topic Re: Merge rows (cross join) in Talend Studio

Merge rows (cross join)

_AnonymousUser — Thu, 08 May 2014 11:50:36 GMT

Hello,
I have the following rows in a file:
Ultrahouse|41|40|1
Megahouse|42|41|2
Superhouse|43|42|3
House|44|43|4
Now i want to use Talend to merge the 4 rows into one like this:
Ultrahouse>Megahouse>Superhouse>House
Thanks
Mr Marui Lamakamapa
Head of Talend Integration

Re: Merge rows (cross join)

Anonymous — Thu, 08 May 2014 11:54:21 GMT

Use tMemorizeRows component...
tFileInputDelimited-->tMemorizeRows-->tJavaRow-->tfileoutDelimited
Refer to post - https://community.talend.com/t5/Design-and-Development/How-to-pick-up-value-of-one-column-where-another-columns-value/td-p/90483 which has similar logic but for different purpose.
Thanks
Vaibhav

Re: Merge rows (cross join)

Anonymous — Thu, 08 May 2014 12:29:52 GMT

Hi ,
Use tDenormalize component.
Thanks,
aishu

Re: Merge rows (cross join)

Anonymous — Thu, 08 May 2014 12:34:30 GMT

Hi ,
tDenormalize component Link: http://www.vikramtakkar.com/2013/01/understanding-tnormalize-and.html
Thanks,
kumar

Re: Merge rows (cross join)

_AnonymousUser — Thu, 08 May 2014 12:55:25 GMT

Unfortunaletyl I dont work with tDenormalize it seems the cross join dont work.
With memorize row I tried but it doesnt work.
Mr Marui Lamakamapa

Re: Merge rows (cross join)

Anonymous — Thu, 08 May 2014 13:17:58 GMT

tMemorizeWorks, you will have to do some tweaking... check the screenshot, used following expression in tjavarow with first column to memorize for 4 times.
output_row.a = a_tMemorizeRows_1+">"+a_tMemorizeRows_1+">"+a_tMemorizeRows_1+">"+a_tMemorizeRows_1;
Vaibhav

Re: Merge rows (cross join)

_AnonymousUser — Thu, 08 May 2014 14:06:49 GMT

Now I realize solution
Thanks to sanvaibhav I was able to do with tMemorice
Thank you alot sanvaibhav
Well support in this forum!
Mr Marui Lamakamapa

Re: Merge rows (cross join)

Anonymous — Thu, 08 May 2014 14:15:34 GMT

create a job like below
read yourinput-----main-->tDenormalize----main---youroutput
now you have to do some changes in tDenormalize components schema take only one column as above your first column for de-normalization.
and map only column to your out put it will give you expected result i have just tested with your given input.