I had a scenario today where I was loading a table and a particular column is known by multiple names in different source systems and thus to different people. In order to make everyone happy on this occasion, I wondered if I could create a normal column for one of the multiple names and then use virtual columns pointing at the normal column, for the other names.
I’m aware there are several ways of skinning this cat and that virtual columns was probably not the best choice in the first place, but I was just playing with an idea and it didn’t quite end up where I thought…so it was interesting in that respect.
The test code:
CREATE TABLE test ( column1 INTEGER NOT NULL ,virtual_column1 INTEGER AS ( column1 ) VIRTUAL NOT NULL ) / CREATE TABLE test ( column1 INTEGER NOT NULL ,virtual_column1 INTEGER AS ( column1 + 0 ) VIRTUAL NOT NULL ) / DROP TABLE test PURGE / CREATE TABLE test ( column1 INTEGER NOT NULL ,virtual_column1 INTEGER AS ( column1 + 0 ) VIRTUAL NOT NULL ,virtual_column2 INTEGER AS ( column1 + 0 ) VIRTUAL NOT NULL ) / CREATE TABLE test ( column1 INTEGER NOT NULL ,virtual_column1 INTEGER AS ( column1 + 0 ) VIRTUAL NOT NULL ,virtual_column2 INTEGER AS ( column1 + 1 - 1 ) VIRTUAL NOT NULL ) / DROP TABLE test PURGE /
And the results:
,virtual_column1 INTEGER AS ( column1 ) VIRTUAL NOT NULL * ERROR at line 4: ORA-54016: Invalid column expression was specified Table created. Table dropped. ,virtual_column2 INTEGER AS ( column1 + 0 ) VIRTUAL NOT NULL * ERROR at line 5: ORA-54015: Duplicate column expression was specified Table created. Table dropped.
I expected it to just work, but I clearly ran in to two problems which scuppered my idea. Firstly, the virtual column can’t simply refer to a normal column without any changes to it, otherwise it fails with ORA-54016. The error isn’t particularly helpful, but eventually I worked out that it was because the column is simply a mapping to a non virtual column. Working around that by adding zero to the numeric column gets it to work, but it’s an ugly hack.
In my scenario there are three different names for this column, depending on the users involved, which then leads on to the next issue, which is that I’d then need two virtual columns pointing at the same source column. Unfortunately if I use the same hack twice, it fails with ORA-54015, because you can’t have two virtual columns with exactly the same expression! A slight variant to the hack and it works, but it’s getting uglier and uglier!
Time to seek out plan B!jQuery(".socialMediaOnEveryPost").addClass("sgmb-left")
I just found this link on OTN to vote for including assertions in a future release of the Oracle database.
A great idea – please vote for it.
One of the most important votes this month…well, I do live in England!
I keep hearing this term lately and I dislike it.
There is no such thing as Unstructured Data. All data has structure. If it didn’t have structure we wouldn’t be able to use it.
What about free text? Well, that’s just a single column value (stored in a CLOB in Oracle, for example) and the free text is, more often than not, on a row with other columns, such as identifiers and timestamps, i.e. yet more structure.
I think what people mean when they use this “marketing foam”TM term is “data that we have not yet defined the structure for”, but in order to use it at some later stage, the structure will need to be defined – that definition process doesn’t actually give the data structure in and of itself, it simply defines what that structure is, in order to be able to use it.
Interestingly, the Wikipedia article for Unstructured Data calls out the imprecise nature of the term:
The term is imprecise for several reasons:
- Structure, while not formally defined, can still be implied.
- Data with some form of structure may still be characterized as unstructured if its structure is not helpful for the processing task at hand.
- Unstructured information might have some structure (semi-structured) or even be highly structured but in ways that are unanticipated or unannounced.
In other words, it does have structure, but maybe we’ve not written it down, or the structure isn’t helpful to processing or is structured in ways we were not expecting – so what?…it’s still structured!
All of the above seem to me to support the view that all data does indeed have structure.
Temporary Post Used For Theme Detection (4a0f3a36-664d-47d9-93c7-24f2096e004d – 3bfe001a-32de-4114-a6b4-4005b770f6d7)
This is a temporary post that was not deleted. Please delete this manually. (31074fcb-66af-42a8-a620-ab6d94cd18a0 – 3bfe001a-32de-4114-a6b4-4005b770f6d7)
An interesting article from Sarah – much good advice there!
A colleague asked if there was a way to do column level dependency tracking recently. He wanted to know for a given view, which tables and the columns on those tables, it was dependent upon, without, of course, reading through the code.
I was vaguely aware that since 11gR1 Oracle has been tracking fine grained (column) dependencies, but couldn’t find a way of seeing the details stored, until I found this interesting article from Rob Van Wijk:
I passed the details on to our DBA who implemented it and it seemed to work, for us. Your mileage may vary, of course.
Some comments on Rob’s blog post, bearing in mind, of course, that it was written in 2008 and refers to 11gR1:
- D_ATTRS contains values other than “000100000A”. I’ve observed this in a basic 12c install and a production 11gR2 install
- D_ATTRS is commented in $ORACLE_HOME/rdbms/admin/dcore.bsq as “/* Finer grain attr. numbers if finer grained */”
- D_REASON is commented in $ORACLE_HOME/rdbms/admin/dcore.bsq as “/* Reason mask of attrs causing invalidation */”. On my basic 12c installation, all rows contain NULL for this value although on a production 11gR2 database I observed a handful of rows with values in this column.
- Noted from the comments against Rob’s article is the opportunity to vote for this feature on OTN here
My colleague asked me yesterday how to enable copy and paste in the command line SQL*Plus window on Windows 7 – a simple enough task…
On the shortcut that starts the command line version of SQL*Plus, right click and bring up the Properties dialog. Nagivate to the Options tab and make sure the QuickEdit mode is checked on, as below:
Now start SQL*Plus and you’ll find that you can hold the left mouse button down whilst dragging a selection area and then pressing return copies the selected text, whilst pressing the right mouse button pastes the copied text.
If you’d prefer to read this from a Microsoft source, try here, where other methods of setting this up are detailed as well as enabling the Autocomplete facility.