Order SQL result with point system - mysql

I'm trying to get my SQL result to be ordered by a certain 'point-system'. For every field that's filled in, the search result gets a certain amount of points. At the end of my SQL I add the following ORDER BY - code, however, if I change one of the '1'-numbers to a higher number, the search result order isn't changed. Why wouldn't that work? (Changed the bedrijfslogo-case to 9 in this example):
Thanks guys!
$sql .= " ORDER BY
case when ID is not null then 1 else 0 end +
case when username is not null then 1 else 0 end +
case when password is not null then 1 else 0 end +
case when voornaam is not null then 1 else 0 end +
case when achternaam is not null then 1 else 0 end +
case when telefoonnummerP is not null then 1 else 0 end +
case when emailP is not null then 1 else 0 end +
case when functie is not null then 1 else 0 end +
case when bedrijfsnaam is not null then 1 else 0 end +
case when bedrijfsslogan is not null then 1 else 0 end +
case when bedrijfslogo is not null then 9 else 0 end +
case when bedrijfsfoto is not null then 1 else 0 end +
case when bedrijfsfoto2 is not null then 1 else 0 end +
case when bedrijfsfoto3 is not null then 1 else 0 end +
case when bedrijfsfoto4 is not null then 1 else 0 end +
case when bedrijfsomschrijving is not null then 1 else 0 end +
case when bedrijfsspecialiteiten is not null then 1 else 0 end +
case when bedrijfsgeschiedenis is not null then 1 else 0 end +
case when openingstijden is not null then 1 else 0 end +
case when kvk is not null then 1 else 0 end +
case when straatnaam is not null then 1 else 0 end +
case when huisnummer is not null then 1 else 0 end +
case when postcode is not null then 1 else 0 end +
case when plaats is not null then 1 else 0 end +
case when nevenvestigingen is not null then 1 else 0 end +
case when telefoonnummer is not null then 1 else 0 end +
case when fax is not null then 1 else 0 end +
case when email is not null then 1 else 0 end +
case when website is not null then 1 else 0 end +
case when twitter is not null then 1 else 0 end +
case when facebook is not null then 1 else 0 end +
case when youtube is not null then 1 else 0 end +
case when linkedin is not null then 1 else 0 end +
case when hoofdrubriek is not null then 1 else 0 end +
case when subrubrieken is not null then 1 else 0 end +
case when merken is not null then 1 else 0 end +
case when diensten is not null then 1 else 0 end +
case when productsoorten is not null then 1 else 0 end +
case when brancheverenigingen is not null then 1 else 0 end
DESC";


Maybe bedrijfslogo is empty in the table? :)
If you add the casing to a view instead it might be easier to find the error.
create view myView as
select *, (CASE when ID is not null then 1 else 0 end + ... n) AS SortPoints
from table
GO
select * from myView order by SortPoints desc

Related

Separating days from MySQL field

Good day, I have a problem wherein the value of days are formatted with MTWThF, ThF, TTh and so on. I am trying to convert into columns. this is my code:
SELECT
dys,
CASE WHEN dys LIKE 'M%' THEN 'M' ELSE '' END AS M,
CASE WHEN (dys LIKE '%T%' AND NOT '%Th%') THEN 'T' ELSE '' END AS T,
CASE WHEN dys LIKE '%W%' THEN 'W' ELSE '' END AS W,
CASE WHEN dys LIKE '%Th%' AND NOT 'T' THEN 'Th' ELSE '' END AS th,
CASE WHEN dys LIKE '%F%' THEN 'F' ELSE '' END AS F,
CASE WHEN (dys LIKE '%S%' AND NOT 'Su') THEN 'S' ELSE '' END AS S,
CASE WHEN (dys LIKE '%Su%' AND NOT 'S') THEN 'Su' ELSE '' END AS Su
FROM sched
GROUP BY dys
and the results are
Tb result
T is till detected on ThF same S from SSu. Do i missing something? Thank you

MySQL average of non-null columns

I've kludged my way through by doing this, but I feel there's probably a cleaner/faster/better solution, and am always keen to learn:
UPDATE myTable SET fAvg =
(COALESCE(f1,0) + COALESCE(f2,0) + COALESCE(f3,0) + COALESCE(f4,0))
/
(
CASE WHEN f1 IS NOT NULL THEN 1 ELSE 0 END +
CASE WHEN f2 IS NOT NULL THEN 1 ELSE 0 END +
CASE WHEN f3 IS NOT NULL THEN 1 ELSE 0 END +
CASE WHEN f4 IS NOT NULL THEN 1 ELSE 0 END
)
So, how can that be improved?

You can simplify the count by taking advantage of the fact the MySQL treats booleans as integers in a numeric context:
UPDATE myTable
SET fAvg = (COALESCE(f1,0) + COALESCE(f2,0) + COALESCE(f3,0) + COALESCE(f4,0)) /
((f1 IS NOT NULL) + (f2 IS NOT NULL) + (f3 IS NOT NULL) + (F4 IS NOT NULL));

MYSQL - Truncated incorrect DOUBLE value: '-'

This UPDATE fails with the error "Truncated incorrect DOUBLE value: '-'" and I cannot figure out why.
UPDATE psych
SET pdqp_adm=CAST((CAST(pdqt_adm AS SIGNED)) - (CAST(pdqf_adm AS SIGNED)) AS CHAR)
WHERE pdqt_adm>0
AND pdqt_adm IS NOT NULL
AND pdqf_adm>0
AND pdqf_adm IS NOT NULL
AND pdqt_adm>=pdqf_adm
All of the columns used here (pdqp_adm, pdqt_adm, pdqf_adm) are VARCHAR(6). I can do this query and the calculation works just fine:
SELECT CAST((CAST(pdqt_adm AS SIGNED)) - (CAST(pdqf_adm AS SIGNED)) AS CHAR)
FROM psych
WHERE pdqt_adm>0
AND pdqt_adm IS NOT NULL
AND pdqf_adm>0
AND pdqf_adm IS NOT NULL
AND pdqt_adm>=pdqf_adm

Ok, this error had nothing to do with the calculated values.
When I run this, I get this same error (only for record with ID 4972):
SELECT p.id, p.pdqt_adm, p.pdqf_adm
FROM psych p
WHERE p.pdqt_adm>0
AND p.pdqt_adm IS NOT NULL
AND p.pdqf_adm>0
AND p.pdqf_adm IS NOT NULL
AND p.id=4972
As it turns out, the 2 columns being used to compare to 0 and to each other both contain the value "-". Now, why this does not affect my SELECT in my question above, but does affect my UPDATE...I have no idea.

The update and select do not necessarily process the table in the same order. You can try:
UPDATE psych
SET pdqp_adm=CAST((CAST(pdqt_adm AS SIGNED)) - (CAST(pdqf_adm AS SIGNED)) AS CHAR)
WHERE (case when pdqt_adm <> '-' then pdqt_adm else 0 end) > 0 AND
pdqt_adm IS NOT NULL AND
(case when pdqf_adm <> '-' then pdqf_adm else 0 end) > 0 AND
pdqf_adm IS NOT NULL AND
(case when pdqt_adm <> '-' then pdqt_adm else 0 end) >= (case when pdqf_adm <> '-' then pdqf_adm else 0 end);

How to remove the seperator if all values are NULL in MySQL CONCAT_WS?

I am trying to use CONCAT_WS to generate a subject line. However, I want to ignore the seperator if a value is null.
Please look at this query
SELECT
CONCAT_WS(" ",
CASE WHEN total_attempts > 0 THEN
CONCAT( call_code_title , " - ", total_attempts, "<sup>",
CASE WHEN total_attempts = 2 THEN "nd"
WHEN total_attempts = 3 THEN "rd"
ELSE "th" END
, "</sup> attempt") ELSE call_code_title END
, "-", program_name) AS callSubject
FROM table
The problem is when "program_name" IS NULL then I always will have "-" at the end of the string. I want to not concat "-" id program_name IS NULL
How do I do that?
Thanks

IFNULL(program_name, concat("-", program_name))
SELECT
CONCAT_WS(" ",
CASE WHEN total_attempts > 0 THEN
CONCAT( call_code_title , " - ", total_attempts, "<sup>",
CASE WHEN total_attempts = 2 THEN "nd"
WHEN total_attempts = 3 THEN "rd"
ELSE "th" END
, "</sup> attempt") ELSE call_code_title END
, IFNULL(program_name, concat("-", program_name)) , " " ) AS callSubject
FROM table

SELECT
CONCAT_WS(" ",
CASE WHEN total_attempts > 0 THEN
CONCAT( call_code_title , " - ", total_attempts, "<sup>",
CASE WHEN total_attempts = 2 THEN "nd"
WHEN total_attempts = 3 THEN "rd"
ELSE "th" END
, "</sup> attempt") ELSE call_code_title END
, CONCAT("-", program_name) ) AS callSubject
FROM table
If program_name is NULL, then CONCAT("-", program_name) will also return NULL, which CONCAT_WS will handle correctly by ignoring what it sees as a single null argument.
CONCAT() returns NULL if any argument is NULL.
— http://dev.mysql.com/doc/refman/5.6/en/string-functions.html#function_concat

SSRS 2008 R2 Get Human Readable Schedule Information from ReportServer DB

I am looking to extract "human readable" schedule information from the ReportServer.dbo.Schedule table using t-sql.
An example of "human readable" follows.
At 6:02 AM every Sun, Mon, Tue, Wed, Thu, Fri, Sat of every week, starting 2/28/2011
There are a bunch of numeric fields in the table which are used to store the schedule, but I would like to convert those to words, as in my example.
Has anyone ever done this with reporting services?

SQL is not great for string manipulation or bitwise operations, and parsing this table requires a moderate bit of both. I'm sure SSRS doesn't do this in SQL: I probably could have written this in half the time and half the lines in C#.
USE ReportServer;
WITH EnhancedSched
AS (
SELECT
dbo.Schedule.ScheduleID ,
dbo.Schedule.Name ,
dbo.Schedule.StartDate ,
dbo.Schedule.Flags ,
dbo.Schedule.NextRunTime ,
dbo.Schedule.LastRunTime ,
dbo.Schedule.EndDate ,
dbo.Schedule.RecurrenceType ,
dbo.Schedule.MinutesInterval ,
dbo.Schedule.DaysInterval ,
dbo.Schedule.WeeksInterval ,
dbo.Schedule.DaysOfWeek ,
dbo.Schedule.DaysOfMonth ,
dbo.Schedule.Month ,
dbo.Schedule.MonthlyWeek ,
dbo.Schedule.State ,
dbo.Schedule.LastRunStatus ,
dbo.Schedule.ScheduledRunTimeout ,
dbo.Schedule.CreatedById ,
dbo.Schedule.EventType ,
dbo.Schedule.EventData ,
dbo.Schedule.Type ,
dbo.Schedule.ConsistancyCheck ,
dbo.Schedule.Path ,
CASE WHEN DaysOfWeek & 1 <> 0 THEN 'Sun, '
ELSE ''
END + CASE WHEN DaysOfWeek & 2 <> 0 THEN 'Mon, '
ELSE ''
END + CASE WHEN DaysOfWeek & 4 <> 0 THEN 'Tue, '
ELSE ''
END + CASE WHEN DaysOfWeek & 8 <> 0 THEN 'Wed, '
ELSE ''
END
+ CASE WHEN DaysOfWeek & 16 <> 0 THEN 'Thu, '
ELSE ''
END + CASE WHEN DaysOfWeek & 32 <> 0 THEN 'Fri, '
ELSE ''
END + CASE WHEN DaysOfWeek & 64 <> 0 THEN 'Sat, '
ELSE ''
END AS DaysOfWeekString ,
CASE WHEN DaysOfMonth & 1 <> 0 THEN '1,'
ELSE ''
END + CASE WHEN DaysOfMonth & 2 <> 0 THEN '2,'
ELSE ''
END + CASE WHEN DaysOfMonth & 4 <> 0 THEN '3,'
ELSE ''
END + CASE WHEN DaysOfMonth & 8 <> 0 THEN '4,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 16 <> 0 THEN '5,'
ELSE ''
END + CASE WHEN DaysOfMonth & 32 <> 0 THEN '6,'
ELSE ''
END + CASE WHEN DaysOfMonth & 64 <> 0 THEN '7,'
ELSE ''
END + CASE WHEN DaysOfMonth & 128 <> 0 THEN '8,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 256 <> 0 THEN '9,'
ELSE ''
END + CASE WHEN DaysOfMonth & 512 <> 0 THEN '10,'
ELSE ''
END + CASE WHEN DaysOfMonth & 1024 <> 0 THEN '11,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 2048 <> 0 THEN '12,'
ELSE ''
END + CASE WHEN DaysOfMonth & 4096 <> 0 THEN '13,'
ELSE ''
END + CASE WHEN DaysOfMonth & 8192 <> 0 THEN '14,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 16384 <> 0 THEN '15,'
ELSE ''
END + CASE WHEN DaysOfMonth & 32768 <> 0 THEN '16,'
ELSE ''
END + CASE WHEN DaysOfMonth & 65536 <> 0 THEN '17,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 131072 <> 0 THEN '18,'
ELSE ''
END + CASE WHEN DaysOfMonth & 262144 <> 0 THEN '19,'
ELSE ''
END + CASE WHEN DaysOfMonth & 524288 <> 0 THEN '20,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 1048576 <> 0 THEN '21,'
ELSE ''
END + CASE WHEN DaysOfMonth & 2097152 <> 0 THEN '22,'
ELSE ''
END + CASE WHEN DaysOfMonth & 4194304 <> 0 THEN '23,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 8388608 <> 0 THEN '24,'
ELSE ''
END + CASE WHEN DaysOfMonth & 16777216 <> 0 THEN '25,'
ELSE ''
END + CASE WHEN DaysOfMonth & 33554432 <> 0 THEN '26,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 67108864 <> 0 THEN '27,'
ELSE ''
END + CASE WHEN DaysOfMonth & 134217728 <> 0 THEN '28,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 268435456 <> 0 THEN '29,'
ELSE ''
END + CASE WHEN DaysOfMonth & 536870912 <> 0 THEN '30,'
ELSE ''
END
+ CASE WHEN DaysOfMonth & 1073741824 <> 0 THEN '31,'
ELSE ''
END AS DaysOfMonthString ,
CASE WHEN Month = 4095 THEN 'every month, '
ELSE CASE WHEN Month & 1 <> 0 THEN 'Jan, '
ELSE ''
END + CASE WHEN Month & 2 <> 0 THEN 'Feb, '
ELSE ''
END + CASE WHEN Month & 4 <> 0 THEN 'Mar, '
ELSE ''
END
+ CASE WHEN Month & 8 <> 0 THEN 'Apr, '
ELSE ''
END + CASE WHEN Month & 16 <> 0 THEN 'May, '
ELSE ''
END + CASE WHEN Month & 32 <> 0 THEN 'Jun, '
ELSE ''
END
+ CASE WHEN Month & 64 <> 0 THEN 'Jul, '
ELSE ''
END + CASE WHEN Month & 128 <> 0 THEN 'Aug, '
ELSE ''
END
+ CASE WHEN Month & 256 <> 0 THEN 'Sep, '
ELSE ''
END + CASE WHEN Month & 512 <> 0 THEN 'Oct, '
ELSE ''
END
+ CASE WHEN Month & 1024 <> 0 THEN 'Nov, '
ELSE ''
END + CASE WHEN Month & 2048 <> 0 THEN 'Dec, '
ELSE ''
END
END AS MonthString ,
CASE MonthlyWeek
WHEN 1 THEN 'first'
WHEN 2 THEN 'second'
WHEN 3 THEN 'third'
WHEN 4 THEN 'fourth'
WHEN 5 THEN 'last'
END AS MonthlyWeekString ,
' starting ' + CONVERT (VARCHAR, StartDate, 101)
+ CASE WHEN EndDate IS NOT NULL
THEN ' and ending ' + CONVERT (VARCHAR, EndDate, 101)
ELSE ''
END AS StartEndString ,
CASE CONVERT(VARCHAR, DATEPART(HOUR, StartDate) % 12)
WHEN 0 THEN '12'
ELSE CONVERT(VARCHAR, DATEPART(HOUR, StartDate) % 12)
END + ':'
+ CASE WHEN DATEPART(MINUTE, StartDate) < 10
THEN '0' + CONVERT(VARCHAR(2), DATEPART(MINUTE,
StartDate))
ELSE CONVERT(VARCHAR(2), DATEPART(MINUTE, StartDate))
END + CASE WHEN DATEPART(HOUR, StartDate) >= 12 THEN ' PM'
ELSE ' AM'
END AS StartTime
FROM
Schedule
),
SuperEnhancedSchedule
AS (
SELECT
EnhancedSched.ScheduleID ,
EnhancedSched.Name ,
EnhancedSched.StartDate ,
EnhancedSched.Flags ,
EnhancedSched.NextRunTime ,
EnhancedSched.LastRunTime ,
EnhancedSched.EndDate ,
EnhancedSched.RecurrenceType ,
EnhancedSched.MinutesInterval ,
EnhancedSched.DaysInterval ,
EnhancedSched.WeeksInterval ,
EnhancedSched.DaysOfWeek ,
EnhancedSched.DaysOfMonth ,
EnhancedSched.Month ,
EnhancedSched.MonthlyWeek ,
EnhancedSched.State ,
EnhancedSched.LastRunStatus ,
EnhancedSched.ScheduledRunTimeout ,
EnhancedSched.CreatedById ,
EnhancedSched.EventType ,
EnhancedSched.EventData ,
EnhancedSched.Type ,
EnhancedSched.ConsistancyCheck ,
EnhancedSched.Path , -- spec what you need.
CASE WHEN RecurrenceType = 1
THEN 'At ' + StartTime + ' on '
+ CONVERT(VARCHAR, StartDate, 101)
WHEN RecurrenceType = 2
THEN 'Every ' + CONVERT(VARCHAR, ( MinutesInterval / 60 ))
+ ' hour(s) and '
+ CONVERT(VARCHAR, ( MinutesInterval % 60 ))
+ ' minute(s), ' + 'starting '
+ CONVERT (VARCHAR, StartDate, 101) + ' at '
+ SUBSTRING(CONVERT(VARCHAR, StartDate, 8), 0, 6)
+ ' ' + SUBSTRING(CONVERT(VARCHAR, StartDate, 109),
25, 2)
+ CASE WHEN EndDate IS NOT NULL
THEN ' and ending '
+ CONVERT (VARCHAR, EndDate, 101)
ELSE ''
END
WHEN RecurrenceType = 3
THEN 'At ' + StartTime + ' every '
+ CASE DaysInterval
WHEN 1 THEN 'day, '
ELSE CONVERT(VARCHAR, DaysInterval) + ' days, '
END + StartEndString
WHEN RecurrenceType = 4
THEN 'At ' + StartTime + ' every '
+ CASE WHEN LEN(DaysOfWeekString) > 1
THEN LEFT(DaysOfWeekString,
LEN(DaysOfWeekString) - 1)
ELSE ''
END + ' of every '
+ CASE WHEN WeeksInterval = 1 THEN ' week,'
ELSE CONVERT(VARCHAR, WeeksInterval)
+ ' weeks,'
END + StartEndString
WHEN RecurrenceType = 5
THEN 'At ' + StartTime + ' on day(s) '
+ CASE WHEN LEN(DaysOfMonthString) > 1
THEN LEFT(DaysOfMonthString,
LEN(DaysOfMonthString) - 1)
ELSE ''
END + ' of ' + MonthString + StartEndString
WHEN RecurrenceType = 6
THEN 'At ' + StartTime + ' on the ' + MonthlyWeekString
+ ' '
+ CASE WHEN LEN(DaysOfWeekString) > 1
THEN LEFT(DaysOfWeekString,
LEN(DaysOfWeekString) - 1)
ELSE ''
END + ' of ' + MonthString + StartEndString
ELSE 'At ' + SUBSTRING(CONVERT(VARCHAR, StartDate, 8), 0,
6) + ' '
+ SUBSTRING(CONVERT(VARCHAR, StartDate, 109), 25, 2)
+ StartEndString
END ScheduleTextDefinition
FROM
EnhancedSched
)
SELECT
*
-- This has the same columns as the native [dbo].Schedule table plus a field called "SheduleTextDefinition"
-- You can use "SuperEnhancedSchedule" in place of the usual SSRS.Schedule table, joining to subscriptions and such.
FROM
SuperEnhancedSchedule

This post may be very old, but it helped me out today!
I found 2 items that I would like to add in the excellent post above by Jamie F for his CTE.
There was a missing entry for day 31 that needs to be added in as part of the EnhancedSched CTE, the below needs to be added to the end of the 'DaysOfMonthString'
+ CASE WHEN DaysOfMonth & 1073741824 <> 0 THEN '31,' ELSE '' END
Also, the 'StartTime' column definition with the modulo 12 makes any time beginning with 12 a zero, so the plain English result shows a start time of 0:30 PM for something that is supposed to say 12:30 pm.
Replace the
CONVERT(VARCHAR, DATEPART(hour, StartDate) % 12)
with
CASE CONVERT(VARCHAR, DATEPART(hour, StartDate) % 12) WHEN 0 THEN '12' ELSE CONVERT(VARCHAR, DATEPART(hour, StartDate) % 12) END
to get the plain English start time to read properly.
Mega thanks to Jamie F's post above, saved my bacon. +1 internets for you good sir.
Sorry for the extra 'answer' post, no rep to comment to up-vote Jamie F's excellent post above.

There's actually a stored procedure in the MSDB database called sp_get_schedule_description that can generate the schedule descriptions. I have the code below writting the schedule ID and the human-readable description to a ScheduleInfo user table. It works very well but the user running the code will need read-access to the msdb database and execute permissions to the SP for it to work.
DECLARE #schedule_description NVARCHAR(255)
DECLARE #freq_type INT
DECLARE #freq_interval INT
DECLARE #freq_subday_type INT
DECLARE #freq_subday_interval INT
DECLARE #freq_relative_interval INT
DECLARE #freq_recurrence_factor INT
DECLARE #active_start_date INT
DECLARE #active_end_date INT
DECLARE #active_start_time INT
DECLARE #active_end_time INT
DECLARE #schedule_id_as_char VARCHAR(10)
DECLARE #scheduleID UNIQUEIDENTIFIER
DECLARE #resultCursor CURSOR
-- Create cursor using records from job schedules in MSDB database
--
SET #resultCursor = CURSOR FOR
SELECT
d.freq_type
,d.freq_interval
,d.freq_subday_type
,d.freq_subday_interval
,d.freq_relative_interval
,d.freq_recurrence_factor
,d.active_start_date
,d.active_end_date
,d.active_start_time
,d.active_end_time
,a.ScheduleID
FROM ReportServer.dbo.Schedule a
JOIN msdb.dbo.sysjobs b on CONVERT(NVARCHAR(128),a.ScheduleID) = b.name
JOIN msdb.dbo.sysjobschedules c on b.job_id = c.job_id
JOIN msdb.dbo.sysschedules d on c.schedule_id = d.schedule_id
OPEN #resultCursor
-- Fetch first record from cursor
--
FETCH NEXT
FROM #resultCursor INTO
#freq_type
,#freq_interval
,#freq_subday_type
,#freq_subday_interval
,#freq_relative_interval
,#freq_recurrence_factor
,#active_start_date
,#active_end_date
,#active_start_time
,#active_end_time
,#scheduleID
-- Loop through cursor and get the rest of the records
--
WHILE ##FETCH_STATUS = 0
BEGIN
-- Call stored prc in MSDB database to get schedule description
--
EXECUTE msdb.dbo.sp_get_schedule_description
#freq_type,
#freq_interval,
#freq_subday_type,
#freq_subday_interval,
#freq_relative_interval,
#freq_recurrence_factor,
#active_start_date,
#active_end_date,
#active_start_time,
#active_end_time,
#schedule_description OUTPUT
-- Insert record to ScheduleInfo table
--
INSERT INTO ScheduleInfo VALUES (#scheduleID, #schedule_description)
-- Get the next record from the cursor
--
FETCH NEXT
FROM #resultCursor INTO
#freq_type
,#freq_interval
,#freq_subday_type
,#freq_subday_interval
,#freq_relative_interval
,#freq_recurrence_factor
,#active_start_date
,#active_end_date
,#active_start_time
,#active_end_time
,#scheduleID
END
--Close cursor
--
CLOSE #resultCursor
DEALLOCATE #resultCursor

I have a solution for this as it came up for a report I am writing.
create function [dbo].[calendarlist](#Value_in as int,#Type as int) returns varchar(200)
as
begin
/*
This code is to work out either the day of the week or the name of a month when given a value
Wrriten by S Manson.
31/01/2012
*/
declare #strings as varchar(200)
declare #Count int
if #Type = 2 --Months
Begin
set #Count =12
end
else if #Type = 1 --Days of Week
Begin
Set #Count = 7
End
else --Days of Month
Begin
Set #Count = 31
End
set #strings = ''
while #Count<>0
begin
if #Value_in>=(select power(2,#count-1))
begin
set #Value_in = #Value_in - (select power(2,#count-1))
If #Type=2
Begin
set #strings = (SELECT DATENAME(mm, DATEADD(month, #count-1, CAST('2008-01-01' AS datetime)))) + ',' + #strings
end
else if #Type = 1
begin
set #strings = (SELECT DATENAME(dw, DATEADD(day, #count-1, CAST('2012-01-01' AS datetime)))) + ',' + #strings
end
else
begin
set #strings = convert(varchar(2),#Count) + ', ' + #strings
end
end
set #count = #count-1
end
if right(#strings,1)=','
set #strings = left(#strings,len(#strings)-1)
return #strings
end