This question already has answers here:
Order of evaluation of boolean expressions in SQL
(4 answers)
Closed 6 years ago.
We have a PeopleSoft report running for years without any issues until recently it started producing erroneous results. In order to correct the issue, we had to group the WHERE clause parameters. Please take a look at the screenshot below:
Query Window Screenshot
The query should not retrieve any data at all, given the parameters. The query on the top is the modified one and the one at the bottom is what the report has been running for over 5 years without any errors. Now I'm having to explain to the business stakeholders why it's behaving like that. Any thoughts?
-- THIS QUERY EXECUTES FINE AND DISPLAYS EXPECTED RESULT (TOP PANE)
SELECT PRODUCT, HM_RETIRE_FLAG, RETIREMENT_DT
FROM FSPROD..PS_HM_PRODUCT_AM
WHERE PRODUCT = '7U3'
AND (HM_RETIRE_FLAG = 'Y' OR (HM_RETIRE_FLAG = 'Y' AND RETIREMENT_DT > getdate()))
-- THIS QUERY DISPLAYS A COMPLETELY DIFFERENT RESULT (BOTTOM PANE)
SELECT PRODUCT, HM_RETIRE_FLAG, RETIREMENT_DT
FROM FSPROD..PS_HM_PRODUCT_AM
WHERE PRODUCT = '7U3'
AND HM_RETIRE_FLAG = 'Y' OR (HM_RETIRE_FLAG = 'Y' AND RETIREMENT_DT > getdate())
I think it's fairly obvious what the difference is. Those queries are not alike in any way.
The first one uses parentheses for the last part of the where clause whereas the last one doesn't.
This means the first one won't return any records as soon as the product='7U3' returns false and the second one will return records as soon as the last or returns true.
See the documentation on precedence to learn how this works.
Also, if next time you post your queries as text instead of screenshots I would have made the effort to point it out in your code :)
Related
I am relatively new to SSRS but have been working with SQL for years. I have been tasked with creating a report that reflects shipped items based on their status. For example, I have x number of items with varying statuses including "IN_TRANSIT", "RECEIVING", "SHIPPED", "WORKING", and "CLOSED". The requestor is asking if I can provide the following options in a report drop down:
"IN_PROCESS" Status filter including all statuses except "CLOSED".
"CLOSED".
Essentially, they want to be able to view all non closed statuses, closed, statuses, or all. Right now, I have it set so you can individually select all statuses, essentially getting them the data they want, just not with the "right" parameters.
My question is, does SSRS provide a way to essentially 'group' the non-closed statuses into one inside the report so that when they select "IN_PROCESS" it sends those non-closed statuses to the SQL query I have built in? The problem with using SQL for this is that the dataset I created to generate the dropdown options provides "CLOSED" and "IN_PROCESS" as it's output options, but when they select "IN_PROCESS" (sending that value to the filter in the report), since it's not an actual status, nothing comes back.
If more information or clarification is required, please let me know.
Thanks ahead of time!
You can create a new column in your SQL query and use a CASE statement to give the value of IN_PROCESS or CLOSED for the applicable status. Then you will just need to the filtering condition to match the SSRS parameter to the new column.
Depending on how often this case is likely to be reused should help determine how to approach it. If it sounds like it might become a regular process.... "Oh can we have another report with the same filter but showing xyz " then take the time to setup correctly and it will save time in the future.
Personally I would add a database table, if possible, that contains the status names and then a status group name (ignoring fully normalising for the sake of simplicity here).
CREATE TABLE StatusGroups(Status varchar(10), StatusGroup varchar(10))
INSERT INTO StatusGroups VALUES
('IN_TRANSIT', 'In Process'),('RECEIVING, 'In Process'),('SHIPPED', 'In Process'),('WORKING', 'In Process'),('CLOSED' 'Closed')
Then a simple view
CREATE MyNewView AS
SELECT t.*, g.StatusGroup
FROM MyTable t
JOIN StatusGroups g on t.STATUS = g.Status
Now change your report dataset query to use this view passing in the report parameter like this...
SELECT *
FROM MyNewView
WHERE StatusGroup = #myReportParameter
Your dataset for your report parameter's available values list could then be something like
SELECT DISTINCT StatusGroup FROM StatusGroups
This way if you every add more status or statusgroup values you can add an entry to this table and everything will work without ever having to edit your report.
I have used the level of detail expression outlined in another post in which I'm trying to mark duplicate records as matched. I'm using a level of detail calculation that seems to work sometimes but not al the time. As you can see in my results sometimes it shows matched even though there is only on Server and Application name combination. Here is the calculations formula that I'm using as well as a view of my results. Any insight would be much appreciated.
Thanks in advance.
This is the matched column calculation...
'''if {fixed str([Servers with Token Deployed])+str([App Name]): count([Number of Records])}>1 then 'Yes' else 'No' End'''
I am trying to figure out how to write a formula in the record selector that would allow me to select records in a specified list....but ONLY if there is also a specific record.
In My example. I am pulling earnings codes for employees from specific payroll transactions. For each Transaction date...each employee will have up to 10 codes.
I have my record selection set as this to narrow down the codes I want to see:
{UPCHKD.EARNDED} in ["01", "02", "BNSQT", "BVMT", "CASHBO", "FLAT", "HOL", "HOLPAY", "WAPFML"]
The issue is that I only want to see the first 8 codes IF there is also the WAPFML code. I can't figure out how to tell the record Selector to pick records that have BOTH WAPFML and any of those other 8 codes.
{UPCHKD.EARNDED} in ["01", "02","BNSQT", "BVMT", "CASHBO", "FLAT", "HOL", "HOLPAY", "WAPFML"] and
{UPCHKH.TRANSDATE} in {?Beginning Check Date} to {?Ending Check Date}
I hoped to see only checks where the WAPFML code existed. But I'm obviously returning checks that may not have that code. Using Group selection doesn't work as then I don't see the lines for the other codes.
Assuming you are grouping on {Employee_ID}, Add a group selection formula of MAX({UPCHKD.EARNDED}, {Employee_ID} ) = "WAPFML"
This takes advantage of the fact that "WAPFML" happens to be the largest alpha value in the set. If that is not the case, a more robust approach is to add the UPCHKD table a second time (with an alias), join on same Emp_ID to the first alias, and add a record selection condition on the 2nd alias forcing it to be "WAPFML"
OH I GOT IT! I was grouping by Employee and then transaction date. I entered Maximum ({UPCHKD.EARNDED}, {UPCHKH.TRANSDATE}) = "WAPFML" and took maximum of each transaction date and BOOM. Which now makes all the sense in the world. Thanks so much MilletSoftware for helping me!
I have a query (relationship between CONTRACT <-> ORDERS) that I decided to break up into 2 parts (contract and orders) so I can reuse in another stored procedure.
When I run the code before the break up, it took around 10 secs to run; however, when I use a function for getting the contract, then pump the data into a temp table first, then join to the other parts it takes 2m:30s - why the difference in time?
The function takes less than a second to run and returns only one row i.e. details of one contract (contract_id is the parameter supplied to the function).
The part that is most effecting the performance the (ORDERS) largest table in the query has 4.1 million rows and joins to a few other tables however; if I just run the sub query for orders in isolation with a particular filter i.e. the contract id it takes less than a second to run and just happens to return zero records based for the contract I am testing on (due to filtering on the type of order it is looking for).
Base on the above information you would think 1 sec at most for the function + 1 sec at most to get the orders + summarize = 2 seconds at most, not 2 and half minutes!
Where am I going wrong, how do I begin to isolate the issue in time difference?
I know someone is going to tell me to paste the code but surely it is an issue of the database vs indexes perhaps vs how the compiler performs when dealing with raw code versus broken up code into parts. Is there an area of the code I can look at before having to post my whole code as I have tried variations of OUTER APPLY vs LEFT JOIN from the contract temp table to the orders subquery and both give me about the same result. Any ideas?
I don't think the issue was with the code but the network I was running it on. Although bizarre in the fact I had 2 versions of the proc running side by side and yesterday or rather before the weekend one was running in 10 secs and it is still running in 10 secs 3 days later and my new version (using the function) was taking anywhere between 2 to 3 minutes. This morning it is running at 2 or 3 seconds!! So I don't know if it is the fact I changed from declaring my table structure and using a table variable instead first to where previously I was using SELECT ... INTO #Contract made the difference or the network or precompiling has an affect. Whatever it is it no longer an issue. Should I delete this post?
I found a solution that another Tech had posted:
New Formula:
{Product.Size} <> “xsm” or IsNull({Product.Size})
Unfortunately, when preview your report, you will find that this doesn’t work. That is not because of a mistake in our logic but rather, what I consider to be a bug in Crystal Reports. If I took this exact same condition and applied it to the database records using a query analyzer or querying tool, I would see the blank records. Unfortunately, Crystal is not allowing the null values to come through even though our formula says that they should.
The trick to circumventing this bug is to put the IsNull() check FIRST in the formula.
Thus, if we rearrange the condition to this:
IsNull({Product.Size}) or {Product.Size} <> "xsm"
WORKED LIKE A CHARM
Problem is, if I select criteria for the second OR statement ({HiredRate.UTRANSDOC}startswith{?TransYN}) and NO for the first ({HiredRate.UTRANSWEB}startswith{?WebYN}) I get only one record that meets the TransYN criteria. If I switch places in the formula putting ({HiredRate.UTRANSDOC}startswith{?TransYN}) 1st I get all of the data.
When I run the SQL query I get all of the data no matter what order they are in. The Crystal Preview only gives me all of the data on the first from the OR section.
The only thing that stands out looking at the data from SQL is that the one record Crystal is returning has YES in the Transdoc field and the Transweb field is blank. All other records show YES for Transdoc and NULL for the Transweb field.
Here is the Crystal Record Selection Formula
{HiredRate.CONTSUPREF} startswith {?LanguageCombo}
and {HiredRate.ONDATE} = {?ProjectDate}
and {HiredRate.ACTVCODE}= "SIG"
and {HiredRate.RESULTCODE} = "CLM"
and (
{HiredRate.UTRANSWEB}startswith{?WebYN}
or {HiredRate.UTRANSDOC}startswith{?TransYN}
or {HiredRate.UTRANLANL0}startswith{?LanloYN}
or {HiredRate.UINTCONSEC}startswith{?InterpYN}
or {HiredRate.UINTCONF}startswith{?IntConfYN}
or {HiredRate.UINTOPI}startswith{?OPIYN}
)
Here is the SQL query Crystal is using:
SELECT HiredRate.DEAR, HiredRate.CONTSUPREF, HiredRate.LASTDATE, HiredRate.CONTACT, HiredRate.USOURCLANG, HiredRate.UTARGLANG, HiredRate.UTRANSDOC, HiredRate.UTRANSWEB, HiredRate.UTRANLANL0, HiredRate.UINTCONSEC, HiredRate.UINTCONF, HiredRate.UINTOPI, HiredRate.ONDATE, HiredRate.ACTVCODE, HiredRate.RESULTCODE
FROM GoldMine_Main.dbo.HiredRate HiredRate
WHERE HiredRate.CONTSUPREF LIKE 'ENG>SPA%' AND (HiredRate.ONDATE>={ts '2012-04-01 00:00:00'} AND HiredRate.ONDATE<{ts '2013-04-06 00:00:00'}) AND HiredRate.ACTVCODE='SIG' AND HiredRate.RESULTCODE='CLM' AND (HiredRate.UTRANSWEB LIKE 'NO%' OR HiredRate.UTRANSDOC LIKE 'YES%' OR HiredRate.UTRANLANL0 LIKE 'NO%' OR HiredRate.UINTCONSEC LIKE 'NO%' OR HiredRate.UINTCONF LIKE 'NO%' OR HiredRate.UINTOPI LIKE 'NO%')
ORDER BY HiredRate.DEAR, HiredRate.CONTACT
This is happening because the {HiredRate.UTRANSWEB} is null - the rest of the expression is therefore evaluating as null in Crystal, even though (logically) it shouldn't.
When the first two or conditions are swapped around, the {HiredRate.UTRANSDOC} condition evaluates as true and the rest of the expression is short-circuited - which is why records are then selected.