DSXchange

Hi All, The logic was not working in the real data since there was an error in the source string itself. Now the results are perfect as desired. Thanks a million to everyone for helping in this issue. Special thanks to Priyadarshi, Battaliou, Craig & Ray. :D I'm marking the topic as resolved. -T...

agpt wrote:Tony,

I think you should use +6

Hi AGPT,

The logic used is length of the variable till the "=" sign which in this case is 7 (Z10HFC=).

Thanks,
Tony

Hi Priyadarshi, Thanks a lot for all the help. The output is perfect in the sample data. The only thing which remains is that when I use the logic in real data, all the output rows are zeros. The actual variable name is "Z10HFC" and the derivation I've used is If Index(DSLink1.SDFTSW,'Z10H...

How about trying the "DateTimeStampTo OraOciWithTime" or "DateTimeStampTo OraOci" routine?

If the Database Server is of low configuration, then the only other option in your hands that you can try is to make the job run in parallel with appropriate partitioning methods so that the run time can be minimised.

-Tony

Hi Ray,

Learnt one more new thing today.Thanks for that. :D

Shah,
The path for the Basic Transformer is

Repository---->Stage Types---->Parallel---->Processing--->Basic Transformer

-Tony

Hi Ray, Thanks a lot for the reply. Actually I'd mentioned the column missing factor and the complexity in the first post. Col A,Col B,Col C,Col D 1234,ab2f,pazz,"&Col E=ZZ&Col F= &Col G=5478&Col H=100.00 2345,cd3f,ddzz,"&Col F=IN&Col G=9801&Col H=250.25&Col...

I even tried customizing the Palette but could not find the a BASIC transformer stage in Processing(Parallel) nor in Processing(Server).

The transformer available under the processing tab of server job is conventionally referred to as "Basic" Transformer.

-Tony

Thanks a million Craig for your support. Now the issue is partially resolved. :D I'm able to get value "ZZ" wherever Col_D contains Col_E as the first variable between the delimiter. The part which is still unsolved is that if Col_E is not present (&Col_F=IN&Col_G=9801&Col_H=25...

Hi Craig,

Thanks a lot for the inputs.

I'm bit confused on the final logic since I haven't worked before on embedded functions. Can you kindly show me how to make the result of first function to take "=" as delimiter and get the final output?

Thanks again,
Tony

Hi Craig, The version we are using in this project is 8.1.1 with no fix packs. Actually the original column names doesn't have spaces. In the sample data file I'd created the columns have underscores only. Sorry for being unclear regarding the same. The approach I'm trying right now as suggested by ...

Hi Priyadarshi, I changed the Col_E to Col E and then ran the job. The derivation is if index(DSLink5.Col D,'Col E=',1) then field(DSLink5.Col D [index(DSLink5.Col D,'Col E=',1) +5,99],'&',1) else 0 Still the result is zeros for all the rows. I've kept "+5" because the length of 'Col E...

Hi Priyadarshi,

antonyraj.deva wrote:
{if index(Col_D,'Col_E=',1) then field(Col_D[index(Col_D,'Col_E=',1) +6 ,99],'&',1) else 0}

Do you really understand what this logic trying to do?

Truly I understood very little of the logic...

Thanks,
Tony

Hi Priyadarshi, {if index(Col_D,'Col_E=',1) then field(Col_D[index(Col_D,'Col_E=',1) +6 , 99 ],'&',1) else 0} In the above logic what exactly is the use of aruguments "+6" and "99"? I guess if I can understand these arguments and give appropriate values for them then the issu...

Hi All, After removing the quote around the column name (Field(Col_D,"&",2)) and running the job the output returned was "Co" (the first two letters after the delimiter) for all the rows. While trying to impart the index logic {if index(Col_D,'Col_E=',1) then field(Col_D[inde...