the example data flow:

Attached files

Pipe_Delim--records_w_Different_Num_Colums.txt

Many Thanks Adrian this worked a treat! hope you are well!

I did have multiple prefixes to output so used a superceded transform node to split into multiple output pins

Martin