db2: select from table without replacement - db2

Hi I would just like to ask a simple question - is there a way in DB2, to select a row from a table (whether that be based on a join or selecting a random row), and then select from the same table again where choosing the last, or any previous rows cannot be selected.
I am thinking I have to loop my code through each row in the table and delete each row I select, but would be interested if anyone has an alternative solution. No code needed but rather describe another approach.
Thanks,
Arron

The simplest way of doing this is to declare a cursor to select all rows from the table then process the
cursor one row at a time. Each row will be selected exactly 1 time (this is pretty much what a cursor is all about).
I suspect that is not the answer you were looking for. You most likely have at least two other constrains on this
selection problem:
You do not want, or cannot, have a single cursor open until the entire table has been processed
You want to have some sort of "randomness" with respect to the order in which rows are selected
The problem of not being able to open and process the entire table under a single cursor can be solved by
maintaining some sort of "state" information between selections. The "state" can be used to determine whether a row
is still eligible for selection on subsequent inquiries. You might add another column to the table to hold the "selected"
state of that row. When a row is inserted its "selected" state is set to "no". On each select operation the state
of the selected row is updated to "yes". The predicate to select new rows then needs to have a WHEN SELECT_STATE = 'no'
added to it to disqualify previously selected rows. If you cannot change the structure of the table you are selecting
from, then add a second table having the same primary key as the selection table plus the "selected" indicator then join
these tables to obtain the required state information.
Another approach is to delete a row once it has been selected.
These or some similar type of state management can be used to solve the selection eligibility problem.
If you need to introduce randomness into the selection process (i.e. make it difficult go guess what
the next row to be selected will be), then you have a very different problem to solve. If this is the case
please ask a new question outlining the approximate size of you table (how many rows) and what the key structure
is (eg. a number between 1 and 100000, a 30 character name etc.)

You can use a cursor, and use the 'delete where current of' feature called positioned-delete. For more information:
http://pic.dhe.ibm.com/infocenter/db2luw/v10r5/topic/com.ibm.db2.luw.sql.ref.doc/doc/r0000939.html
http://mysite.verizon.net/Graeme_Birchall/cookbook/DB2V97CK.PDF page 55

Related

In Tableau, Is it possible to give Column number in Column Shelf

I'm using Tableau. So, instead of giving the [Column_name], Is it possible to give [Column_number] in column shelf?
- Hariharasudhan. R.
No -- for good reason.
Think of the data source as a template for a potential SQL (or MDX or TQL) query; specifying tables, joins, unions and possibly some where/having clauses for data source filters.
The actual SQL generated for any particular view will be an (optimized) query that only selects columns that are actually needed for that particular view, adds where/having clauses based on the filters being used etc.
So a column doesn't have a fixed number. The same column may be the first field selected in one situation, the last field in another situation, and left off completely in another.
If you want to change the name of column shelf:
Create a duplicate of variable and change original with duplicates and assign name as your wish by right-click on Edit Aliases and change as per your requirement.
Go to Data Source
On the middle right corner check on Show aliases
Go to column and right click on it. Go to Rename

Displaying Record Once in SSRS Table

I've created a table from a procedure that will display a summary of all incentives from a given collection operation or recruiter in my report. But if the collection operation or recruiter has multiple incentives assigned to their drives during the time period, it will repeat the same collection operation or recruiter on each line. If possible, I'd like to only display the first time it occurs and then leave a blank space until it hits a new collection operation/recruiter.
See screenshot for example:
Is there a way to create an expression to handle this? Thanks,
The answer is Groupings. If you add a row grouping on your "Collection Operation" column it should do a trick. To add a group Right click on your details row. just like you would to add a row. There should be an option to add a parent group. Select that and tell it you want to group on "Collection Operation" and click "Ok". You may have to adjust your formatting a little, but I think this will achieve what you are going for.
I just noticed the recruiter Column. To make that only display once you will need to use an expression. Now that you have a loop you should be able to write an expression that determines your position within the group and blank it out when it isn't the first row.

Is there any way to avoid PostgreSQL placing the updated row as the last row?

Well, my problem is that each time that I make an update of a row, this row goes to the last place in the table. It doesn't really matter where was placed before.
I've read in this post Postgresql: row number changes on update that rows in a relational table are not sorted. Then, why when I execute a select * from table; do I always get the same order?
Anyway, I don't want to start a discussion about that, just to know if is there any way to don't let update sentence place the row in the last place.
Edit for more info:
I don't really want to get all results at all. I have programmed 2 buttons in Java, next and previous and, being still a begginer, the only way that I had to get the next or the previous row was to use select * from table limit 1 and adding offset num++ or offset num-- depending of the button clicked. So, when I execute the update, I lose the initial order (insertion order).
Thanks.
You could make some space in your tables for updates. Change the fill factor from the default 100%, no space for updates left on a page, to something less to create space for updates.
From the manual (create table):
fillfactor (integer)
The fillfactor for a table is a percentage
between 10 and 100. 100 (complete packing) is the default. When a
smaller fillfactor is specified, INSERT operations pack table pages
only to the indicated percentage; the remaining space on each page is
reserved for updating rows on that page. This gives UPDATE a chance to
place the updated copy of a row on the same page as the original,
which is more efficient than placing it on a different page. For a
table whose entries are never updated, complete packing is the best
choice, but in heavily updated tables smaller fillfactors are
appropriate. This parameter cannot be set for TOAST tables.
But without an ORDER BY in your query, there is no guarantee that a result set will be sorted the way you expect it to be sorted. No fill factor can change that.

Erroneous duplicate rows in BIRT report

I have a problem with a BIRT report I'm working on where I have a nested table in the report. The outer table contains data to do with an item on an invoice, while the inner table contains stuff to do with price banding for labor charges. I've written a separate DataSet which gets the inner data, bound by parameters to data in the outer table. Now, when I preview the inner DataSet in BIRT using the defaults I've given it, it returns two rows of data for that bill number & item number - a normal rate & an overtime rate if you like. When I run the report in full over the same data, the outer table stuff is fine, but the inner table just repeats the same row over twice - it's just the first row repeating.
This is sorta what the table looks like in layout view:
Item Description Rate Quantity Item total
[item] [desc] [rate] [quantity] [total]
...where the price & quantity are in the inner table.
I'd have expected to see something like:
Item Description Rate Quantity Item Total
1 Callout $40 1 $40
2 Labor $30 4.5 $185
$50 1
but instead I get more like:
Item Description Rate Quantity Item Total
1 Callout $40 1 $40
2 Labor $30 4.5 $185
$30 4.5
...even though querying the database & previewing the inner data set based on the same input criteria show the expected result.
Has anyone else had experience like this? I have a hunch it's to do with bindings, but not sure what.
One way to get this behavior is by accidentally replacing a table-level binding with a column-level binding.
For example, define a table by dragging a data set into the report. Select the entire table (use the outline view, or select something in the table and then click on the "Table" button that pops up just below the grid.) Then go to the Binding tab. Note that the data set and column bindings are all filled in.
Now select just one field in the Detail row. On the Binding tab, note that the Data Set is blank, and no column binding is shown. Someone who is confused by this (as I was) might then edit the column's binding and specify the same Data Set that was used to create the table. If you do this you will only see a single value repeated in that column when you run the report. (I believe the overridden column is binding to a second instance of the data set, not the one the table is iterating over.)
Not sure your question can be answered withou looking at the data and the design. But it is important to note that the results you see in the dataset preview, and not neccisarly what you would see if the query was run fully. I have seen difference with 7 records returned. I thought as it was only 7 it would be the same on full run, but it's not. The preview is not just a top 500 query, it has some other (not sure what) filters also.
To problem solve if it is your query or your binding.
If you are using a SQL database. Run the SQL in a SSMS query and see if you get the same results you do when run in the innner table.
Altentively, create a new test report, copy over your dataset and use with a stand alone table.
I think I sorted it, & this is the most bizarre thing: On the child table I'd been deleting the header & footer row & just leaving the detail row in, in the layout view. Last thing today, just before I was going to go home, I tried again - deleted the table for about the 70th time that day, replaced it, re-did the parameter bindings all exactly as before, but this time I left the header row & footer intact. Clicked the preview tab, voila, all shows up correctly. So, since I didn't need the header or footer on the child table, I went into properties, clicked Hide this element, preview again - all good. No difference to the data bindings, no difference to mappings or anything else, no change to the data sets - the only difference was leaving the header & footer in place but hidden.
Contemplating making a bug report, tbh.

Hide duplicate row SSRS 2008 R2

Duplicated data is coming in my report because source table has duplicate data. Without creating group, I want to hide duplicate data writing expression. So what I did: I select table row and put a expression for hidden property of table row.
The expression was like =(Previous(Fields!ID.Value) = Fields!ID.Value)
but it did not work ... Still duplicate data is showing. So tell me how to suppress duplicate rows in ssrs writing expression not by grouping.
You probably should try these options first:
Try to clean the duplicate data at the source.
Change your source query so the duplicates don't appear in the dataset. (e.g. SELECT DISTINCT)
If not, on the row's Visibility Hidden property you can use the Previous function:
=iif(Fields!YourField.Value = Previous(Fields!YourField.Value), True, False)
You would have to sort on the YourField column for it to work.
I was putting the expression above also until I started using the "Hide Duplicates" line in the properties pane. You basically just select the row, in the dropdown choose your Dataset and that's it. any duplicates will be hidden. Also if you just want to hide certain textboxes duplicates you can do the same as i stated earlier except click on the textbox and not the row. Just another alternative, i'm aware you said using an expression.
You can do it using expression or "Hide Duplicates" options from cell or row properties.
Expressions :
=IIF(Fields!YourField.Value = Previous(Fields!YourField.Value), True, False)
Hide Duplicates Steps:
Select row or cell
Click on F4 key on your Keyboard
Look for "Hide Duplicates"
Choose your DataSet from the dropdownlist
Done, I hope that helps
As an alternative option, you can do it by setting row groups. In a report I was trying to create, the Hide Duplicates property wouldn't behave correctly because of using the same dataset multiple times in a list container.
All you need to do is set the row group properties for the default row group (rightclick the grey row header, and go to Row Group then Group Properties), and add Group expressions on the General tab. Add as many as you need for each field. It's like the Remove Duplicates tool in Microsoft Excel
Sometimes the Hide Duplicates option does apply to the report content. If you add a =Sum(Field!Field_Name.Value) sum around field in the cell, it suppresses the copy from previous record. Of course, strictly speaking this is a solution where the incoming data set has NULL rows for the cells with the issue.
Since the using of Previous function in SSRS compare to the only record previous to it, thus it might cause the duplicate of records still shown if the repeated records not next to each other.
Use the sorting on each table you apply the Previous function, it should resolve the "non next to each other" duplicate records as well.