I am newbie to postgresql and I need a postgresql version of this function - could you help me please?
CREATE FUNCTION [dbo].[getpersonname]
(#commid INT)
RETURNS VARCHAR(255)
AS
BEGIN
DECLARE #personnname varchar(255)
SELECT
#personnname = COALESCE(#personname + ',', '') +
ISNULL(CONVERT(VARCHAR(50), pers.personnname ),'')
FROM
dbo.comlink comli
INNER JOIN
dbo.person pers ON personid= comli_personid
WHERE
comli.comli_commid = #commid
RETURN #personnname
END
If I understand that code correctly it returns all names separated with commas.
That can easily be done with a simple SQL statement in Postgres:
create FUNCTION get_person_name(p_commid int)
RETURNS text
AS
$$
SELECT string_agg(pers.personnname, ',')
FROM dbo.comlink comli
JOIN dbo.person pers ON personid = comli_personid
WHERE comli.comli_commid= p_commid;
$$
language sql;
Related
I am working on a query that I need to modify so that a string is passed to in(). The view table is being used by some other view table and ultimately by a stored procedure. The string values must be in ' '.
select region, county, name
from vw_main
where state - 'MD'
and building_id in ('101', '102') -- pass the string into in()
The values for the building_id will be entered at the stored procedure level upon its execution.
Please check below scripts which will give you answer.
Way 1: Split CSV value using XML and directly use select query in where condition
DECLARE #StrBuildingIDs VARCHAR(1000)
SET #StrBuildingIDs = '101,102'
SELECT
vm.region,
vm.county,
vm.name
FROM vw_main vm
WHERE vm.state = 'MD'
AND vm.building_id IN
(
SELECT
l.value('.','VARCHAR(20)') AS Building_Id
FROM
(
SELECT CAST('<a>' + REPLACE(#StrBuildingIDs,',','</a><a>') + '</a>') AS BuildIDXML
) x
CROSS APPLY x.BuildIDXML.nodes('a') Split(l)
)
Way 2: Split CSV value using XML, Create Variable Table and use that in where condition
DECLARE #StrBuildingIDs VARCHAR(1000)
SET #StrBuildingIDs = '101,102'
DECLARE #TblBuildingID TABLE(BuildingId INT)
INSERT INTO #TblBuildingID(BuildingId)
SELECT
l.value('.','VARCHAR(20)') AS Building_Id
FROM
(
SELECT CAST('<a>' + REPLACE(#StrBuildingIDs,',','</a><a>') + '</a>') AS BuildIDXML
) x
CROSS APPLY x.BuildIDXML.nodes('a') Split(l)
SELECT
vm.region,
vm.county,
vm.name
FROM vw_main AS vm
WHERE vm.state = 'MD'
AND vm.building_id IN
(
SELECT
BuildingId
FROM #TblBuildingID
)
Way 3: Split CSV value using XML, Create Variable Table and use that in INNER JOIN
Assuming the input string is not end-user input, you can do this. That is, derived or pulled from another table or other controlled source.
DECLARE #in nvarchar(some length) = N'''a'',''b'',''c'''
declare #stmt nvarchar(4000) = N'
select region, county, name
from vw_main
where state = ''MD''
and building_id in ({instr})'
set #stmt = replace(#stmt, N'{instr}', #instr)
exec sp_executesql #stmt=#stmt;
If the input is from an end-user, this is safer:
declare # table (a int, b char)
insert into #(a, b) values (1,'A'), (2, 'B')
declare #str varchar(50) = 'A,B'
select t.* from # t
join (select * from string_split(#str, ',')) s(b)
on t.b = s.b
You may like it better anyway, since there's no dynamic sql involved. However you must be running SQL Server 2016 or higher.
I have a stored procedure in a program that is not performing well. Its truncated version follows. The MyQuotes table has an IDENTITY column called QuoteId.
CREATE PROCEDURE InsertQuote
(#BinderNumber VARCHAR(50) = NULL,
#OtherValue VARCHAR(50))
AS
INSERT INTO MyQuotes (BinderNumber, OtherValue)
VALUES (#BinderNumber, #OtherValue);
DECLARE #QuoteId INT
SELECT #QuoteId = CONVERT(INT, SCOPE_IDENTITY());
IF #BinderNumber IS NULL
UPDATE MyQuotes
SET BinderNumber = 'ABC' + CONVERT(VARCHAR(10),#QuoteId)
WHERE QuoteId = #QuoteId;
SELECT #QuoteId AS QuoteId;
I feel like the section where we derive the binder number from the scope_identity() can be done much, much, cleaner. And I kind of think we should have been doing this in the C# code rather than the SQL, but since that die is cast, I wanted to fish for more learned opinions than my own on how you would change this query to populate that value.
The following update avoids needing the id:
UPDATE MyQuotes SET
BinderNumber = 'ABC' + CONVERT(VARCHAR(10), QuoteId)
WHERE BinderNumber is null;
If selecting QuoteId as a return query is required then using scope_identity() is as good a way as any.
Dale's answer is better, however this can be useful way too:
DECLARE #Output TABLE (ID INT);
INSERT INTO MyQuotes (BinderNumber, OtherValue) VALUES (#BinderNumber, #OtherValue) OUTPUT inserted.ID INTO #Output (ID);
UPDATE q SET q.BinderNumber = 'ABC' + CONVERT(VARCHAR(10),o.ID)
FROM MyQuotes q
INNER JOIN #Output o ON o.ID = q.ID
;
Also, if BinderNumber is always linked to ID, it would be better to just create computed column
AS 'ABC' + CONVERT(VARCHAR(10),ID)
I am passing string values from my code like '12th Standard/Ordinary National Diploma,Higher National Diploma' to SQL query, but I am not getting any values and nothing showing any result.
My SQL query:
declare #qua varchar(250),#final varchar(250),#Qualification varchar(250)
set #Qualification= '12th Standard/Ordinary National Diploma,Higher National Diploma'
set #qua =replace(#Qualification,',',''',''')
set #final= ''''+#qua+''''
select * from mytablename in(#final)
Result: Data is not displaying
Thank you in advance.
Instead do it using a table variable like
declare #tbl table(qual varchar(250));
insert into #tbl
select '12th Standard/Ordinary National Diploma'
union
select 'Higher National Diploma';
select * from mytablename where somecolumn in(select qual from #tbl);
Despite trying to put quote marks in there, you're still only passing a single string to the IN. The string just contains embedded quotes and SQL Server is looking for that single long string.
You also don't seem to be comparing a column for the IN.
Your best bet is to pass in multiple string variables, but if that's not possible then you'll have to write a function that parses a single string into a resultset and use that. For example:
SELECT
Column1, -- Because we never use SELECT *
Column2
FROM
MyTableName
WHERE
qualification IN (SELECT qualification FROM dbo.fn_ParseString(#qualifications))
You can insert all your search criteria in one table and then can easily do a lookup on the main table, example below:
DECLARE #MyTable TABLE (Name VARCHAR(10), Qualification VARCHAR(50))
DECLARE #Search TABLE (Qualifications VARCHAR(50))
INSERT INTO #MyTable VALUES ('User1','12th Standard'), ('User2','Some Education'),
('User3','Ordinary National Diploma'), ('User4','Some Degree'),
('User5','Higher National Diploma')
INSERT INTO #Search VALUES ('12th Standard'),('Ordinary National Diploma'),('Higher National Diploma')
SELECT MT.*
FROM #MyTable MT
INNER JOIN (SELECT Qualifications FROM #Search) S ON S.Qualifications = MT.Qualification
As previous said, you are passing a string with commas, not comma separated values. It needs to be split up into separate values.
You can do this by passing the qualification string into XML which you can use to turn it into separate rows of data.
The IN parameter will then accept the data as separate values.
DECLARE #Qualifications as varchar(150) = '12th Standard/Ordinary National Diploma,Higher National Diploma'
Declare #Xml XML;
SET #Xml = N'<root><r>' + replace(#Qualifications, char(44),'</r><r>') + '</r></root>';
select *
from MyTableName
Where MyTableName.Qualification in
(select r.value('.','varchar(max)') as item
from #Xml.nodes('//root/r') as records(r))
Alternatively you can create a table-valued function that splits according to input like in your case its ',' and then INNER JOIN with the returnColumnname and that particular column that you want to filter
SELECT COLUMNS, . . . .
FROM MyTableName mtn
INNER JOIN dbo.FNASplitToTable(#qualifications, ',') csvTable
ON csvTable.returnColumnName = mtn.somecolumn
Table Valued function might be like:
CREATE FUNCTION dbo.FNASplitToTable (#string varchar(MAX), #splitType CHAR(1))
RETURNS #result TABLE(Value VARCHAR(100))
AS
BEGIN
DECLARE #x XML
SELECT #x = CAST('<A>' + REPLACE(#string, #splitType, '</A><A>') + '</A>' AS XML)
INSERT INTO #result
SELECT LTRIM(t.value('.', 'VARCHAR(100)')) AS inVal
FROM #x.nodes('/A') AS x(t)
RETURN
END
GO
For example, I have a table
create table T (
A int,
B numeric(10,3),
C nvarchar(10),
D datetime,
E varbinary(8)
)
Update: This is just one of the example table. Any table can be used as input for generating the SQL string.
Is there an easy way to dynamically generate the following Sql for a row? (Any built-in function to make the Quotes, prefix easier?)
'declare
#A int = 1,
#B numeric(10,3) = 0.01,
#C nvarchar(10) = N''abcd'',
#D = ''10/1/2013'',
#E = 0x9123'
No, there isn't. The closest you might get, but which still will require manual changes, is by using SQL Server Management Studio. Expand the database and the table.
Right-click the table, then select Script table As, Insert To, and then selecting a new query window. This will generate output that will give you a starting point, but it's not generating variables. You'd have to either script that yourself, or edit the generated INSERT statement.
Example code:
INSERT INTO MyDB].[dbo].[Table1]
([A]
,[B]
,[C]
VALUES
(<A, int,>
,<B, float,>
,<C, nvarchar(10),>
)
GO
Not sure what specifically you are trying to achieve… There is no built in function for something like this but you can try to create one easily using query similar to this one…
select 'DECLARE #A int = ' + TableA.A +
', #B numeric(10,3) = ' + TableA.B +
', #C nvarchar(10) = N''' + TableA.C +
''', #D = ''' + TableA.D + ''''
from TableA
WHERE PrimaryKeyColumn = some_value
Just cleanup the query above and convert it into a function that returns nvarchar
If you want to dynamically generate table definitions too that’s possible too but you’ll have to use system views to create this for any given table.
Try something like this and work your way from here
select T.name, C.name, TY.name, C.column_id
from sys.tables T
inner join sys.columns C on T.object_id = C.object_id
inner join sys.types TY on TY.system_type_id = C.system_type_id
where T.name = 'TableName'
order by C.column_id asc
I just wonder if anyone knows how to automatize views creation after running DROP ... CASCADE?
Now I'm trying to drop view at first with classic DROP VIEW myview statement and if I cannot drop the view because other objects still depend on it then checking out all the objects names that postgres lists and save their creates and then I run drop with cascade. Sometimes it's like over a dozen objects. But maybe you have got some idea to handle this issue in more automated way?
Maybe anybody has got some function?
Next step... (continuation of my previous answer).
function save_views(objectname text) stores views depending on objectname (view or table) in table saved_views.
function restore_views() restores views from table saved_views.
create or replace function save_views_oid(objectid oid)
returns void language plpgsql as $$
declare
r record;
begin
for r in
select distinct c.oid, c.relname, n.nspname
from pg_depend d
join pg_rewrite w on w.oid = d.objid
join pg_class c on c.oid = w.ev_class
join pg_namespace n on n.oid = c.relnamespace
where d.refclassid = 'pg_class'::regclass
and d.classid = 'pg_rewrite'::regclass
and d.refobjid = objectid
and c.oid <> objectid
loop
insert into saved_views values (
'CREATE VIEW ' || r.nspname || '.' || r.relname ||
' AS ' || pg_get_viewdef(r.oid, 'f'));
perform save_views_oid(r.oid);
end loop;
end; $$;
create or replace function save_views(objectname text)
returns void language plpgsql as $$
begin
create table if not exists saved_views(viewbody text);
truncate saved_views;
perform save_views_oid(objectname::regclass);
end; $$;
create or replace function restore_views()
returns void language plpgsql as $$
declare
viewtext text;
begin
for viewtext in
select viewbody from saved_views
loop
execute viewtext;
end loop;
drop table saved_views;
end; $$;
Test:
select save_views('my_view'); -- may be save_views('my_schema.my_view');
select * from saved_views;
Use:
select save_views('my_view');
drop view my_view cascade;
create view my_view as ...
select restore_views();
Table pg_depend contains all necessary informations, but it is not so easy to interpret them. Here you have sketch recursive function to retrieve dependencies of pg_class object in text format. You can tune up the function to your needs (and show us results:).
create or replace function dependency
(class_id regclass, obj_id regclass, obj_subid integer, dep_type "char")
returns setof text language plpgsql as $$
declare
r record;
begin
return query
select pg_describe_object(class_id, obj_id, obj_subid)
|| ' ('|| dep_type|| ')';
for r in
select classid, objid, objsubid, deptype
from pg_depend
where class_id = refclassid
and obj_id = refobjid
and (obj_subid = refobjsubid or obj_subid = 0)
loop
return query select dependency(r.classid, r.objid, r.objsubid, r.deptype);
end loop;
end; $$;
use:
select dependency('pg_class'::regclass, 'my_view'::regclass, 0, ' ');
select dependency('pg_class'::regclass, 'my_table'::regclass, 0, ' ');