"[SOLVED] Importing: text with fixed length attributes"

Ugo

Hello,

I have looked through the import components and Googled but cannot find a way to
read in a simple text file with fixed length attributes. How can one do this?

Apologies if this is a "no brainer" but I could not find a simple direct way to do this.

TIA,
Hugo

Skirzynski

Hey,

You probably need the "Read CSV" operator. This operator can read a structured data set from a text file. Use the wizard to import and configure this operator correctly. For instance it is important to specify the separator so the operator knows where the value for an attribute begins and stops.

Best regards
Marcin

Ugo

Hi Marcin,

I already had a look at the CSV reader but it requires the use of a delimiter.
The file I have has no such delimiters. As an example, assume I have a line:

AABBCCCC

In this case I have 3 attributes with lengths 2, 2 and 3 respectively.
The attribute value would be AA, BB and CCCC. Note that no separator
exists.

Write now I am preparing AWK scripts to deal with this but I assumed
Rapidminer can deal with this type of data easily.

Thanks for the feedback.

awchisholm

Hello

You could read the text file using the regular expression "\r\n" to read each complete line.

Then use the operator Generate Extract to split each line into the required components using regular expressions.

Here's an example

< =“query_type”价值=参数的关键"Regular Expression"/>

The text file contains this

AABBCCCC
ABCDBBDS
ABDBQBDD
AASHHFGU

and the result looks like this

a1 a2 a3
AA BBC CCC
AB CDB BDS
AB DBQ BDD
AA SHH FGU

regards

Andrew

Ugo

Hi Andrew,

Exactly what I was looking for.

Thank you.
Hugo F.