I have a column (datatype nvarchar(max), not my choice, legacy) with various different uses to the end user dependent on other factors.
I was trying to narrow down to certain specific data within that column given earlier clauses within some sample (but highly representative) data I noticed that the remaining data was either a 1,2,3,4, or a date
I initially added some nested NULLIF along with an IS NOT NULL
AND NULLIF(NULLIF(NULLIF(NULLIF([ColumnName],'1'),'2'),'3'),'4') IS NOT NULL
The 1 is as a string because in the global data there are some strings so removing the single brackets creates an implied conversion to int which many records would fail. This got me down to 25 records in my sample, the records with dates and the target records
I then thought I'd add an ISDATE to isolate those records
AND NULLIF(NULLIF(NULLIF(NULLIF(ISDATE([ColumnName]),'1'),'2'),'3'),'4') IS NOT NULL
This then returned 60k or so records, which was not the behaviour I expected.
I ran the following queries to see if there was any incompatibility with the two commands inline but the returned as expected
SELECT NULLIF(ISDATE('06/01/2022'),1)
returned NULL.
SELECT NULLIF(ISDATE('06/01/2022'),'1')
in case it didn't like a string, but returned NULL.
SELECT NULLIF(NULLIF(ISDATE('06/01/2022'),'1'),'2')
in case it didn't like the nest, but returned NULL.
So why does it not NULL the values that present as dates, and also why does it negate the other NULLIF commands in the outer parts of the nest?
Turns out I'm an idiot
I forgot to catch all the times that the ISDATE() returned a 0, so it was returning all the values that the NULLIFs were trying to catch
AND ISDATE(ISNULL(NULLIF(NULLIF(NULLIF(NULLIF([2nd Ref],'1'),'2'),'3'),'4'),'06/01/2022')) = 0
So changing the order here helped, creating NULLs for the expected values, changing those into a known date, then returning where it isn't a date
That should now work as expected. Oh well, someone might one day find it useful
edit: There are a couple of records that still have dates in them, I will also use the TRYCONVERT as Jeroen suggested
Related
Here's the thing, I have a stored procedure which must select products from the database, but only those which fit in between the start product code, and the finish product code which the user chooses and get sent to the procedure.
Now the product codes are varchars, and they may contain numbers AND letters, like 01, 01B and 02A may be product codes, and the procedure must be able to select all the values in between the user selected ranges.
I'm having trouble with such where clause doing it like this:
...
SELECT ...
WHERE ... AND
ProductCode >= #IniProCod
AND
ProductCode <= #FinProCod
...
The available product codes are shown to the user via a previous query which returns only available product codes, so there should be no problem with non existant values.
I also tried evaluating them as text, although they are received as varchars by the procedure anyway, and since this is inside a dynamic SQL query it actually looks something like this:
...
SELECT ...
WHERE ... AND
ProductCode >= '''+#IniProCod+'''
AND
ProductCode <= '''+#FinProCod+''''
...
And when I try to execute it it returns either the first row, or a set of rows that has very little to nothing to do with the actual values I'm sending to the procedure, so I don't even think it's comparing the values as strings at all.
I leave out code that has nothing to do with the issue, the query actually works alright without this range where clause.
So any ideas?
Edit:
I also tried using BETWEEN but got the same outcome.
And I'm very certain it is just not comparing the variables as text even when they are declared as varchars and I'm using apostrophes on them, and this is only happening on this procedure. I tested this same clause on a clean select query and it does behave as it should, comparing each character for what they are. If I set the initial values as 0 and 29 respectively it will give me a range from 01 to 29 which are the values contained within that range, but when these same values are given to the stored procedure it will return a range from 01 which is the first value, to 03F, which makes no sense to me.
Most likely, you'll need to have your application figure out the list of product codes based on the start & end codes, then send that list to the stored procedure. Your SQL would just contain something like AND ProductCode IN (#myProductCodeList).
I've read lots of examples of JOIN syntax, but none of them work for my problem.
I have a table of quotes and a table of sales enquiries. One quote may result in multiple enquiries, or it may not result in any enquiries. I want to create a list of all quotes and include details of any resulting enquiry. I'm expecting multiple result rows for a quote that resulted in many enquiries. I'm expecting a single row with empty or null enquiries fields, for those quotes that didn't give rise to any enquiries.
The connecting data is in the quotes table, where there's a field called 'activity' that contains the id of any or all enquiries that resulted. It's updated each tim e anew enquiry comes in.
So I tried this:
SELECT q.*, e.id, e.price
FROM quotes as q
LEFT JOIN enquiries as e
ON INSTR(q.activity, e.id) >'0'
WHERE q.date > '2013-07-01'
But every row in my results includes enquiries data. I can't seem to get it to include the quotes that never resulted in anything. I thought LEFT JOIN was supposed to give me all of the quotes regardless of enquiries, and include the enquiries data where there was a match. But every example I've seen is just joining on a.id = b.id, so I suspect that my INSTR() match criteria might be messing things up.
As previous commentators have suggested the issue will be down to the join with Instr. The return value from INSTR of many RDBMSs is an integer value. When you therefore test the value of INSTR against '0' you won't get a match. Also, if Instr doesn't find a match you may get something else returned like MS Access where Null is a possible return value. This is obviously all speculation and we really need to see an example of your data and the issue to confirm if this is the actual problem. In the absence of any more info this is the best you are going to get:
Without knowing which DB you are using I've included a few links for INSTR:
MySql,
Oracle,
MS Access (returns variant Long),
SQL Server - No Instr Function - CharIndex
I think your problem might be somewhere else, because this seems to work fine for me. I assumed the list of enquiries was just a comma separated string. See http://sqlfiddle.com/#!4/71ce1/1
Get rid of the single quotes around the 0, but that doesn't make any difference. Also, you shouldn't be relying on the default date format, but using TO_DATE.You don't say what DBMS you're using, but I tried both Oracle and MySQL.
what am I doing wrong with my sql query? It always return an empty rows even if there is a value exist.
Here is my query:
SELECT *
FROM users
WHERE user_theme_id IN ( 9735, 9325, 4128 )
AND ( user_date_created BETWEEN '2013-06-04' AND '2013-06-10' );
I tried to cut my original query one by one, I got a result. Here is the first one:
SELECT * FROM users WHERE user_theme_id IN (9735, 9325, 4128 );
I got 3 rows for this result. See attached snapshot:
Now, the next query that I run is this:
SELECT *
FROM users
WHERE user_date_created BETWEEN '2013-06-04' AND '2013-06-10';
I do get 3 results on this. See attached snapshot:
By the way, this sql that uses BETWEEN should suppose return 4 rows but it only return 3. It doesn't return the data which has the created date of 2013-06-10 08:27:43
What am I doing wrong with my original query Why does it always return an empty rows?
If you are getting results by separately running different where clauses doesn't guarantee that AND 2 where clauses will return an answer.
There has to be intersection of rows to get result while AND.
You should validate your data and see if overlapping exists.
I have able to make it work by not using the SQL BETWEEN operators but instead COMPARISON OPERATORS like: >= || <=
I have read it from W3schools.com, the SQL between can produce different results in different databases.
This is the content:
Notice that the BETWEEN operator can produce different result in different databases!
In some databases, BETWEEN selects fields that are between and excluding the test values.
In other databases, BETWEEN selects fields that are between and including the test values.
And in other databases, BETWEEN selects fields between the test values, including the first test value and excluding the last test value.
Therefore: Check how your database treats the BETWEEN operator!
That is what happened in the issue that I am facing. The first field was being treated as part of the test values and the 2nd field was being excluded. Using the comparison operators give accurate result.
I am using SQL server 2005, querying with Web Developer 2010, and the min function appears to be returning more than one value (for each ID returned, see below). Ideally I would like it to just return the one for each ID.
SELECT Production.WorksOrderOperations.WorksOrderNumber,
MIN(Production.WorksOrderOperations.OperationNumber) AS Expr1,
Production.Resources.ResourceCode,
Production.Resources.ResourceDescription,
Production.WorksOrderExcel_ExcelExport_View.PartNumber,
Production.WorksOrderOperations.PlannedQuantity,
Production.WorksOrderOperations.PlannedSetTime,
Production.WorksOrderOperations.PlannedRunTime
FROM Production.WorksOrderOperations
INNER JOIN Production.Resources
ON Production.WorksOrderOperations.ResourceID = Production.Resources.ResourceID
INNER JOIN Production.WorksOrderExcel_ExcelExport_View
ON Production.WorksOrderOperations.WorksOrderNumber = Production.WorksOrderExcel_ExcelExport_View.WorksOrderNumber
WHERE Production.WorksOrderOperations.WorksOrderNumber IN
( SELECT WorksOrderNumber
FROM Production.WorksOrderExcel_ExcelExport_View AS WorksOrderExcel_ExcelExport_View_1
WHERE (WorksOrderSuffixStatus = 'Proposed'))
AND Production.Resources.ResourceCode IN ('1303', '1604')
GROUP BY Production.WorksOrderOperations.WorksOrderNumber,
Production.Resources.ResourceCode,
Production.Resources.ResourceDescription,
Production.WorksOrderExcel_ExcelExport_View.PartNumber,
Production.WorksOrderOperations.PlannedQuantity,
Production.WorksOrderOperations.PlannedSetTime,
Production.WorksOrderOperations.PlannedRunTime
If you can get your head around it, I am selecting certain columns from multiple tables where the WorksOrderNumber is also contained within a subquery, and numerous other conditions.
Result set looks a little like this, have blurred out irrelevant data.
http://i.stack.imgur.com/5UFIp.png (Wouldn't let me embed image).
The highlighted rows are NOT supposed to be there, I cannot explicitly filter them out, as this result set will be updated daily and it is likely to happen with a different record.
I have tried casting and converting the OperationNumber to numerous other data types, varchar type returns '100' instead of the '30'. Also tried searching search engines, no one seems to have the same problem.
I did not structure the tables (they're horribly normalised), and it is not possible to restructure them.
Any ideas appreciated, many thanks.
The MIN function returns the minimum within the group.
If you want the minimum for each ID you need to get group on just ID.
I assume that by "ID" you are referring to Production.WorksOrderOperations.WorksOrderNumber.
You can add this as a "table" in your SQL:
(SELECT Production.WorksOrderOperations.WorksOrderNumber,
MIN(Production.WorksOrderOperations.OperationNumber)
FROM Production.WorksOrderOperations
GROUP BY Production.WorksOrderOperations.WorksOrderNumber)
I'm having trouble with Microsoft Access 2003, it's complaining about this statement:
select cardnr
from change
where year(date)<2009
group by cardnr
having max(time+date) = (time+date) and cardto='VIP'
What I want to do is, for every distinct cardnr in the table change, to find the row with the latest (time+date) that is before year 2009, and then just select the rows with cardto='VIP'.
This validator says it's OK, Access says it's not OK.
This is the message I get: "you tried to execute a query that does not include the specified expression 'max(time+date)=time+date and cardto='VIP' and cardnr=' as part of an aggregate function."
Could someone please explain what I'm doing wrong and the right way to do it? Thanks
Note: The field and table names are translated and do not collide with any reserved words, I have no trouble with the names.
Try to think of it like this - HAVING is applied after the aggregation is done.
Therefore it can not compare to unaggregated expressions (neither for time+date, nor for cardto).
However, to get the last (principle is the same for getting rows related to other aggregated functions as weel) time and date you can do something like:
SELECT cardnr
FROM change main
WHERE time+date IN (SELECT MAX(time+date)
FROM change sub
WHERE sub.cardnr = main.cardnr AND
year(date)<2009
AND cardto='VIP')
(assuming that date part on your time field is the same for all the records; having two fields for date/time is not in your best interest and also using reserved words for field names can backfire in certain cases)
It works because the subquery is filtered only on the records that you are interested in from the outer query.
Applying the same year(date)<200 and cardto='VIP' to the outer query can improve performance further.