Showing posts with label everybody. Show all posts
Showing posts with label everybody. Show all posts

Monday, March 26, 2012

Query Problem

Hi everybody,

I have a table storing the test dates, student names and the grade of
the students. I am now going to find out those dates with student1 got
A and student2 got B and vice versa (i.e. student1 got B and student2
got A). How can I do that? I think it is quite complicate for me.
Please help. I'm using SQL server 2000.

FRANKLINPlease post your DDL + INSERT statements of your sample data. Here's an
untested solution:

select
TheDate
from
MyTable
where
Grade in ('A', 'B')
and Student in ('Student1', 'Student2')
group by
TheDate
having
(sum (case when Grade = 'A' and Student = 'Student1' then 1 else 0
end) = 1
and sum (case when Grade = 'B' and Student = 'Student2' then 1 else 0
end) = 1)
or (sum (case when Grade = 'A' and Student = 'Student2' then 1 else 0
end) = 1
and sum (case when Grade = 'B' and Student = 'Student1' then 1 else 0
end) = 1)

Assumes only one grade per student per date.

--
Tom

----------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada
www.pinpub.com
..
"肥權" <lokalun@.gmail.com> wrote in message
news:1142790320.623759.278150@.g10g2000cwb.googlegr oups.com...
Hi everybody,

I have a table storing the test dates, student names and the grade of
the students. I am now going to find out those dates with student1 got
A and student2 got B and vice versa (i.e. student1 got B and student2
got A). How can I do that? I think it is quite complicate for me.
Please help. I'm using SQL server 2000.

FRANKLIN|||Another alternative, again untested:

select
TheDate
from
MyTable
where
Grade in ('A', 'B')
and Student in ('Student1', 'Student2')
group by
TheDate
having
sum (case
when Grade = 'A' and Student = 'Student1' then 1
when Grade = 'B' and Student = 'Student2' then 1
else 0 end) = 2
or sum (case
when Grade = 'A' and Student = 'Student2' then 1
when Grade = 'B' and Student = 'Student1' then 1
else 0 end) = 2

Same assumption as before.

--
Tom

----------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada
www.pinpub.com
..
"Tom Moreau" <tom@.dont.spam.me.cips.ca> wrote in message
news:najTf.1845$ji6.54932@.news20.bellglobal.com...
Please post your DDL + INSERT statements of your sample data. Here's an
untested solution:

select
TheDate
from
MyTable
where
Grade in ('A', 'B')
and Student in ('Student1', 'Student2')
group by
TheDate
having
(sum (case when Grade = 'A' and Student = 'Student1' then 1 else 0
end) = 1
and sum (case when Grade = 'B' and Student = 'Student2' then 1 else 0
end) = 1)
or (sum (case when Grade = 'A' and Student = 'Student2' then 1 else 0
end) = 1
and sum (case when Grade = 'B' and Student = 'Student1' then 1 else 0
end) = 1)

Assumes only one grade per student per date.

--
Tom

----------------
Thomas A. Moreau, BSc, PhD, MCSE, MCDBA
SQL Server MVP
Columnist, SQL Server Professional
Toronto, ON Canada
www.pinpub.com
..
"肥權" <lokalun@.gmail.com> wrote in message
news:1142790320.623759.278150@.g10g2000cwb.googlegr oups.com...
Hi everybody,

I have a table storing the test dates, student names and the grade of
the students. I am now going to find out those dates with student1 got
A and student2 got B and vice versa (i.e. student1 got B and student2
got A). How can I do that? I think it is quite complicate for me.
Please help. I'm using SQL server 2000.

FRANKLIN|||Please post DDL, so that people do not have to guess what the keys,
constraints, Declarative Referential Integrity, data types, etc. in
your schema are. Sample data is also a good idea, along with clear
specifications. Here is my guess

CREATE TABLE Gradebook
(student_name CHAR(15) NOT NULL,
test_date DATETIME NOT NULL,
grade CHAR(1) NOT NULL,
PRIMARY KEY (student_name, test_date));

>> I am now going to find out those dates with student1 got A and student2 got B and vice versa (i.e. student1 got B and student2 got A).<<

untested

WITH X(student_name, grade, test_date)
SELECT student_name, grade, test_date
FROM Gradebook
WHERE student_name IN ('student1', 'student2')
AND grade IN ('A', 'B')
SELECT
FROM X AS X1, X AS X2
WHERE X1.test_date = X2.test_date
AND X1.student_name <> X2.student_name
AND X1.grade <> X2.grade;|||FRANKLIN,

i repeat the sentence about the necessity to post ddl and sample data
and want to suggest the following solution of your problem:

CREATE TABLE Gradebook
(student_name CHAR(15) NOT NULL,
test_date DATETIME NOT NULL,
grade CHAR(1) NOT NULL,
PRIMARY KEY (student_name, test_date));

INSERT INTO Gradebook
SELECT 'Student1', '2006-01-01', 'A' UNION ALL
SELECT 'Student2', '2006-01-01', 'B' UNION ALL
SELECT 'Student3', '2006-01-01', 'A' UNION ALL
SELECT 'Student1', '2006-01-03', 'C' UNION ALL
SELECT 'Student2', '2006-01-03', 'A' UNION ALL
SELECT 'Student3', '2006-01-03', 'B' UNION ALL
SELECT 'Student1', '2006-01-07', 'A' UNION ALL
SELECT 'Student2', '2006-01-07', 'C' UNION ALL
SELECT 'Student3', '2006-01-07', 'A' UNION ALL
SELECT 'Student1', '2006-01-09', 'A' UNION ALL
SELECT 'Student2', '2006-01-09', 'B';

SET STATISTICS IO ON;

-- SELECT DISTINCT G1.test_date
-- FROM Gradebook AS G1, Gradebook AS G2
-- WHERE G1.test_date = G2.test_date
-- AND ( ( G1.student_name = 'Student1' AND G1.grade = 'A'
-- AND G2.student_name = 'Student2' AND G2.grade = 'B')
-- OR( G1.student_name = 'Student2' AND G1.grade = 'B'
-- AND G2.student_name = 'Student1' AND G2.grade = 'A'));

SELECT G1.test_date
FROM Gradebook AS G1
WHERE G1.student_name = 'Student1'
AND G1.grade IN('A', 'B')
AND EXISTS(SELECT *
FROM Gradebook AS G2
WHERE G2.student_name = 'Student2'
AND G2.test_date = G1.test_date
AND G2.grade = CASE G1.grade
WHEN 'A' THEN 'B'
WHEN 'B' THEN 'A' END);

SET STATISTICS IO OFF;

DROP TABLE Gradebook;

--
Andrey Odegov
avodeGOV@.yandex.ru
(remove GOV to respond)|||homework.

Wednesday, March 7, 2012

Query on XML column: fn:lower-case not working?

Hi everybody, I'm looking to issue a query to an XML column like this:
SELECT *
FROM dbo.ContactRecords
WHERE XmlContent.exist(
'declare namespace
my="http://schemas.microsoft.com/office/infopath/2003/myXSD/2004-03-09T14-09
-40";
// my:myFields[my:AddressInfo[fn:contains(f
n:lower-case(my:Company[1]),"lego"
)]]')
= 1
(i skipped some other conditions, but they are working)
--> SQL 2005 tells me: There is no function
'{http://www.w3.org/2004/07/xpath-functions}:lower-case()'
however, there is one...
How can I compare string values in SQL 2005 case-insensitive?
I appreciate your help, have a good day,
MichaelHello michael.hofer@.getronics.com,
Well, you've stumbled into the chasm caused by Microsoft trying to get a
product released and the W3C XQuery WG trying to get things perfect the firs
t
time. There's a number of functions that just aren't implemented in MS's
XQuery heap but are in the spec. We might get them in service packs, or we
might get them in the next version of SQL Server. String-Upper and String-L
ower
are a couple of the common ones that we'd like to have but don't.
Interestingly enough, I see that contains() has a optional collation pattern
.
I try to find a collation that ignored case to test with but didn't have
much immediate luck. If you might want to go down that path. Or not. No prom
ises.
I'd like to offer a suggestion other than .value(path,'varchar(n)') like
'%pattern%' (e.g., do the comparsion in T-SQL as shown below) since this
is fairly ugly for the kind of stuff you're doing. But it does work.
declare @.x xml
set @.x = '<book>Ender''s Game</book><book>Xenocide</book><book>Children of
the Mind</book><book>The Hive Queen</book><book>The Hegemon</book>'
select p.x.value('.','varchar(100)')
from @.x.nodes('/book') as p(x)
where p.x.value('.','varchar(100)') like '%the%'
Thank you,
Kent Tegels
DevelopMentor
http://staff.develop.com/ktegels/