DSXchange

I hope i should be more clear about my requirement. I am reading the lines from seq. file and group of lines forms a record in table. If line contains 'ABC' then that is starting of new record, which means we have to commit the group of lines we have inserted so for in table. Eg: i have 25 lines in ...

Hi I have a scenario where i need to commit the database insertion only if certain condition meets. Eg: i am reading a file line by line and started inserting the same into a table and if the line containe 'ABC' then commit the records inserted sofar and start new insertion. Any help would be great.

You Can use soem thing like Job1: Source Table-->Tranformer-->Sequencialfile(not hash file) Job2: Sequential file -->PIvot stage -->hashfile You can even combine Job1 and 2 in a single job. In you sequential file if you have something like a:1,b:12,c:123 ... and if you know the max number of fields ...

In my job the source file path has been defined as #SrcDir#/#SrcFile#,
Job has individual parameters like SrcDir, SrcFile etc. More over in the director logs i am seeing my hardcoded file name rather than passed file name missing note.

The spaces are getting deleted when i am editing this message with dummy values. I tried this as well still it fails because job can not get the file name from commadn prompt. ./dsjob -server Server1 -user User1 -password password -run -param SrcDir=/dvl/etl/projects/Project1/Source -param SrcFile=P...

I tried with both single Quote and double quote options for the params.I have 4 other paramters which i wanted to take the default ones that i defined in the job.(So i am not passing them) ./dsjob -server Server1 -user User1 -password password -run -param "SrcDir=/dvl/etl/projects/Project1/Sour...

Chulett - It has space aftre warn .
kduke - I tried as you suggested, but still job fails.

Any other suggestions?

Hi All I am trying to execute the sequencer from command prompt and i used following command. ./dsjob -server Servername -user username -password password -run -param SrcDir="Dir path" -param SrcFile="FileName"-warn 5000 Project Jobname The job has been kicked off, but failed bec...

I am trying to design a job which does following. 1) Get source flat file(havingng approx. 1-2M records) and separate them into 2 files (file1,file2) based on category. 2) key columns are same in these files.Load file 1 records to table1. File 2 is having all comma delimited fields(variable number o...

Hi I have a text file with approximately 20 variable length columns max. First 10 columns are always present and the next 10 columns are variable i.e i can get 14 columns in a row and 18 columns in next row etc.. when i am defining the sequencial file read i defined 20 columns by default and want to...

Ok Do like this. Procedure 1: 1)Job1 - read the file with comma(,) delimitor, in the transformer pick all the fields except | separated column(remember all pipe separated field will be read into single column here, so make sure to give enough length for this field) write these field to one file -fil...

Hi All, Can some one through some information on how i can control the datastage when to commit the transcation. Example: Lets say we have flight table and passenger table and if i want to commit the data for each flight with all the passenger information at once(2 different tables, one transaction ...

I end up being using TRIM function to delete the leading zeros and it works fine now. But if i define SQL data type as smallint,integer it should work but appenrently it did not.

Hi All, I am trying to read data from flat file and looking for record existance in hashedfile(which is populated from DB) . My hashed file contains a numeric field and i am getting it as 3, 4 etc. From the flat file i am getting input as 003, even though i defined the field as smallInt. Because of ...

I should be able to solve the issue, the date format was the culprint.

Thanks for the assitance.