Originally posted by: SystemAdmin

Hi,
I am very new to WTX(started working a day ago).
I have to create a map that will pull data from CSV file and put it into a XML file.
My first problem is how do i make a tree type for a CSV file, a row of which is pasted below.
"
r1c1,r1c2,r1c3,r1c4,r1c5,r1c6,r1c7,r1c8,r1c9,r1c10,r1c11
r2c1,r2c2,r2c3,r2c4,r2c5,r2c6,r2c7,r2c8,r2c9,r2c10,r2c11
"
I have created the mtt for XML from the schema.
With the current csv tree type,when i run the map, i get a message "Input valid but unknown data found"
The output xml that is created has proper structure but no data.
I would really appriciate help in
1. Creating the tree type for csv file.
2. How do I run the map automatically whenever the csv file is created.
3. How do i provide the name of input file/output file at runtime.
4. Any pointers to relevent documentation.

Thanks.

#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: Kepa_Screen

1- You must create 10 text fiels. Then create a group called line. Populate "line" with the 10 text fields ( infile, separator, literal ',') . Create a new group called file and populate it with group line. Use terminator literal <NL>.

2- For this, you must use event server. You must create a new system, put the map in the system, and put a "trigger" at input card for .csv file .

3- You can do this in several ways, usign modifier -O for output file, if using event server, you can use wildcards.
( i.e.
inputfile = 'abc*.txt'
and outputfile = 'def*.log'

imagine
real input filename is 'abc123.txt'
then real output filename would be = 'def123.txt'
)

4- I think you must read all documentation ( Mercator on.line library ) that was installed with WTX
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: SystemAdmin

Its working but still not ok. I am sure there is some problem with my mtt files.
Could you please have a look at them?
In the output xml, just b4 the first character of the input file is entered, i am getting these junk characters "ÿþ". and the end of line char is not recognized..as the last record is coming as "r1c11 r2c1". r2c1 is Row2 Col1.it shoud have gone in the next node.

Thanks for your quick response.
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: Kepa_Screen

Yes, sorry for my mistake.
You must tell how much lines are in the group file.

This could be "1:s", you must open File group, then on line, with right button, "SET RANGE" --> min: 1 - max: s

So in your map, you must use a Funjavascript:postPressed()
Submitctional Map in order to treat an unique record.
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

hi Kepa,
This helped to a greater extent.
The next problem is that the output xml is getting generated only for the first record.
A snapshot showing i/p, o/p cards and the i/p and o/p results is attached.
I have a feeling that there might be some problem with my output mtt file.

Also could you please point me out to some tutorial that talks about creating type trees.

Thanks for your assistance.
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: Kepa_Screen

Hi again:

You must use a functional map in order to treat more than one line in your file.
The first thing to do is to put SET RANGE 1:s to Line group in File Group

File
Lines (1:s)

Then, at the output card rule, you must put something like :

File
Lines (1:s ) = F_treat_records(Line:Input)

Then create the functional map with the wizard, in order to pass a single record to the functional map.

Then in functional map output, you only have to map field by field.

I attach another picture showing this ( in my case, output type tree is the same than input type tree ).

Picture1 : how to create a functional map
Picture2 : you only have to map field by field, because at functional map level you only have one record.


#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: Kepa_Screen

here is the second picture
#DataExchange
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender

Originally posted by: SystemAdmin

Hi Kepa,
I tried whatever you suggested. and now my maps looks similar to snapshots that you had posted...but still i am getting only 1 record populated in the xml output file.
Could you please have a look at the snap of the functional map(pict1) and execution map(pict2) and help me figure out what is wrong.
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: SystemAdmin

#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: Kepa_Screen

Could you send to me your input and output .mtt, and also an input file ?
I can see your mapping at first picture, so i could try to make a similar map.

It's very important to put AUDIT Logs ON, also TRACE on, when you are developing, so you can see where it's wrong.

If you are trying to map a collection of records from a file, then the output should be another file, just another collection of records...are you sure that your output card type tree is also 1:s ?

You can send to me your i/o .mtt and the input file ? ( kampanita(dot)gmail.com )
I'll try to help you.

#IBMSterlingTransformationExtender
#DataExchange
#IBM-Websphere-Transformation-Extender

Originally posted by: vickey

Arora,

Nice that Kepa is helping you. I hope this issue is resolved.

I would like to let know one thumb rule in Mercator. when you want to iterate objects more than once then the input range should be either 0:S or 1:S. and the output range should also be either 0:s or 1:S.

In the pic you attached if you closely look you are using a fucntional map where the range of the output is 1:1.

The schema you imported to create the type tree is incorrect. Correct you schema to loop the Segments and then reimport.

I have attached the pic where the range should be 1:S.

Let me know if you any more questions.

-Ramakrishna.
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: SystemAdmin

hi kepa/krishnan,
I modified the outpot mtt file as adviced. Things now seem to be working ok..as long as there is only 1 rec in the input csv file. If there are more than 1 line, I get an error trace like

"
(Level 2: Offset 62, len 5, comp 12 of 12, #1, DI 0000000C:)
Data at offset 62 ('r1c12') was found to be of TYPE
X'0010' (Path Field Data).

(Level 1: Offset 0, len 67, comp 1 of 1, #1, DI 0000000D:)
Data at offset 0 ('r1c1,r1c2,r1c3,r') was found to be of TYPE
X'0003' (Line Data).

(Level 1: Offset 67, len 0, comp 1 of 1, #2, DI 0000000E:)
COMPONENT number 1 of TYPE X'0002' (File Data):
occurrence 2 is optional and has no content.

(Level 0: Offset 0, len 67, comp 1 of 0, #1, DI 0000000E:)
Data at offset 0 ('r1c1,r1c2,r1c3,r') was found to be of TYPE
X'0002' (File Data).

INPUT 1 was valid but contained 138 bytes of unknown data at the end.
"
I am using NL as the record terminal char. I tried using <CR><LF> also, but that also resulted in the same type of error msg.

Could you please suggest if this is some prob with my input CSV file or something wrong in my mtt files?

#DataExchange
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender

Originally posted by: vickey

Hi Arora,

Can you send me the components. So that I can correct it and send u back the components.

you can send me the Maps, Input & Output Type tree, XML schema and Input files.

You can zip them and send them to my email ID vikashrk@gmail.com.

-Ramakrishna.
#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: janhess

Have you defined your file as having multiple records?
#IBM-Websphere-Transformation-Extender
#DataExchange
#IBMSterlingTransformationExtender

Originally posted by: shilpaAllam

INPUT 1 was valid but contained 138 bytes of unknown data at the end.
"
I am using NL as the record terminal char. I tried using <CR><LF> also, but that also resulted in the same type of error msg.

Could you please suggest if this is some prob with my input CSV file or something wrong in my mtt files?

 

I have similar type of error, can you provide me a solution on it?

#IBMSterlingTransformationExtender
#IBM-Websphere-Transformation-Extender
#DataExchange

Originally posted by: PaulBrett

Please start a new post.

Thank you.

Paul

Follow me on Twitter

#DataExchange
#IBM-Websphere-Transformation-Extender
#IBMSterlingTransformationExtender