Related
hello I would like to find a solution to solve my problem in a single request if possible.
For the moment I take all the records then I go through the lines one by one to eliminate what I don't want.
I have 2 tables : first table with links
the second with the prefered label for the url
the second table must be consulted keeping only the row with maximum priority
priority rules are
the current user then
the user group and finally
everyone.
if the hidden column is true, exclude any reference to the url
here is the expected result.
Unfortunately, I don't see any other solution than to multiply the conditions on several selects and unions.
if you have a idea to solve my problem, thank you in advance for your help
It appears as though you can rely on pref_id for the preference ordering, correct? If so, you could try:
SELECT *
FROM table2
INNER JOIN table1 ON table2.url_id = table1.url_id
QUALIFY ROW_NUMBER() OVER (
PARTITION BY table1.url
ORDER BY pref_id ASC
) = 1
This will partition by the url and then provide only the one with lowest pref_id.
I didn't test this SQL as I wasn't sure which RDBMS you're running on, but I used Rasgo to translate the SQL.
maybe of interest in this tricky query:
select so.*, table1.url from
(select distinct t.url_id,
(select pref_id from table2 s where s.url_id = t.url_id order by "user" is null, "group" is null limit 1) pref_id
from table2 t
where not exists(select 1 from table2 s where s.hide and s.url_id = t.url_id)
) ids
join table2 so on so.pref_id = ids.pref_id
join table1 ON table1.url_id = ids.url_id
order by so.url_id;
here is my solution but i think there is better to do.
in the condition's select, I built a column which gives a level note according to the priorities
DECLARE #CUR_USER VARCHAR(10) = 'ROBERT'
DECLARE #CUR_GROUP VARCHAR(10) = 'DEV'
DECLARE #TABLE1 TABLE (
URL_ID INT
,URLNAME VARCHAR(100)
);
DECLARE #TABLE2 TABLE (
PREF_ID INT
,URL_ID INT
,FAVORITE_LABEL VARCHAR(100)
,USER_GROUP VARCHAR(10)
,USER_CODE VARCHAR(10)
,HIDE_URL DECIMAL(1, 0) DEFAULT 0
);
INSERT INTO #TABLE1
VALUES
(1, 'https://stackoverflow.com/')
,(2, 'https://www.microsoft.com/')
,(3, 'https://www.apple.com/')
,(4, 'https://www.wikipedia.org/')
;
INSERT INTO #TABLE2
VALUES
(1000, 1, 'find everything', NULL, 'ROBERT', 0)
,(1001, 1, 'a question ? find the answer here', 'DEV', NULL, 0)
,(1002, 1, 'StackOverFlow', NULL, NULL, 0)
,(1003, 2, 'Windows', 'DEV', NULL, 0)
,(1004, 2, 'Microsoft', NULL, NULL, 0)
,(1005, 3, 'Apple', NULL, NULL, 0)
,(1006, 4, 'Free encyclopedia', NULL, 'ROBERT', 1)
,(1007, 4, 'Wikipedia', NULL, NULL, 0)
,(1008, 1, 'StackOverFlow FOR MAT', 'MAT', NULL, 0)
,(1009, 2, 'Microsoft FOR MAT', 'MAT', NULL, 0)
,(1010, 3, 'Apple', 'MAT', NULL, 1)
,(1011, 4, 'Wikipedia FOR MAT', 'MAT', NULL, 0)
,(1012, 1, 'StackOverFlow', NULL, 'JEAN', 1)
,(1013, 2, 'Microsoft ', NULL, 'JEAN', 0)
,(1014, 3, 'Apple', NULL, 'JEAN', 0)
,(1015, 4, 'great encyclopedia', NULL, 'JEAN', 0)
;
SELECT t2.* ,t1.URLName
FROM #TABLE1 t1
INNER JOIN #TABLE2 t2 ON t1.URL_ID = t2.URL_ID
WHERE EXISTS (
SELECT 1
FROM (
SELECT TOP (1) test.PREF_ID
,CASE
-- if I do not comment this case: jean from the MAT group will not see apple
-- WHEN Hide_Url = 1
-- THEN 3
WHEN USER_code IS NOT NULL
THEN 2
WHEN USER_GROUP IS NOT NULL
THEN 1
ELSE 0
END AS ROW_LEVEL
FROM #TABLE2 test
WHERE (
(
test.USER_GROUP IS NULL
AND test.user_group IS NULL
AND test.USER_code IS NULL
)
OR (test.USER_GROUP = #CUR_GROUP)
OR (test.USER_code = #CUR_USER)
)
AND t2.URL_ID = test.URL_ID
ORDER BY ROW_LEVEL DESC
) test
WHERE test.PREF_ID = t2.PREF_ID
AND Hide_Url = 0
)
Simply use an ORDER BY clause that puts the preferred row first. You can use this in the window function ROW_NUMBER and work with this or use a lateral top(1) join with CROSS APPLY.
select *
from urls
cross apply
(
select top(1) *
from labels
where labels.url_id = urls.url_id
where [Group] is not null or [user] is not null or hide is not null
order by
case when [Group] is null then 2 else 1 end,
case when [user] is null then 2 else 1 end,
case when hide is null then 2 else 1 end
) top_labels
order by urls.url_id;
I try to find a way to let the SGBD perform a population of merge fields within a long text.
Create the structure :
CREATE TABLE [dbo].[store]
(
[id] [int] NOT NULL,
[text] [nvarchar](MAX) NOT NULL
)
CREATE TABLE [dbo].[statement]
(
[id] [int] NOT NULL,
[store_id] [int] NOT NULL
)
CREATE TABLE [dbo].[statement_merges]
(
[statement_id] [int] NOT NULL,
[merge_field] [nvarchar](30) NOT NULL,
[user_data] [nvarchar](MAX) NOT NULL
)
Now, create test values
INSERT INTO [store] (id, text)
VALUES (1, 'Waw, stackoverflow is an amazing library of lost people in the IT hell, and i have the feeling that $$PERC_SAT$$ of the users found a solution, personally I asked $$ASKED$$ questions.')
INSERT INTO [statement] (id, store_id)
VALUES (1, 1)
INSERT INTO [statement_merges] (statement_id, merge_field, user_data)
VALUES (1, '$$PERC_SAT$$', '85%')
INSERT INTO [statement_merges] (statement_id, merge_field, user_data)
VALUES (1, '$$ASKED$$', '12')
At the time being my app is delivering the final statement, looping through merges, replacing in the stored text and output
Waw, stackoverflow is an amazing library of lost people in the IT
hell, and i have the feeling that 85% of the users found a solution,
personally I asked 12 questions.
I try to find a way to be code-independent and serve the output in a single query, as u understood, select a statement in which the stored text have been populated with user data. I hope I'm clear.
I looked on TRANSLATE function but it looks like a char replacement, so I have two choices :
I try a recursive function, replacing one by one until no merge_fields is found in the calculated text; but I have doubts about the performance of this approach;
There is a magic to do that but I need your knowledge...
Consider that I want this because the real texts are very long, and I don't want to store it more than once in my database. You can imagine a 3 pages contract with only 12 parameters, like start date, invoiced amount, etc... Everything else cant be changed for compliance.
Thank you for your time!
EDIT :
Thanks to Randy's help, this looks to do the trick :
WITH cte_replace_tokens AS (
SELECT replace(r.text, m.merge_field, m.user_data) as [final], m.merge_field, s.id, 1 AS i
FROM store r
INNER JOIN statement s ON s.store_id = r.id
INNER JOIN statement_merges m ON m.statement_id = s.id
WHERE m.statement_id = 1
UNION ALL
SELECT replace(r.final, m.merge_field, m.user_data) as [final], m.merge_field, r.id, r.i + 1 AS i
FROM cte_replace_tokens r
INNER JOIN statement_merges m ON m.statement_id = r.id
WHERE m.merge_field > r.merge_field
)
select TOP 1 final from cte_replace_tokens ORDER BY i DESC
I will check with a bigger database if the performance is good...
At least, I can "populate" one statement, I need to figure out to be able to extract a list as well.
Thanks again !
If a record is updated more than once by the same update, the last wins. None of the updates are affected by the others - no cumulative effect. It is possible to trick SQL using a local variable to get cumulative effects in some cases, but it's tricky and not recommended. (Order becomes important and is not reliable in an update.)
One alternate is recursion in a CTE. Generate a new record from the prior as each token is replaced until there are no tokens. Here is a working example that replaces 1 with A, 2 with B, etc. (I wonder if there is some tricky xml that can do this as well.)
if not object_id('tempdb..#Raw') is null drop table #Raw
CREATE TABLE #Raw(
[test] [varchar](100) NOT NULL PRIMARY KEY CLUSTERED,
)
if not object_id('tempdb..#Token') is null drop table #Token
CREATE TABLE #Token(
[id] [int] NOT NULL PRIMARY KEY CLUSTERED,
[token] [char](1) NOT NULL,
[value] [char](1) NOT NULL,
)
insert into #Raw values('123456'), ('1122334456')
insert into #Token values(1, '1', 'A'), (2, '2', 'B'), (3, '3', 'C'), (4, '4', 'D'), (5, '5', 'E'), (6, '6', 'F');
WITH cte_replace_tokens AS (
SELECT r.test, replace(r.test, l.token, l.value) as [final], l.id
FROM [Raw] r
CROSS JOIN #Token l
WHERE l.id = 1
UNION ALL
SELECT r.test, replace(r.final, l.token, l.value) as [final], l.id
FROM cte_replace_tokens r
CROSS JOIN #Token l
WHERE l.id = r.id + 1
)
select * from cte_replace_tokens where id = 6
It's not recommended to do such tasks inside sql engine but if you want to do that, you need to do it in a loop using cursor in a function or stored procedure like so :
DECLARE #merge_field nvarchar(30)
, #user_data nvarchar(MAX)
, #statementid INT = 1
, #text varchar(MAX) = 'Waw, stackoverflow is an amazing library of lost people in the IT hell, and i have the feeling that $$PERC_SAT$$ of the users found a solution, personally I asked $$ASKED$$ questions.'
DECLARE merge_statements CURSOR FAST_FORWARD
FOR SELECT
sm.merge_field
, sm.user_data
FROM dbo.statement_merges AS sm
WHERE sm.statement_id = #statementid
OPEN merge_statements
FETCH NEXT FROM merge_statements
INTO #merge_field , #user_data
WHILE ##FETCH_STATUS = 0
BEGIN
set #text = REPLACE(#text , #merge_field, #user_data )
FETCH NEXT FROM merge_statements
INTO #merge_field , #user_data
END
CLOSE merge_statements
DEALLOCATE merge_statements
SELECT #text
Here is a recursive solution.
SQL Fiddle
MS SQL Server 2017 Schema Setup:
CREATE TABLE [dbo].[store]
(
[id] [int] NOT NULL,
[text] [nvarchar](MAX) NOT NULL
)
CREATE TABLE [dbo].[statement]
(
[id] [int] NOT NULL,
[store_id] [int] NOT NULL
)
CREATE TABLE [dbo].[statement_merges]
(
[statement_id] [int] NOT NULL,
[merge_field] [nvarchar](30) NOT NULL,
[user_data] [nvarchar](MAX) NOT NULL
)
INSERT INTO store (id, text)
VALUES (1, '$$(*)$$, stackoverflow...$$PERC_SAT$$...$$ASKED$$ questions.')
INSERT INTO store (id, text)
VALUES (2, 'Use The #_#')
INSERT INTO statement (id, store_id) VALUES (1, 1)
INSERT INTO statement (id, store_id) VALUES (2, 2)
INSERT INTO statement_merges (statement_id, merge_field, user_data) VALUES (1, '$$PERC_SAT$$', '85%')
INSERT INTO statement_merges (statement_id, merge_field, user_data) VALUES (1, '$$ASKED$$', '12')
INSERT INTO statement_merges (statement_id, merge_field, user_data) VALUES (1, '$$(*)$$', 'Wow')
INSERT INTO statement_merges (statement_id, merge_field, user_data) VALUES (2, ' #_#', 'Flux!')
Query 1:
;WITH Normalized AS
(
SELECT
store_id=store.id,
store.text,
sm.merge_field,
sm.user_data,
RowNumber = ROW_NUMBER() OVER(PARTITION BY store.id,sm.statement_id ORDER BY merge_field),
statement_id = st.id
FROM
store store
INNER JOIN statement st ON st.store_id = store.id
INNER JOIN statement_merges sm ON sm.statement_id = st.id
)
, Recurse AS
(
SELECT
store_id, statement_id, old_text = text, merge_field,user_data, RowNumber,
Iteration=1,
new_text = REPLACE(text, merge_field, user_data)
FROM
Normalized
WHERE
RowNumber=1
UNION ALL
SELECT
n.store_id, n.statement_id, r.old_text, n.merge_field, n.user_data,
RowNumber=r.RowNumber+1,
Iteration=Iteration+1,
new_text = REPLACE(r.new_text, n.merge_field, n.user_data)
FROM
Normalized n
INNER JOIN Recurse r ON r.RowNumber = n.RowNumber AND r.statement_id = n.statement_id
)
,ReverseOnIteration AS
(
SELECT *,
ReverseIteration = ROW_NUMBER() OVER(PARTITION BY statement_id ORDER BY Iteration DESC)
FROM
Recurse
)
SELECT
store_id, statement_id, new_text, old_text
FROM
ReverseOnIteration
WHERE
ReverseIteration=1
Results:
| store_id | statement_id | new_text | old_text |
|----------|--------------|------------------------------------------|--------------------------------------------------------------|
| 1 | 1 | Wow, stackoverflow...85%...12 questions. | $$(*)$$, stackoverflow...$$PERC_SAT$$...$$ASKED$$ questions. |
| 2 | 2 | Use TheFlux! | Use The #_# |
With the help of Randy, I think I've achieved what I wanted to do !
Known the fact that my real case is a contract, in which there are several statements that may be :
free text
stored text without any merges
stored text with one or
several merges
this CTE does the job !
WITH cte_replace_tokens AS (
-- The initial query dont join on merges neither on store because can be a free text
SELECT COALESCE(r.text, s.part_text) AS [final], CAST('' AS NVARCHAR) AS merge_field, s.id, 1 AS i, s.contract_id
FROM statement s
LEFT JOIN store r ON s.store_id = r.id
UNION ALL
-- We loop till the last merge field, output contains iteration to be able to keep the last record ( all fields updated )
SELECT replace(r.final, m.merge_field, m.user_data) as [final], m.merge_field, r.id, r.i + 1 AS i, r.contract_id
FROM cte_replace_tokens r
INNER JOIN statement_merges m ON m.statement_id = r.id
WHERE m.merge_field > r.merge_field AND r.final LIKE '%' + m.merge_field + '%'
-- spare lost replacements by forcing only one merge_field per loop
AND NOT EXISTS( SELECT mm.statement_id FROM statement_merges mm WHERE mm.statement_id = m.statement_id AND mm.merge_field > r.merge_field AND mm.merge_field < m.merge_field)
)
select s.id,
(select top 1 final from cte_replace_tokens t WHERE t.contract_id = s.contract_id AND t.id = s.id ORDER BY i DESC) as res
FROM statement s
where contract_id = 1
If the CTE solution with a cross join is too slow, an alternate solution would be to build a scalar fn dynamically that has every REPLACE required from the token table. One scalar fn call per record then is order(N). I get the same result as before.
The function is simple and likely not to be too long, depending upon how big the token table becomes...256 MB batch limit. I've seen attempts to dynamically create queries to improve performance backfire - moved the problem to compile time. Should not be a problem here.
if not object_id('tempdb..#Raw') is null drop table #Raw
CREATE TABLE #Raw(
[test] [varchar](100) NOT NULL PRIMARY KEY CLUSTERED,
)
if not object_id('tempdb..#Token') is null drop table #Token
CREATE TABLE #Token(
[id] [int] NOT NULL PRIMARY KEY CLUSTERED,
[token] [char](1) NOT NULL,
[value] [char](1) NOT NULL,
)
insert into #Raw values('123456'), ('1122334456')
insert into #Token values(1, '1', 'A'), (2, '2', 'B'), (3, '3', 'C'), (4, '4', 'D'), (5, '5', 'E'), (6, '6', 'F');
DECLARE #sql varchar(max) = 'CREATE FUNCTION dbo.fn_ReplaceTokens(#raw varchar(8000)) RETURNS varchar(8000) AS BEGIN RETURN ';
WITH cte_replace_statement AS (
SELECT a.id, CAST('replace(#raw,''' + a.token + ''',''' + a.value + ''')' as varchar(max)) as [statement]
FROM #Token a
WHERE a.id = 1
UNION ALL
SELECT n.id, CAST(replace(l.[statement], '#raw', 'replace(#raw,''' + n.token + ''',''' + n.value + ''')') as varchar(max)) as [statement]
FROM #Token n
INNER JOIN cte_replace_statement l
ON n.id = l.id + 1
)
select #sql += [statement] + ' END' from cte_replace_statement where id = 6
print #sql
if not object_id('dbo.fn_ReplaceTokens') is null drop function dbo.fn_ReplaceTokens
execute (#sql)
SELECT r.test, dbo.fn_ReplaceTokens(r.test) as [final] FROM [Raw] r
I need to filter a table based in a sub table data.
I'll exemplify with a hypnotic data to be easier to explain:
Master table: Cars
Sub table: Attributes (like Color, car type, accessories)
These attributes have an id (idOption) and the selected value (idList)
So, in an example, I need to filter all the cars with the color (idOption = 10) yellow (idList = 45). I can't filter this directly because the search need to consider the other option's results (which include the types, accessories.
When I use NOT IN for just one table, it works. But when I use merging the 2 tables with INNER JOIN, it does not work.
So in summary, I need to filter the 3 idOption (when is not NULL) with a given value, and this needs to reflect in the main table, grouped by product.
Table Cars:
idProduct | Description
1 Product A
2 Product B
3 Product C
Table Attributes:
idRow idProduct idOption idList
---------------------------------------
1 1 10 45
2 2 10 46
3 3 10 47
4 1 11 10
5 2 11 98
6 1 14 56
7 3 16 28
8 2 20 55
This is the stored procedure that I created which is not working:
ALTER PROCEDURE [dbo].[SP_GET_TestSearch]
(#Param1 BIGINT = NULL,
#PValue1 BIGINT = NULL,
#Param2 BIGINT = NULL,
#PValue2 BIGINT = NULL,
#Param3 BIGINT = NULL,
#PValue3 BIGINT = NULL)
AS
SET NOCOUNT ON;
SELECT
Cars.idProduct,
Cars.[Description]
FROM
Cars
INNER JOIN
Attributes ON Cars.idProduct = Attributes.idProduct
WHERE
((#Param1 IS NULL OR (idOption NOT IN (#Param1)))
AND
(#Param2 IS NULL OR (idOption NOT IN (#Param2)))
AND
(#Param3 IS NULL OR (idOption NOT IN (#Param3))))
OR
(idOption = ISNULL(#Param1, NULL)
AND idList = ISNULL(#PValue1, NULL))
OR
(idOption = ISNULL(#Param2, NULL)
AND idList = ISNULL(#PValue2, NULL))
OR
(idOption = ISNULL(#Param3, NULL)
AND idList = ISNULL(#PValue3, NULL))
GROUP BY
Cars.idProduct, Cars.[Description]
The following code demonstrates how to implement the logic of excluding vehicles from query results if they have any "bad" property values. The rejection is handled by ... where not exists ... which is used to check each car against the "bad" property values.
Rather than using an assortment of (hopefully) paired parameters to pass the undesirable properties, the values are passed in a table. The stored procedure to implement this ought to use a table-valued parameter (TVP) to pass the table.
-- Sample data.
declare #Cars as Table ( CarId Int Identity, Description VarChar(16) );
insert into #Cars ( Description ) values
( 'Esplanade' ), ( 'Tankigator' ), ( 'Land Yacht' );
select * from #Cars;
declare #Properties as Table ( PropertyId Int Identity, Description VarChar(16) );
insert into #Properties ( Description ) values
( 'Turbochargers' ), ( 'Superchargers' ), ( 'Hyperchargers' ), ( 'Color' ), ( 'Spare Tires' );
select * from #Properties;
declare #CarProperties as Table ( CarId Int, PropertyId Int, PropertyValue Int );
insert into #CarProperties ( CarId, PropertyId, PropertyValue ) values
( 1, 1, 1 ), ( 1, 4, 24 ), ( 1, 4, 42 ), -- Two tone!
( 2, 2, 1 ), ( 2, 4, 7 ),
( 3, 1, 2 ), ( 3, 4, 0 ), ( 3, 5, 6 );
select C.CarId, C.Description as CarDescription,
P.PropertyId, P.Description as PropertyDescription,
CP.PropertyValue
from #Cars as C inner join
#CarProperties as CP on CP.CarId = C.CarId inner join
#Properties as P on P.PropertyId = CP.PropertyId
order by C.CarId, P.PropertyId;
-- Test data: Avoid vehicles that have _any_ of these property values.
-- This should be passed to the stored procedure as a table-value parameter (TVP).
declare #BadProperties as Table ( PropertyId Int, PropertyValue Int );
insert into #BadProperties ( PropertyId, PropertyValue ) values
( 2, 1 ), ( 2, 2 ), ( 2, 4 ),
( 4, 62 ), ( 4, 666 );
select BP.PropertyId, BP.PropertyValue, P.Description
from #BadProperties as BP inner join
#Properties as P on P.PropertyId = BP.PropertyId;
-- Query the data.
select C.CarId, C.Description as CarDescription
from #Cars as C
where not exists (
select 42
from #CarProperties as CP inner join
#BadProperties as BP on BP.PropertyId = CP.PropertyId and BP.PropertyValue = CP.PropertyValue
where CP.CarId = C.CarId )
order by C.CarId;
A few things here.
Firstly, this kind of catch all procedure is a bit of an anti pattern for all sorts of reasons, see here for a full explanation:- https://sqlinthewild.co.za/index.php/2018/03/13/revisiting-catch-all-queries/
Secondly, you need to be very careful of using NOT IN with nullable values in a list: http://www.sqlbadpractices.com/using-not-in-operator-with-null-values/
I've added the DDL for the tables:-
IF OBJECT_ID('Attributes') IS NOT NULL
DROP TABLE Attributes;
IF OBJECT_ID('Cars') IS NOT NULL
DROP TABLE Cars;
IF OBJECT_ID('SP_GET_TestSearch') IS NOT NULL
DROP PROCEDURE SP_GET_TestSearch
CREATE TABLE Cars
(idProduct INT PRIMARY KEY
, Description VARCHAR(20) NOT NULL);
CREATE TABLE Attributes
(idRow INT PRIMARY KEY
, idProduct INT NOT NULL FOREIGN KEY REFERENCES dbo.Cars(idProduct)
, idOption INT NOT NULL
, idList INT NOT NULL);
INSERT INTO dbo.Cars
VALUES
(1, 'Product A')
,(2 , 'Product B')
,(3, 'Product C');
INSERT INTO dbo.Attributes
(
idRow,
idProduct,
idOption,
idList
)
VALUES (1,1,10,45)
,(2,2,10,46)
,(3,3,10,47)
,(4,1,11,10)
,(5,2,11,98)
,(6,1,14,56)
,(7,3,16,28)
,(8,2,20,55);
GO
The issue with your query, is that the first part of the block is always evaluated to TRUE for any idOption that you don't specify:-
((#Param1 IS NULL OR (idOption NOT IN (#Param1)))
AND
(#Param2 IS NULL OR (idOption NOT IN (#Param2)))
AND
(#Param3 IS NULL OR (idOption NOT IN (#Param3))))
To explain; if I pass in the following:-
DECLARE #Param1 BIGINT
, #Param2 BIGINT
, #Param3 BIGINT
, #PValue1 BIGINT
, #PValue2 BIGINT
, #PValue3 BIGINT;
SET #Param1 = 11
SET #Pvalue1 = 42
SET #Param2 = 11
SET #Pvalue2 = 10
SET #Param3 = 14
SET #PValue3= 56
EXEC dbo.SP_GET_TestSearch #Param1, #PValue1, #Param2, #PValue2, #Param3, #PValue3
Then you effectively have WHERE idOption NOT IN (11,14) as the evaluation for the first part of the clause, so all other rows are returned.
I suspect you really want the WHERE clause to be:-
WHERE
(#Param1 IS NULL AND #Param2 IS NULL AND #Param3 IS NULL)
OR
(idOption = #Param1
AND idList = #PValue1)
OR
(idOption = #Param2
AND idList = #PValue2)
OR
(idOption = #Param3
AND idList = #PValue3)
The Structure
I have 2 tables that link to each other. One is a set of values and a nullable foreign key that points to the Id of the other table, which contains 2 foreign keys back to the other table.
HierarchicalTable
Id LeftId RightId SomeValue
1 1 2 some value
2 3 4 top level in tree
3 5 6 incorrect hierarchy 1
4 7 8 incorrect result top level
IntermediateTable
Id SomeValue HierarchicalTableId
1 some value NULL
2 value NULL
3 NULL 1
4 value NULL
5 incorrect result 1 NULL
6 incorrect result 3 NULL
7 incorrect result 3 NULL
8 NULL 3
Each table points down the hierarchy. Here is this structure graphed out for the Hierarchical Table records 1 & 2 and their IntermediateTable values:
(H : HierarchicalTable, I : IntermediateTable)
H-2
/ \
I-3 I-4
/
H-1
/ \
I-1 I-2
The Problem
I need to be able to send in an Id for a given HierarchicalTable and get all the HierarchicalTable records below it. So, for the structure above, if I pass 1 into a query, I should just get H-1 (and from that, I can load the related IntermediateTable values). If I pass 2, I should get H-2 and H-1 (and, again, use those to load the relevant IntermediateTable values).
The Attempts
I've tried using a CTE, but there are a few main things that are different from the examples I've seen:
In my structure, the objects point down to their children, instead of up to their parent
I have the Id of the top object, not the Id of the bottom object.
My hierarchy is split across 2 tables. This shouldn't be a big issue once I understand the algorithm to find the results I need, but this could be causing additional confusion for me.
If I run this query:
declare #TargetId bigint = 2
;
with test as (
select h.*
from dbo.hierarchicaltable h
inner join dbo.intermediatetable i
on (h.leftid = i.id or h.rightid = i.id)
union all
select h.*
from dbo.hierarchicaltable h
where h.id = #TargetId
)
select distinct *
from test
I get all 4 records in the HierarchicalTable, instead of just records 1 & 2. I'm not sure if what I want is possible to do with a CTE.
Try this:
I'm build entire tree with both tables, then filter (only hierarchicaltable records).
DECLARE #HierarchicalTable TABLE(
Id INT,
LeftId INT,
RightId INT,
SomeValue VARCHAR(MAX)
)
INSERT INTO #HierarchicalTable
VALUES
(1, 1, 2, 'some value '),
(2, 3, 4, 'top level in tree '),
(3, 5, 6, 'incorrect hierarchy 1 '),
(4, 7, 8, 'incorrect result top level')
DECLARE #IntermediateTable TABLE(
Id INT,
SomeValue VARCHAR(MAX),
HierarchicalTableId INT
)
INSERT INTO #IntermediateTable
VALUES
(1, 'some value' ,NULL ),
(2, 'value ' ,NULL ),
(3, NULL ,1 ),
(4, 'value ' ,NULL ),
(5, 'incorrect result 1' ,NULL ),
(6, 'incorrect result 3' ,NULL ),
(7, 'incorrect result 3' ,NULL ),
(8, NULL ,3 )
DECLARE #TargetId INT = 2;
WITH CTE AS (
SELECT Id AS ResultId, LeftId, RightId, NULL AS HierarchicalTableId
FROM #HierarchicalTable
WHERE Id = #TargetId
UNION ALL
SELECT C.Id AS ResultId, C.LeftId, C.RightId, NULL AS HierarchicalTableId
FROM #HierarchicalTable C
INNER JOIN CTE P ON P.HierarchicalTableId = C.Id
UNION ALL
SELECT NULL AS ResultId, NULL AS LeftId, NULL AS RightId, C.HierarchicalTableId
FROM #IntermediateTable C
INNER JOIN CTE P ON P.LeftId = C.Id OR P.RightId = C.Id
)
SELECT *
FROM CTE
WHERE ResultId IS NOT NULL
I was wondering if I have SQL Server 2008 table that was created like this:
CREATE TABLE tbl (id INT PRIMARY KEY,
dvt NVARCHAR(32),
d0 TINYINT,
d1 TINYINT,
d2 TINYINT);
INSERT INTO tbl (id, dvt, d0, d1, d2)
VALUES(1, '1', NULL, NULL, NULL);
INSERT INTO tbl (id, dvt, d0, d1, d2)
VALUES(2, '', NULL, NULL, NULL);
INSERT INTO tbl (id, dvt, d0, d1, d2)
VALUES(3, '2,5', NULL, NULL, NULL);
INSERT INTO tbl (id, dvt, d0, d1, d2)
VALUES(4, '13, 34, 45, 5', NULL, NULL, NULL);
INSERT INTO tbl (id, dvt, d0, d1, d2)
VALUES(5, '1,8, 10', NULL, NULL, NULL);
I need to take the string from the 'dvt' column and split it into 'd0', 'd1' and 'd2' columns. The 'dvt' value can be separated by commas.
I can do this using C# and a tokenization function but I was wondering if it's possible to do the same using SQL?
Columns BEFORE:
1, "1", NULL, NULL, NULL
2, "", NULL, NULL, NULL
3, "2,5", NULL, NULL, NULL
4, "13, 34, 45, 5", NULL, NULL, NULL
5, "1,8, 10", NULL, NULL, NULL
Columns AFTER:
1, "1", 1, NULL, NULL
2, "", NULL, NULL, NULL
3, "2,5", 2, 5, NULL
4, "13, 34, 45, 5", 13, 34, 45 -- 5 is discarded
5, "1,8, 10", 1, 8, 10
The main problem with this type of code is re-use of calculations.
SQL Server is good at caching results (If you type the exact same CHARINDEX() caluculation 5 times, it only calculates once and re-uses that result 4 times).
That's little consolation for the poor coder who has to type or maintain that code though.
SQL Server 2005 onward has CROSS APPLY that does help somewhat. The logic is repeated, but the results can be referenced repeatedly, rather that the calculation typed repeatedly.
SELECT
*,
SUBSTRING(dvt, 1, ISNULL(comma1.pos-1, LEN(dvt)) ) AS item1,
SUBSTRING(dvt, comma1.pos+1, ISNULL(comma2.pos-1, LEN(dvt))-comma1.pos) AS item2,
SUBSTRING(dvt, comma2.pos+1, ISNULL(comma3.pos-1, LEN(dvt))-comma2.pos) AS item3
FROM
(
SELECT 'ab,c,def,hij' AS dvt
UNION ALL
SELECT 'xyz,abc' AS dvt
)
AS data
OUTER APPLY
(SELECT NULLIF(CHARINDEX(',', data.dvt, 1 ), 0) AS pos ) AS comma1
OUTER APPLY
(SELECT NULLIF(CHARINDEX(',', data.dvt, comma1.pos+1), 0) AS pos WHERE comma1.pos > 0) AS comma2
OUTER APPLY
(SELECT NULLIF(CHARINDEX(',', data.dvt, comma2.pos+1), 0) AS pos WHERE comma2.pos > 0) AS comma3
OUTER APPLY
(SELECT NULLIF(CHARINDEX(',', data.dvt, comma3.pos+1), 0) AS pos WHERE comma3.pos > 0) AS comma4
Another option is to simply write a table valued user defined function that does this (even when the result of the function is always one row). Then you simply CROSS APPLY that function.
Try something like this
;WITH Vals AS (
SELECT id,
dvt,
CAST('<r>'+REPLACE(dvt,',','</r><r>')+'</r>' AS XML).query('/r[1]').value('.','varchar(max)') d1,
CAST('<r>'+REPLACE(dvt,',','</r><r>')+'</r>' AS XML).query('/r[2]').value('.','varchar(max)') d2,
CAST('<r>'+REPLACE(dvt,',','</r><r>')+'</r>' AS XML).query('/r[3]').value('.','varchar(max)') d3
FROM tbl
)
SELECT id,
dvt,
CASE WHEN d1 = '' THEN NULL ELSE d1 END d1,
CASE WHEN d2 = '' THEN NULL ELSE d2 END d2,
CASE WHEN d3 = '' THEN NULL ELSE d3 END d3
FROM Vals
It's possible.
You could do it with some repetitive calls to CHARINDEX and checking for nulls but it may be better and clearer to write a FUNCTION to split the string.
I needed String tokenizer for Sybase; separation by 1 or more spaces in name data
Name date clean and has no commas or other special characters
declare #test varchar(60)
select #test=str_replace(lower(rtrim('Jayanta Narayan Choudhuri'))," ",",")
exec sp_splitwords #test
This is based on a neat hint by Kenny Lucas from http://www.sql9.com/?id=102
drop proc sp_splitwords
go
create proc sp_splitwords(#instr varchar(80)) as
begin
declare #pos int,
#word varchar(80),
#list varchar(81)
create table #words(word varchar(80))
select #list = #instr + ','
set #pos = patindex('%,,%',#list)
while #pos > 0
begin
select #list = str_replace(#list,',,',',')
set #pos = patindex('%,,%',#list)
end
set #pos = patindex('%,%',#list)
while #pos > 0
begin
set #word = substring(#list, 1,#pos-1)
set #list = substring(#list, #pos+1,len(#list)-#pos)
if NOT( #word is null OR LEN(#word) = 0 )
insert into #words (word) values (#word)
set #pos = patindex('%,%',#list)
end
select * from #words
order by len(word) desc
drop table #words
end
I could port Metaphone SQL Function to Sybase
http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=125724
Sybase allows recursion in functions from a beautiful workaround
http://www.sypron.nl/quiz2008a.html#jan08
CREATE FUNCTION Metaphone2 (#str VARCHAR(100))
RETURNS VARCHAR(25) AS
BEGIN
RETURN #str
END
DROP FUNCTION Metaphone2
GO
CREATE FUNCTION Metaphone2 (#str VARCHAR(100))
RETURNS VARCHAR(25) AS
BEGIN
RETURN dbo.Metaphone(#str)
END
Changed 1 line of Function pasted from http://www.sqlteam.com/forums/topic.asp?TOPIC_ID=125724
Combination of metaphone and string tokenizer means I can fuzzy search names
first middle and surname and rotations thereof