Up: Component summary

CSVSplit

Divides the content of the given CSV to an array of multiple CSV files. Possible partition rules are:

• Split file in to N parts, equal number of rows, by:
• saving the first M lines as the first output, or
• saving every Mth line as the first output
• Split file by regular expression group match
• Split file by unique label column

The component acts as the inverse operation to CSVListJoin.

Inputs

Outputs

Parameters

Test cases

Test network

import anduril.tools._
import anduril.builtin._
import org.anduril.runtime._
 
object CSVSplitNetwork {
// CSVSplit testcases
 
/** Define a CSV */
val input = INPUT(path="csv.csv")
 
/** Simple splitting to two parts, filling the first file fully */
val split_order_first = CSVSplit(in=input)
 
val rejoin_splitted = CSVListJoin( in1=split_order_first.out(1),
                                   in2=split_order_first.out(2),
                                   fileCol="")
 
/*  Now,  input should equal  rejoin_splitted */
 
/** Sparse filling */
val split_order_sparse = CSVSplit(in=input,
                                  order="sparse")
                             
/** Split file by unique values in a column */
val split_column = CSVSplit(in=input,
                            labelCol="Well")
 
/** Split file by unique regexp in a column */
val split_regexp = CSVSplit(in=input,
                             labelCol="File",
                             regexp=".*_(.*).tif")
 
/** Join two of the outputs of the splitting. Note, the keys follow 
 *  unique values in the labelCol */
val rejoin_regexp_splitted = CSVListJoin( in1=split_regexp.out("15ADF48D5BC479534681"),
                                          in2=split_regexp.out("E7359B7DCAADD496A950"),
                                          fileCol="")
 
OUTPUT(split_column)
}