DSXchange

The same is true in most databases. Mixed case identifier names and identifier names that contain non-alphanumeric characters (except underscore) need to be quoted.

The value you give to @ID in your UPDATE statement must be the same string (the name of the sequence) that you give in the KeyMgtGetNextValueConcurrent() function. Display the contents of SDKSequences to see what I mean.

No.

Because of the tags, XML files are substantially larger than sequential files for the same volume of data. For that reason I prefer to use the latter. However, XML can transport the metadata along with the data in the same file, which may prove useful in some circumstances. Apart from that, I think ...

There's no such thing as DataStage MQ. MQ Series is a separate product in the IBM WebSphere group. Plug-in stages for DataStage exist to allow you to read from and to write to MQ Series queues. Depending on your DataStage version, the MQ Series server may need to be co-resident with the DataStage se...

Many of the error codes can be resolved via a lookup on the SYS.MESSAGE hashed file. However you need a six digit key. Fmt(ErrorCode, "R%6") Apart from that most of the others are not retrievable using software. Some are documented in header files, but searching these has to be by ...

(a) a "list of things" loop in a job sequence (b) a server job that iterates through the list (by reading the file) and invokes another job to process the data using UtilityRunJob() function from a Transformer stage The second approach is necessary if you have a version earlier than 7.5, which is th...

I haven't had the chance to check yet. It may be that you need to quote the entire thing (remember parameters with spaces?).

The version 8 product spent over a year in beta testing so that as many pitfalls as possible would be eliminated. Word reaching me is that that part of the upgrade process (upgrade of server jobs) is as simple as it appears - indeed, they update in place, you don't even need to import them. Though o...

Did you try the approach I outlined in my second paragraph?

They are stored in separate fields in the record in DS_JOBOBJECTS, so my best guess would be a line terminator character or field mark. Or it just may not be possible. Any reason for preferring user-defined over generated SQL?

It depends on a number of factors. For example, if you are reading via ODBC then the limit is 400 columns by default, though this can be changed by specifying MAXFETCHCOLS in the uvodbc.config file at least for that data source. You might also need to increase MAXFETCHBUFF, which defaults to 8KB per...

Do you know in advance how many rows there will be? Is the change condition a change of case, or the appearance of the space character? That is, could you read into three columns as a space delimited record? If you could split the file into three based on the space, then you could write the rows int...

No. Because I would have to guess, even what scripting language you are using. Type dsjob alone to get the overall syntax. Then type dsjob with any of the options to get more information, for example dsjob -linkinfo to learn the remainder of the syntax to be used with this particular option. You nee...