topic Re: Multischema Input-XML output in Talend Studio

Multischema Input-XML output

Anonymous — Sat, 16 Nov 2024 13:14:31 GMT

I have a multischema input file. Each row is differentiated by a 2 byte identifier. I have created metadata for each schema. I am having difficulties proceeding further. How do I do validations like trailer is present or the counts in trailer are equal to the number of rows. How should I map and insert it in the database and produce XML's. Is there a sample project for the same? I am a newbie in the world or Talend

Re: Multischema Input-XML output

Anonymous — Fri, 22 Oct 2010 15:47:02 GMT

You can split the file into individual flows using tFileInputMSPositional.
Probably easier to then drop them into separate files e.g. header, trailer, type1, type2 etc.
Subjob 1 - test for missing trailer
To test if trailer exists you can then use tFileProperties>tFilterRow>tdie - in the filter test for a size of 0. If the size is zero then the file is empty and the tdie will kill the process.
Subjob 2 - Check counts
Use tFileRowCount on the detail file(s). (note the context Variable of "Count" for each one).
use tRowGenerator to create a single record with a dummy key (e.g. set to 1) and a count field (which is the sum of the file counts)
Input the trailer and create a similar record with same dummy key and trailer count.
Join using tMap and create a record with just the two separate counts
Use tFilterRow to check if the two fields match. If they don't then use tdie to kill the job.
Subjob 3 - Create XML
Use tFileOutputXML to create your output files.

Re: Multischema Input-XML output

Anonymous — Mon, 25 Oct 2010 17:29:48 GMT

Thanks Rick.
Can you tell me how do you use the tFilterrow component to test file size?

Re: Multischema Input-XML output

Anonymous — Mon, 25 Oct 2010 17:51:25 GMT

If you look at the schema for tFileProperties, you will see a field called "size". This is the size of the file.
So in tFilterRow add a new condition (green "+" sign)>Select InputColumn=size, Operator="equals" and set the Value column to 0 (zero).
Now go back to the canvas and connect the "Filter" output of tFilterRow to tDie. Change the error message in tDie to something sensible and the subjob/job will fail when the file is empty.
Rick

Re: Multischema Input-XML output

Anonymous — Mon, 01 Nov 2010 19:55:15 GMT

I have another validation to be performed.
I have two variables (A & B) in Table00
I have to check values in these variables.
Either A or B can have values. Whichever has values it should be inserted in a new variable C.
If both the variables A or B have values. it should give an error.
Thanks.

Re: Multischema Input-XML output

Anonymous — Tue, 02 Nov 2010 09:06:54 GMT

Hi gnaik,
What happens if both are blank? This could affect how you do this.
Regards
Rick

Re: Multischema Input-XML output

Anonymous — Tue, 02 Nov 2010 20:32:21 GMT

Hi gnaik,
What happens if both are blank? This could affect how you do this.
Regards
Rick

Hi Rick,
If both are blank the variable C is also blank.
Thanks,
G

Re: Multischema Input-XML output

Anonymous — Wed, 03 Nov 2010 07:42:13 GMT

Hi gnaik,
If you are doing this assignment in tmap I would recommend writing a code routine in Java, that way you can get the code to fail elegantly. If you haven't written one before then take a look at the string methods, they are a good place to start.
Regards,
Rick