DSXchange

Currently your DOS command is "d:\progra~1\winzip\winzip32.exe -min -e c:\File_Name.zip c:\ ", I wanted to have this command force some sort of text into the buffer, so I wondered what would happen if you changed that to "d:\progra~1\winzip\winzip32.exe -min -e c:\File_Name.zip c:\ | dir"

When you do an explicit conversion, i.e. "StringToDecimal" the implicit conversion warning will not appear. If you are 100% certain that your implicit conversion are correct you can always deprecate the warning message using the Director log entries and either turn them to informational (recommended...

What happens if you force some output, i.e. add "| dir" to your command line?

Columns not passed to the next stage are dropped. They are no longer present in the data stream and thus consume no memory resources or other overhead. They cannot be re-instated further downstream unless they are created as new columns and in that case they won't have any data in them. Dropped colu...

The consensus in the past was that this is not possible, DataSets are not like tables or hashed files.

Columns that are not copied over explicitly to the next stage are dropped unless you have RCP turned on.

If you change your commit frequency to 1 so that each row is immediately committed and the error goes away then you have answered the locking question positively.

The caveat is that 7.5x2 (for PX) can run on XP PRO but it is not officially supported.

The link "direction" can be retrieved using the DSGetStageInfo() routine call and using the DSJ.LINKTYPES code.

If it is a dummy file then create a dummy key column as well. A hashed file needs at least one defined key column.

All you need to create a hashed file is an output hashed file stage. This applies to any sort of hashed file, be it a "normal" one or a "dummy" one.

You cannot use DataStage to read .pst files. The format is not only quite complex but oftentimes these files are encrypted as well. I did a quick google search and there are products out there that will convert .pst files to text ones; or you could even try an Outlook export. Once you have a text fi...

The COUNT() function might not be the best since it finds any and all occurrences and not just those at the end of the string, but you can use the RIGHT() function to extract the last 2 or 3 characters in a string and do your comparison on that.

If you are doing mainly a 1 table to 1 table migration with identical columns then you can create generic PX jobs to do this for you using what is called "runtime column propagation" and won't need to create that many jobs.

You can try the following to see what might be happening in your environment: Enter the TCL environment in your project and execute the command "VLIST DSD_BP DSD_RUN.B". This command will show the pseudocode source for your job and the first column will be the line number. If you could post the outp...