Wie kann ich meinen Kollationskonflikt besser lösen?

Question

Ich habe ein Sortierproblem mit meiner Datenbank und ich habe meine eigene Lösung entwickelt.

Lösung:

DECLARE @new_collation varchar(128), 
    @conflict_collation varchar(128), 
    @cmd_holder varchar(2000), 
    @cmd_complete varchar(2000), 
    @schema varchar(128), 
    @table_name varchar(128), 
    @constraints_name varchar(128), 
    @column_name varchar(128), 
    @definition varchar(256), 
    @data_type varchar(128), 
    @type varchar(5), 
    @length varchar(4), 
    @nullability varchar(8), 
    @db_name varchar(10) 

SET @new_collation = 'SQL_Latin1_General_CP1_CI_AS' 
SET @conflict_collation = 'French_CI_AS' 

CREATE TABLE #LIST_CONSTRAINT(
    constraints_name VARCHAR(128), 
    table_name VARCHAR(128), 
    definition VARCHAR(256), 
    type VARCHAR(10)) 

INSERT INTO #LIST_CONSTRAINT 
SELECT c.name AS constraints_name, o.name AS table_name, definition, 'CH' AS type 
FROM sys.check_constraints c 
INNER JOIN sysobjects o ON id = parent_object_id 

INSERT INTO #LIST_CONSTRAINT 
SELECT i.name AS index_name, o.name AS table_name, c.name AS field_name, 'UQ' AS type 
FROM sys.indexes i 
INNER JOIN sys.index_columns ic 
    ON i.object_id = ic.object_id and i.index_id = ic.index_id 
INNER JOIN sys.columns c 
    ON ic.object_id = c.object_id and ic.column_id = c.column_id 
INNER JOIN sys.objects o 
    ON i.object_id = o.object_id 
WHERE is_unique_constraint = 1 

SET @cmd_holder = 'ALTER TABLE $table_name DROP CONSTRAINT $constraints_name' 

DECLARE column_cursor 
CURSOR FOR SELECT constraints_name, table_name FROM #LIST_CONSTRAINT GROUP BY constraints_name, table_name 
OPEN column_cursor 
FETCH NEXT FROM column_cursor INTO @constraints_name, @table_name 

WHILE @@Fetch_Status = 0 
    BEGIN 
    SELECT @cmd_complete = @cmd_holder, 
      @cmd_complete = Replace(@cmd_complete, '$table_name', '[' + @table_name +']'), 
      @cmd_complete = Replace(@cmd_complete, '$constraints_name', @constraints_name) 

    --PRINT @cmd_complete 
    EXEC(@cmd_complete) 
    FETCH NEXT FROM column_cursor INTO @constraints_name, @table_name 
    END 
CLOSE column_cursor 
DEALLOCATE column_cursor 

SELECT @db_name = DB_NAME() 
EXEC('ALTER DATABASE ' + @db_name + ' COLLATE ' + @new_collation) 

SET @cmd_holder = 'ALTER TABLE $schema.$table_name ALTER COLUMN $column_name $data_type($length) COLLATE $new_collation $nullability' 

DECLARE column_cursor CURSOR 
    FOR SELECT table_schema, 
       table_name, 
       column_name, 
       data_type, 
       CASE WHEN character_maximum_length = -1 THEN 'max' 
        ELSE Convert(varchar(4), character_maximum_length) 
       END As length, 
       CASE WHEN is_nullable = 'YES' THEN 'NULL' 
        ELSE 'NOT NULL' 
       END As nullability 
     FROM information_schema.columns 
     INNER JOIN sysobjects ON name = table_name 
     WHERE collation_name = @conflict_collation AND xtype = 'U' 
     AND table_name NOT IN ('dtproperties', 'Exotics', 'ContractAccountsBalance', 'TechnicalParameters', 'SavingProducts', 'GeneralParameters') 
OPEN column_cursor 

FETCH NEXT FROM column_cursor INTO @schema, @table_name, @column_name, @data_type, @length, @nullability 

WHILE @@Fetch_Status = 0 
    BEGIN 
    SELECT @cmd_complete = @cmd_holder, 
      @cmd_complete = Replace(@cmd_complete, '$schema', @schema), 
      @cmd_complete = Replace(@cmd_complete, '$table_name', '[' + @table_name +']'), 
      @cmd_complete = Replace(@cmd_complete, '$column_name', '[' + @column_name +']'), 
      @cmd_complete = Replace(@cmd_complete, '$data_type', @data_type), 
      @cmd_complete = Replace(@cmd_complete, '$length', @length), 
      @cmd_complete = Replace(@cmd_complete, '$new_collation', @new_collation), 
      @cmd_complete = Replace(@cmd_complete, '$nullability', @nullability), 
      @cmd_complete = Replace(@cmd_complete, 'text(*)', 'text') 

    --PRINT @cmd_complete 
    EXEC(@cmd_complete) 

    FETCH NEXT FROM column_cursor INTO @schema, @table_name, @column_name, @data_type, @length, @nullability 
    END 
CLOSE column_cursor 
DEALLOCATE column_cursor 

DECLARE @name_constraints VARCHAR(128) 

DECLARE column_cursor 
CURSOR FOR SELECT constraints_name, table_name, definition, [type] FROM #LIST_CONSTRAINT 
OPEN column_cursor 
FETCH NEXT FROM column_cursor INTO @constraints_name, @table_name, @definition, @type 

WHILE @@Fetch_Status = 0 
    BEGIN 
    IF @type = 'CH' 
    SET @cmd_holder = 'ALTER TABLE $table_name WITH NOCHECK ADD CONSTRAINT $constraints_name CHECK NOT FOR REPLICATION $definition ALTER TABLE $table_name CHECK CONSTRAINT $constraints_name' 

    IF @type = 'UQ' 
    BEGIN 
     SET @cmd_holder = 'ALTER TABLE $table_name ADD CONSTRAINT $constraints_name UNIQUE NONCLUSTERED ($definition) WITH(STATISTICS_NORECOMPUTE = OFF, IGNORE_DUP_KEY = OFF, ALLOW_ROW_LOCKS = ON, ALLOW_PAGE_LOCKS = ON)' 

     SET @definition = ''; 

     SELECT @definition = @definition + definition + ', ' 
     FROM #LIST_CONSTRAINT 
     WHERE constraints_name = @constraints_name 

     SELECT @definition = SUBSTRING(@definition, 1, LEN(@definition) - 1)   
    END 


    SELECT @cmd_complete = @cmd_holder, 
      @cmd_complete = Replace(@cmd_complete, '$table_name', '[' + @table_name +']'), 
      @cmd_complete = Replace(@cmd_complete, '$constraints_name', @constraints_name), 
      @cmd_complete = Replace(@cmd_complete, '$definition', @definition) 

    --PRINT @cmd_complete 
    IF (@name_constraints <> @constraints_name) 
     EXEC(@cmd_complete) 

    SET @name_constraints = @constraints_name 

    FETCH NEXT FROM column_cursor INTO @constraints_name, @table_name, @definition, @type 
    END 
CLOSE column_cursor 
DEALLOCATE column_cursor 

DROP TABLE #LIST_CONSTRAINT 

Hat jemand eine andere Lösung?

Kann mir jemand Ratschläge geben, um meinen Code zu optimieren?

Answer 1

Ich habe gesehen, wo die Tempdb eine andere Standard-Kollatierung als Ihre Datenbank hat, und ich musste "COLLATE DATABASE_DEFAULT" zu meinen Vergleichen hinzufügen, z.

Create table #tmp 
(
     mailbox varchar(50) not null 
) 
. . . 
Select t.mailbox, count(*) 
from #tmp t inner join processed_email e 
on t.mailbox = e.mailbox 

wird

Select t.mailbox, count(*) 
from #tmp t inner join processed_email e 
on t.mailbox COLLATE DATABASE_DEFAULT = e.mailbox COLLATE DATABASE_DEFAULT 

Answer 2

Binary Worrier-Lösung für die Ausführung von Code mit verschiedenen Sortierungen nützlich ist. Beispielsweise verwenden temporäre Tabellen und temporäre Variablen die Tempdb-Sortierung.

Um die Sortierung zu ändern, hat MS eine KB "How to transfer a database from one collation to another collation in SQL Server" zusammengestellt. Es erwähnt jedoch immer noch die Verwendung von DTS, was möglicherweise nicht möglich ist. Ich kann keinen aktualisierten Knowledge Base-Artikel finden.