Conflicto de intercalación

Question

siempre recibí el mensaje "No se puede resolver el conflicto de intercalación entre" SQL_Latin1_General_CP1_CI_AS "y" Latin1_General_CI_AI "en la operación igual". cuando ejecuto el script que vino del servidor MSSQL 2005. por cierto yo solía MSSQL 2008.

aquí es el guión

USE [database1] 
GO 
SET ANSI_NULLS ON 
GO 
SET QUOTED_IDENTIFIER ON 
GO 
--exec rsp_SOA '', '', '', '', '', '', '40050000', '40050000', '05/01/2010', '05/31/2010', '000-00','Dont Show Note' 

ALTER procedure [dbo].[rsp_SOA] 
    @cCompName VARCHAR(100) , 
    @cAddress1 VARCHAR(200) , 
    @cAddress2 VARCHAR(200), 
    @cReportTitle VARCHAR(200), 
    @cCriteria1 VARCHAR(200), 
    @cCriteria VARCHAR(200), 

    @cFrom VARCHAR(25), 
    @cTo VARCHAR(25), 
    @dDateFrom VARCHAR(10), 
    @dDateTo VARCHAR(10), 
    @cCompID VARCHAR(10), 
    @cFilter VARCHAR(30) 
as 

declare @cSQL VARCHAR(200), 
     @cSQL1 VARCHAR(200), 
     @cmd VARCHAR(8000), 
     @cmd1 VARCHAR(8000), 
     @cTempTable varchar (50), 
     @cTempTable1 varchar (50), 
     @nInterval VARCHAR(3), 
     @nCurrent INTEGER, 
     @nInterval1 varchar(3), 
     @nInterval2 varchar(3), 
     @nInterval3 varchar(3), 
     @nInterval4 varchar(3), 
     @nInterval5 varchar(3), 
     @dd INTEGER, 
     @cValue1 VARCHAR(100), 
     @cValue2 VARCHAR(100), 
     @cValue3 VARCHAR(100), 
     @cValue4 VARCHAR(100), 
     @cValue5 VARCHAR(100) 

set @nInterval = 30 
set @nInterval1 = cast(@nInterval * 1 as varchar(3)) 
set @nInterval2 = cast(@nInterval * 2 as varchar(3)) 
set @nInterval3 = cast(@nInterval * 3 as varchar(3)) 
set @nInterval4 = cast(@nInterval * 4 as varchar(3)) 
set @nInterval5 = cast(@nInterval * 5 as varchar(3)) 

SET @cValue1 = CONVERT(VARCHAR(10),1,110) + ' - ' + CONVERT(VARCHAR(10),@nInterval1,110) + ' Days' 
SET @cValue2 = CONVERT(VARCHAR(10),@nInterval1 + 1,110) + ' - ' + CONVERT(VARCHAR(10),@nInterval2,110) + ' Days' 
SET @cValue3 = CONVERT(VARCHAR(10),@nInterval2 + 1,110) + ' - ' + CONVERT(VARCHAR(10),@nInterval3,110) + ' Days' 
SET @cValue4 = CONVERT(VARCHAR(10),@nInterval3 + 1,110) + ' - ' + CONVERT(VARCHAR(10),@nInterval4,110) + ' Days' 
SET @cValue5 = 'Above ' + CONVERT(VARCHAR(10),@nInterval4,110) + ' Days' 

create table ##Interval 
(
    cCompID varchar(20), 
    nOrder int, 
    cInterval varchar(20), 
    cCode varchar(20) 
) 


--------------------------- 
--get all the clients 
--------------------------- 
SELECT * INTO ##temp1 FROM 
(
    select cCode from client_customer where cCode = @cFrom union all 
    select cGroupCode from client_customer where cCode = @cFrom union all 
    select cBillingCompany from client_customer where cCode = @cFrom union all 
    select cArea from client_customer where cCode = @cFrom 
)a 

--------------------------- 
--determining the balance of the invoices 
--------------------------- 
SELECT * INTO ##temp2 FROM 
(
    (
     select a.cCompanyID, a.cInvNo, sum(a.nGross) as nSales, 0 as nPaid, 0 as nDebit, 0 as nCredit, 0 as nReturns 
     from sales a 
     where a.lCancelled = 0 and a.cPayType<>'Cash' 
      and a.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, a.cInvNo 
    ) 
    union all 
    (
     select a.cCompanyID, a.cInvNo, 0 as nSales, sum(a.nApplied) as nPaid, 0 as nDebit, 0 as nCredit, 0 as nReturns 
     from pr_t a 
     left outer join pr b on a.cTranNo = b.cTranNo and a.cCompanyID = b.cCompanyID 
     left outer join sales c on a.cInvNo = c.cInvNo and a.cCompanyID = c.cCompanyID 
     where b.lCancelled = 0 
      and c.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, a.cInvNo 
    ) 
    union all 
    (
     select a.cCompanyID, a.cInvNo, 0 as nSales, 0 as nPaid, sum(a.nDebit) as nDebit, 0 as nCredit, 0 as nReturns 
     from ar_t a 
     left outer join ar b on a.cTranNo = b.cTranNo and a.cCompanyID = b.cCompanyID 
     where b.cType = 'Debit' and b.lCancelled = 0 and b.lApproved = 1 
      and b.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, a.cInvNo 
    ) 
    union all 
    (
     select a.cCompanyID, a.cInvNo, 0 as nSales, 0 as nPaid, 0 as nDebit, sum(a.nCredit) as nCredit, 0 as nReturns 
     from ar_t a 
     left outer join ar b on a.cTranNo = b.cTranNo and a.cCompanyID = b.cCompanyID 
     where b.cType = 'Credit' and b.lSalesReturn = 0 and b.lCancelled = 0 and b.lApproved = 1 
      and b.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, a.cInvNo 
    ) 
    union all 
    (
     select a.cCompanyID, a.cInvNo, 
     0 as nSales, 0 as nPaid, 0 as nDebit, 0 as nCredit, sum(a.nCredit) as nReturns 
     from ar_t a 
     left outer join ar b on a.cTranNo = b.cTranNo and a.cCompanyID = b.cCompanyID 
     where b.cType = 'Credit' and b.lSalesReturn = 1 and b.lCancelled = 0 and b.lApproved = 1 
      and b.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, a.cInvNo 
    ) 
)a 


----------------------- 
--main script for creating the temp ##AR 
----------------------- 
select a.*, b.cRefNo, b.dRefDate, b.nBeg, b.nEnd, b.nConsumed, b.nConsumedKG, b.nPrice, b.nAmount, b.nVat, b.nWHT, b.nNet, 
      case when a.nDue <=0 then 'Current' 
       when a.nDue > 0 and a.nDue <= @nInterval1 then @cValue1 
       when a.nDue > @nInterval1 and a.nDue <= @nInterval2 then @cValue2 
       when a.nDue > @nInterval2 and a.nDue <= @nInterval3 then @cValue3 
       when a.nDue > @nInterval3 and a.nDue <= @nInterval4 then @cValue4 
       when a.nDue > @nInterval4 then @cValue5 
      end as cInterval,b.cTenantName into ##AR 
    from 
    (
     select a.cCompanyID, b.cCode, a.cInvNo, (isnull(sum(a.nSales), 0) - isnull(sum(a.nPaid), 0) + isnull(sum(a.nDebit), 0) - isnull(sum(a.nCredit), 0) - isnull(sum(a.nReturns), 0)) as nBalance, 
       b.dDate, b.cSMan, b.nGross, b.cTerm, c.cValue, (cast((cast(convert(varchar(20),getdate(),101) as datetime) - b.dDate) as integer) - cast(c.cValue as integer)) as nDue, dateadd(d,cast(c.cValue as integer),b.dDate) as dDueDate 
     from 
     (
      select * FROM ##temp2 --determining the balance of the invoices 
     ) a 
     left outer join sales b on a.cInvNo = b.cInvNo and a.cCompanyID = b.cCompanyID 
     left outer join parameter_user c on b.cTerm = c.cParamName and c.cType = 'TERMS' 
     where b.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
     group by a.cCompanyID, b.cCode, a.cInvNo, b.dDate, b.cSMan, b.nGross, b.cTerm, c.cValue 
     having (isnull(sum(a.nSales), 0) - isnull(sum(a.nPaid), 0) + isnull(sum(a.nDebit), 0) - isnull(sum(a.nCredit), 0) - isnull(sum(a.nReturns), 0)) > 0 
    )a 
    left outer join 
    (
     select a.cCompanyID, a.cInvNo, b.cRefNo, d.dDate as dRefDate, 
      case when a.cType = 'Meter' then b.nMeterIn 
       when a.cType = 'Weight' then b.nWeightOut 
      else 0 end as nBeg, 
      case when a.cType = 'Meter' then b.nMeterOut 
       when a.cType = 'Weight' then b.nWeightIn 
      else 0 end as nEnd, b.nConsumed, 
      case when a.cType = 'Meter' then b.nConsumedKG 
       when a.cType = 'Weight' then (b.nWeightOut - b.nWeightIn) 
      else 0 end as nConsumedKG, b.nPrice, (b.nAmount/1.12) as nAmount, 
       (b.nAmount - (b.nAmount/1.12)) as nVat, 
       (CASE WHEN e.lEWT=1 THEN ((b.nAmount/1.12) * e.nEWT/100) ELSE 0 END) as nWHT, b.nAmount as nNet,f.cTenantName 


     from sales a left outer join sales_t b on a.cInvNo = b.cInvNo and a.cCompanyID = b.cCompanyID 
      left outer join item c on b.cItemNo = c.cItemNo and a.cCompanyID = c.cCompanyID 
      left outer join dr d on d.cDRNo = b.cRefNo 
      left outer JOIN CLIENT_CUSTOMER e ON a.cCode = e.cCode 
      left outer JOIN METER_READING_T f ON b.cMRNo = f.cTransNo AND b.nMeterIn = f.nMeterIn 
     where c.cType <> 'CYLINDER' and a.cType <> 'Invoice' and a.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 

     union all 

     select a.cCompanyID, a.cInvNo, d.cDRNo as cRefNo, d.dDate as dRefDate, 0 as nBeg, 
      0 as nEnd, b.nConsumed, 0 as nConsumedKG, b.nPrice, b.nAmount, 0 as nVat, 0 as nWHT, 0 as nNet,'' as cTenantName 
     from sales a left outer join sales_t b on a.cInvNo = b.cInvNo and a.cCompanyID = b.cCompanyID 
      left outer join item c on b.cItemNo = c.cItemNo and a.cCompanyID = c.cCompanyID 
      left outer join dr d on a.cInvNo = d.cInvNo   
     where c.cType <> 'CYLINDER' and a.cType = 'Invoice' and d.cDRNo is null and a.cCode in (select * FROM ##temp1) --get all the clients 
      and a.cCompanyID = @cCompID 
    )b on a.cInvNo = b.cInvNo and a.cCompanyID = b.cCompanyID 


----------------------- 
--main script for creating the temp ##Interval 
----------------------- 
insert into ##Interval 
select distinct @cCompID, 1, 'Above 120 Days', cCode from ##AR 
insert into ##Interval 
select distinct @cCompID, 2, '91 - 120 Days', cCode from ##AR 
insert into ##Interval 
select distinct @cCompID, 3, '61 - 90 Days', cCode from ##AR 
insert into ##Interval 
select distinct @cCompID, 4, '31 - 60 Days', cCode from ##AR 
insert into ##Interval 
select distinct @cCompID, 5, '1 - 30 Days', cCode from ##AR 
insert into ##Interval 
select distinct @cCompID, 6, 'Current', cCode from ##AR 




-------------------------------------- 
--displaying the result 
-------------------------------------- 

select a.nOrder, a.cInterval, a.cCode, c.cName, (c.cFirstName + ' ' + case when isnull(c.cMiddleInitial,'')='' then '' else c.cMiddleInitial + ' ' end + c.cLastName) as cCustomerName, (case when isnull(c.cBusinessName,'')='' then c.cName else c.cBusinessName end) as cBusinessName, 
     c.cAddress, c.cLastName, c.cJobTitle, d.cCompanyName, d.cAddress1, d.cAddress2, d.cPhone, d.cfax, d.cEmail, c.cTerm, b.cInvNo, b.nBalance, b.dDate, b.nGross, b.cRefNo, b.dRefDate, b.nBeg, b.nEnd, b.nConsumed, b.nCOnsumedKG, 
     b.nPrice, b.nAmount, b.nVat, b.nWHT, b.nNet, isnull(e.nTotalPDC,0) as nTotalPDC, f.nTotalAR, f.nTotalPastDue, c.nLimit, @dDateFrom as dStartDate, @dDateTo as dEndDate,--g.cTenantName, 
     b.cTenantName,(CASE WHEN @cFilter = 'Show Note' THEN 1 ELSE 0 END) AS lNote 
from ##Interval a 
left outer join ##AR b on a.cInterval = b.cInterval and a.cCode = b.cCode 
left outer join client_customer c on a.cCode = c.cCode 
left outer join company d on a.cCompID = d.cCompanyID 
left outer join 
(
    select a.cCode, sum(a.nAmount) as nTotalPDC 
    from checks a 
    where a.lDeposited=0 and a.cTransType = 'COL' and a.cCode = @cFrom and a.cCompanyID = @cCompID 
    group by a.cCode 
)e on a.cCode=e.cCode 
left outer join 
(
    --select a.cCode, sum(a.nBalance) as nTotalAR, sum(case when a.cInterval <> 'Current' then isnull(a.nBalance,0) else 0 end) as nTotalPastDue 
    --from ##AR a 
    --group by a.cCode 

    SELECT a.cCode,SUM(nTotalAR) AS nTotalAR,SUM(nTotalPastDue) AS nTotalPastDue 
    FROM 
     (select distinct a.cCode, (a.nBalance) as nTotalAR, 
      (case when a.cInterval <> 'Current' then isnull(a.nBalance,0) else 0 end) as nTotalPastDue 
     from ##AR a) a 
    GROUP BY a.cCode 

)f on a.cCode=f.cCode 
order by a.nOrder, a.cCode 


drop table ##temp1 
drop table ##temp2 
drop table ##Interval 
drop table ##AR 


/* 
    select * from ##temp1 
    select * from ##temp2 
    select * from ##Interval 
    select * from ##AR 
*/ 

el servidor SQL siempre señalando el error de esta línea

-------------------------------------- 
--displaying the result 
-------------------------------------- 

select a.nOrder, a.cInterval, a.cCode, c.cName, (c.cFirstName + ' ' + case when isnull(c.cMiddleInitial,'')='' then '' else c.cMiddleInitial + ' ' end + c.cLastName) as cCustomerName, (case when isnull(c.cBusinessName,'')='' then c.cName else c.cBusinessName end) as cBusinessName, 
     c.cAddress, c.cLastName, c.cJobTitle, d.cCompanyName, d.cAddress1, d.cAddress2, d.cPhone, d.cfax, d.cEmail, c.cTerm, b.cInvNo, b.nBalance, b.dDate, b.nGross, b.cRefNo, b.dRefDate, b.nBeg, b.nEnd, b.nConsumed, b.nCOnsumedKG, 
     b.nPrice, b.nAmount, b.nVat, b.nWHT, b.nNet, isnull(e.nTotalPDC,0) as nTotalPDC, f.nTotalAR, f.nTotalPastDue, c.nLimit, @dDateFrom as dStartDate, @dDateTo as dEndDate,--g.cTenantName, 
     b.cTenantName,(CASE WHEN @cFilter = 'Show Note' THEN 1 ELSE 0 END) AS lNote 

Answer 1

Es probablemente sea:

1. Hay son algunas columnas o tablas con caracteres específicos creadas con una COLLACIÓN diferente, entonces estás haciendo una compa rison en columnas de diferente colación

2. Está ejecutando una secuencia de comandos en una base de datos que es una recopilación diferente de su servidor, y la secuencia de comandos utiliza TEMP TABLE. Las tablas temporales se crean en la base de datos tempdb, por lo tanto, obtienen la COLLATION de tempdb (es decir, la intercalación del servidor). La mayoría de los scripts no esperan esto, por lo que las comparaciones con las columnas de tablas temporales no especifican una intercalación.

Puedes solucionar este problema ya sea cambiando las tablas/columnas para el partido colaciones, o cambiar su SQL de modo que especifique el tipo de colación a utilizar en cada comparación de cadenas.

p. Ej. cambiar esta situación:

select * 
from someTable 
inner join #t on someTable.col1 = #t.col1 
where #t.col2 = 'blah' collate database default 

a esto:

select * 
from someTable 
inner join #t on someTable.col1 = #t.col1 collate database default 
where #t.col2 = 'blah' collate database default 

ACTUALIZACIÓN:

Basado en el guión que has enviado, supongo que es mi segunda sugerencia anterior.

SQL Server dice que el problema está en su última consulta masiva. Sugiero uno o más de los siguientes:

• para identificar la consulta en particular eso es un problema que sugeriría usted explota las subconsultas y tienen la secuencia de comandos ejecutar por separado. Solo para la depuración.

• Intente llamar al sp_help '##temp1' y para las otras tablas una vez creadas, y compare las intercalaciones de columnas con las de sus tablas principales.

• Añadir collate database default al final de cada una de las comparaciones (cuando sea necesario si se utiliza la sugerencia sp_help)

• crear las tablas temporales de forma explícita en lugar de a través de SELECT INTO. De esta forma puede especificar la intercalación de las columnas. O use variables de tabla en su lugar.

Answer 2

Cuando se comparan las columnas/agregados con diferentes colaciones debe Coherencia al menos uno de ellos para el cotejo que desee, utilizando la expresión COLLATE.Por ejemplo, supongamos que tiene una columna c1 con intercalación Latin1_General_CI_AI en la tabla Table1 y una columna c2 con intercalación SQL_Latin1_General_CP1_CI_AS en la tabla Table2 y que desea la unión de los dos:

SELECT c1 COLLATE SQL_Latin1_General_CP1_CI_AS 
FROM Table1 
UNION ALL 
SELECT c2 
FROM Table2; 

Answer 3

Parece que esta pregunta es una vieja pregunta, pero En cualquier caso, daré respuesta, porque recientemente tuve un problema similar, ¡y pensé que podría salvar a alguien más de la pena!

Compruebe su tabla para crear scripts para la base de datos porque pueden estar configurando las intercalaciones en las columnas en el momento de la creación. Lo cual no necesariamente causa un error hasta más adelante en el ciclo de desarrollo. es decir, como la instancia se identifica aquí en que usted está tratando de preformas una comparación de igualdad

select a.nOrder, a.cInterval, a.cCode, c.cName, (c.cFirstName + ' ' + case when isnull(c.cMiddleInitial,'')='' then '' else c.cMiddleInitial + ' ' end + c.cLastName) as cCustomerName, (case when **isnull(c.cBusinessName,'')=''** then c.cName else c.cBusinessName end) as cBusinessName,   c.cAddress, c.cLastName, c.cJobTitle, d.cCompanyName, d.cAddress1, d.cAddress2, d.cPhone, d.cfax, d.cEmail, c.cTerm, b.cInvNo, b.nBalance, b.dDate, b.nGross, b.cRefNo, b.dRefDate, b.nBeg, b.nEnd, b.nConsumed, b.nCOnsumedKG,   b.nPrice, b.nAmount, b.nVat, b.nWHT, b.nNet, isnull(e.nTotalPDC,0) as nTotalPDC, f.nTotalAR, f.nTotalPastDue, c.nLimit, @dDateFrom as dStartDate, @dDateTo as dEndDate,--g.cTenantName,   b.cTenantName,(CASE WHEN @cFilter = 'Show Note' THEN 1 ELSE 0 END) AS lNot 

He resaltado en negrita. Mi problema parece provenir del hecho de que algunas de nuestras tablas se construyeron en SQL Management Studio y algunas tablas se construyeron en Visual Studio, que tenía diferentes configuraciones de intercalación habilitadas. Tardó bastante tiempo para resolver esto también. Fue solo después de revisar los scripts de la base de datos que lo recogí. Ahora hemos pasado al uso de la edición del desarrollador de bases de datos en Visual Studio Premium, que ahora ha simplificado nuestra gestión de configuración de bases de datos y pruebas de unidades