What package is best to create a table for a presentation that shows the difference between the means of two variables? I've been using gt_summary but had trouble generating summary stats for multiple variables so used vtable instead which is great for a newbie, but I can't find a way to display the difference between means. Thank you!
Related
during development, there are many codes/SQLs can be re-used, which is used to implement some kind of business logical. To avoid re-inventing the wheel, I propose to create a package, which contains these code/SQL/functions; then, any new code can call the functions/codes in this package and no need to write these frequently used codes again.
I wonder if the design is a decent solution for the scenario: avoid re-inventing wheel and decrease the lines of the code
PL/SQL is no different than any other programming language as for code reusability. If you can reuse - create apropritate function that can be called later.
In PL/SQL you just need to know how to create functions in case if there is huge data amounts in return. So you either need to use temporary tables or custom table types or create views or materialized views.
I have came across an ODI project and there seems to be a lot of user defined KMs and I don't understand why they were modified? Is there any particular scenario where existing KM doesn't work?
There are a lot of reasons for writing your own KM or modify the existing ones, for example:
log in your own paths/tables;
read metadata from flex fields (Metadata like: default values for some columns, base table name used for temporary tables, type of load: full/incremental etc);
make different transformations/stage steps, different from the standard KM's;
customize your CKM: make error table - where you see your rows in error, make correct result table and so on;
for modifying KM you may have want to save the temporary tables with your own standard and so on.
The benefits of writing KM's is that the limit is your imagination (or almost). You can do plenty of stuff. The standard KM's are very good but there are some moments when you reach the limit with them and from there you should create your own.
Hope that this helps you.
I've looked through GotReportViewer.com in some detail, but I just can't find enough of a clue to really work out how to do this.
I need to effectively join two datatables (client-side) on a primary key, and show some information from one table in one area, and other (listed) information from the other in another.
From what I read, it's possible (though I haven't managed it yet) to join the tables together to form a third, and use filters to achieve this, although it seems like a lot of work when you're building the datasources dynamically as I am.
Is it possible to generate two datasources that share a key (one implementing the 'many' part of the 1:many relationship), and create a master-detail report that uses both?
If there are any simple tutorials (C# or VB would be fine if I need code) for this I'd appreciate a pointer. There are several unanswered questions here on similar topics to this, which is a worry!
TIA
I gave up on Microsoft's reporting tools, switched to Crystal, and life became a lot easier. Importing an XSD generates a data source, in which tables can be easily linked.
Job done.
I'm looking for a means of somehow exporting or visualizing the parse tree of an arbitrary PL/SQL object in Oracle in such a way as to allow comparison between the parse trees of two similar objects. Is this possible?
In theory this should be possible (and will need lots of time to implement!):
PL/SQL is parsed and stored in the DB as a tree-like structure PLUS m-code. The structure is some variant of DIANA plus IDL which in turn is something Ada uses to store the parse results too (PL/SQL is rather similar to Ada). For reference see http://download.oracle.com/docs/cd/B14117_01/appdev.101/b10807/e_limits.htm and http://cs.felk.cvut.cz/10gr2/appdev.102/b14261/tuning.htm
Another option is to build your own parser - a starting point is http://it.toolbox.com/blogs/oracle-guide/building-a-plsql-code-parser-using-plsql-part-1-29704
A very helpful resource is http://www.blackhat.com/presentations/bh-usa-06/BH-US-06-Finnigan.pdf - you will find lots of information on how to extract DIANA-/IDL-information from an Oracle DB which should help you achieve your goal.
I have one "big" TOracleDataSet which I can't change 'cause it's using in many different parts of huge project. I want to add just one record to this dataset for using in another grid. The solve way I see it is create another one oracle data set which will combine wanted record and another ones from "big" dataset. In other words, "small" dataset includes "big" dataset.
Try this maybe?
TxQuery Project
This was a commercial project at one time, but the auther was convinced to release it opensource.
TxQuery component is a TDataSet descendant component that can be used to query one or more TDataSet descendant components using SQL statements. It is implemented in Delphi 100% source code, no DLL required, because it implements its own SQL syntax parser and SQL engine.
That quote was taken from the page of the current maintainer, I believe, Chau Chee Yang .
Either this or maybe TClientDataset might be your best options.