DSXchange

Then it's not the same issue (the original is on Windows). Please begin a new thread. ODBC on UNIX works differently from ODBC on Windows.

You could investigate using the Schema File property with your Sequential File stage, if you can get an accurate definition for the aberrant file's structure.

You have a business rule problem; ambiguous dates. For example, what is the date of 20110406 ? Is it April 6th or June 4th? You must resolve that before you can do anything sensible with the data.

There is no limit to the number of parameters that a job may have. When you add environment variable parameters to a job, you can use special tokens for their default values. $PROJDEF means "use the default value defined for the project (in the Administrator client" $ENV means "obtain...

The status mnemonics are also defined in JOBCONTROL.H, so you don't have to include dsapi.h.

You can create a reject-handling link to capture those lines from the file that do not match the designated metadata in your output link from the Sequential File stage.

This is not the same problem and really warranted a new thread. Actually it really warranted a search, in which you would have found that you need to set the Record Delimiter String property to "DOS style".

The connector checks your design metadata with what's really out there. If there's a difference, you get this failure. You need to make sure that the metadata accurately represent the structure of the data.

There are no restrictions on the name (you don't need any particular suffix) and DataStage imposes no restriction on the size of the file. 32-bit operating systems usually impose a 2GiB size limit on files. Even 64-bit operating systems may need to have "large file support" enabled.

Use Matches operator in a constraint expression. If you might need signed integers, extend the pattern.

What date picture string did you specify with the TO_DATE() function? Was it "YYYYDDMM" or was it "YYYYMMDD"?

Welcome aboard. The correct answer to your question is "the right ones". Every place is different, has different BI requirements. So, in the requirements gathering phase, you're garnering corporate knowledge about what's needed in reports (which may be a subset of what's wanted in reports)...

Did you build the stage variable expression using the expression editor (selecting operands and operators from menus) or manually?

Incidentally, a sufficient expression would be Boolean expressions return 1 for "true" and 0 for "false".

No, it only preserves the space against the likelihood that it will need to be re-used. Every record is marked as "free space" and there are no rollback tools. (There are tools where a support analyst, for example, could undelete a single record, but nothing automatic.)

DataStage is almost identical from a user's perspective - the differences are minor, such as no SQL Server Enterprise stage on UNIX platforms (no surprise there). As Vincent noted, it's mainly going to be about your in-house skill sets and whether you need to access things like Excel without having ...