Cannot bind 2 schemas to a tFileOutputMSDelimited

[Anonymous]

Hi,

I have a specific problem with a Multi Schema Delimited File.

I receive a Multi Schema file, and i need to make some transformations to the data using tmaps (1 for each schema).

This works fine but the problem is that i cannot bind them to their initial shape in a tFileOutputMSDelimited

Here is a picture to illustrate my problem : 

(in red the bind i can't make)

My data is set like following :

E;X;Y;Z

P;X;Y;M;T;Z

P;X;Y;M;T;Z

P;X;Y;M;T;Z

E;X;Y;Z

P;X;Y;M;T;Z

P;X;Y;M;T;Z

...

And i need to keep the same order for my output file.

That restriction forbids me to do the following thing :

Here are my settings for my tFileInputMSDelimited :

If you have any suggestions, it's welcome.

Best Regards,

Pierre

[akumar2301 · Accepted solution]

You have to use tHashinput/output component . because 

https://help.talend.com/reader/~V1UYTyXpjBKjP5PCMy4EA/khJeOaSeNMnFt0o~DMx6qQ

Attaching an example job how the order can be maintained. Change path accordingly.

Hope this is useful.

tMultidelimiter_0.1.zip
in.txt

[Anonymous]

it does not work, 

i only have the following outputs :

E;X1;Y1;Z
E;X1;Y1;Z

P schema lines are missing

Also, i am not allowed to use temporary memory for performance reasons i guess.

[akumar2301]

aah it should work .

It might be because Schema for P type might have been removed .

Can you please redefine it and check. 

I am getting correct output.

Instead of Hash Component , you could write these in csv files and read it.

[Anonymous]

Ok,

Thank you for your support, i found an alternative solution to my problem.

I set my sequencer in tJava

Input each lines of the Multi Schéma file

tMap1 : - I build my schemas sorting lines with substring function.

             - I set the sequencer to each outputs to obtain at the end the result : 

1|E|X|Y|Z

1|P|A|B|C|D|E|F

2|E|X|Y|Z

2|P|A|B|C|D|E|F

...

I make my transformations to the data at tMap2 tMap3 tMap4

Fetching correctly the datawith tFileOutputMSDelimited with sequencer "Id" as Key Column

If this helps someone in future.

Pierre