Skip to main content
Announcements
Join us at Qlik Connect for 3 magical days of learning, networking,and inspiration! REGISTER TODAY and save!
cancel
Showing results for 
Search instead for 
Did you mean: 
Not applicable

conditional unwrap during load

Hello,

does anyone know of a possibility to use conditional unwrapping during the transformation step?

I can only specify ROW conditions but need to be able to specify column conditions.

Right now i do it manually and not "automated":

[....]

  Unwrap(Col, Pos(Top, 6)),

  Unwrap(Col, Pos(Top, 6)),

  Unwrap(Col, Pos(Top, 6)),

  Unwrap(Col, Pos(Top, 6)),

  Unwrap(Col, Pos(Top, 6)),

  Unwrap(Col, Pos(Top, 6)),

[....]

What i need to do is unwrap after every 6th column if there is still data in following column.

Is there any way to do this?

My table looks like this:

abcabca
32123453213212213312

What i hope to get is:

abc
3212345
3213212213

.... and so on.

Thank you!

1 Solution

Accepted Solutions
Gysbert_Wassenaar

Sure. Use a generic load as I said. Or do something like five loads, one for each field and join the lot.

Table2:

load rowno() as row,

subfield(@1,': ',2) as Workstep

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Workstep' ;

join

load rowno() as row,

subfield(@1,': ',2) as Operation

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Operation' ;

join

load rowno() as row,

subfield(@1,': ',2) as Start

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Start' ;

join

load rowno() as row,

subfield(@1,': ',2) as Stop

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Stop' ;

join

load rowno() as row,

subfield(@1,': ',2) as Duration

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Duration' ;

drop field row;


talk is cheap, supply exceeds demand

View solution in original post

7 Replies
Gysbert_Wassenaar

Perhaps a transpose is enough to get the result you want. See attached example


talk is cheap, supply exceeds demand
Not applicable
Author

hmm i don't really get it.. i mean the result of your example is correct but i don't know how you get there..

My original table is already transposed to get the structure i provided.

right now it looks like this:

DATALOAD2_TXT:
LOAD left(filename(), index(filename(), '_')-1) as SN,
[Workstep:],
[Operation:],
[Start:],
[Stop:],
[Duration:]
FROM
[$(vFileName)]       (
txt, codepage is 1252, embedded labels, delimiter is spaces, msq, header is 1 lines, filters(
Transpose(),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Unwrap(Col, Pos(Top, 6)),
Remove(Row, RowCnd(CellValue, 1, StrCnd(null))),
Remove(Row, RowCnd(Compound,
RowCnd(CellValue, 1, StrCnd(equal, 'Workstep:')),
RowCnd(Interval, Pos(Top, 2), Pos(Bottom, 1), Select(1, 0))
))
));


The source file (txt) has the following structure:


Workstep: EM04=HMHD654036_NC=654036_NC=EDM_1

Operation: EM04=006110500002=0004=2

Start: 22.10.13,16:05:23

Stop: 22.10.13,16:12:01

Duration: 000:06:38

Workstep: EM04=HMHD654036_NC=654036_NC=EDM_2

Operation: EM04=006110500002=0004=2

Start: 22.10.13,16:12:20

Stop: 22.10.13,16:22:04

Duration: 000:09:44

and so on.

Gysbert_Wassenaar

Perhaps like this then:

Table1:

LOAD subfield(@1,': ',1) as Key,

    subfield(@1,': ',2) as Value,

    ceil(rowno()/5) as Index

FROM [$(vFileName)] (txt, codepage is 1252, no labels, delimiter is '|', msq);

Or maybe you want to a generic load instead. In that case add the keyword GENERIC in front of LOAD in the load statement.


talk is cheap, supply exceeds demand
Not applicable
Author

thank you, but this only displays the data correctly in a pivot chart.

My goal was to get the QV Table in the correct format.

Is there no unwrap solution?

Gysbert_Wassenaar

Sure. Use a generic load as I said. Or do something like five loads, one for each field and join the lot.

Table2:

load rowno() as row,

subfield(@1,': ',2) as Workstep

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Workstep' ;

join

load rowno() as row,

subfield(@1,': ',2) as Operation

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Operation' ;

join

load rowno() as row,

subfield(@1,': ',2) as Start

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Start' ;

join

load rowno() as row,

subfield(@1,': ',2) as Stop

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Stop' ;

join

load rowno() as row,

subfield(@1,': ',2) as Duration

FROM

[$(vFileName)]

(txt, codepage is 1252, no labels, delimiter is '|', msq)

where subfield(@1,': ',1) = 'Duration' ;

drop field row;


talk is cheap, supply exceeds demand
Not applicable
Author

thank you so much this worked!

could you kindly tell me why "delimiter is '|'" works and "delimiter is spaces" not?...

Gysbert_Wassenaar

delimiter is spaces will work too. You just need to add the : to the match string in the where clause:

Table2: 

load rowno() as row, 

@2 as Workstep 

FROM 

[$(vFileName)] 

(txt, codepage is 1252, no labels, delimiter is spaces, msq) 

where @1= 'Workstep:' ; 


talk is cheap, supply exceeds demand