FiveTech Software tech support forums

If you don’t use relations what do you use for ex. to display fields of different "related" workareas?

seek whatever = select * from .... where (seek expression)

Code: Select all  Expand view

LOCAL a :=  { { "CUSTOMER", {"CUSTOMER1",  "NAME"} ,;
                                           {"CUSTOMER2",  "NAME", "WHERE CITY = 'Barcelona' "} ,;
                                           {"LUCAS",          "STREET", "WHERE NOTES = 'varis' "} ,;
                         }  }
 

Code: Select all  Expand view

SET AUTOPEN ON

USE access.mdb VIA "ADORDD" TABLE "CUSTOMER"  ACCESS  NEW  ALIAS "CUSTOMER"
 

Code: Select all  Expand view

INDEX ON FIELD->FSERIE + StrZero (   FIELD->FNUMERO, 20   ) TAG CAB

INDEX ON FIELD->FNUMERO               TAG ALB   FOR FIELD->FP != "S"

INDEX ON FIELD->TIPDOC    TAG AUXILIAR3  FOR  !EMPTY(   FIELD->TIPDOC   )

INDEX ON DTOS( FIELD->FECHA )         TAG CARTERA3    FOR FIELD->COBRO = 0

lucasdebeltran wrote:Thank you Antonio, there is no rush.

I would like to ask you how I define such indexes with functions and conditions. For instance:

Code: Select all  Expand view

INDEX ON FIELD->FSERIE + StrZero (   FIELD->FNUMERO, 20   ) TAG CAB

INDEX ON FIELD->FNUMERO               TAG ALB   FOR FIELD->FP != "S"

INDEX ON FIELD->TIPDOC    TAG AUXILIAR3  FOR  !EMPTY(   FIELD->TIPDOC   )

INDEX ON DTOS( FIELD->FECHA )         TAG CARTERA3    FOR FIELD->COBRO = 0

Thank you.

Code: Select all  Expand view

{....
{"CAB",{"FSERIE,NUMERO"}} //I dont know strzero
{"ALB",{"FNUMERO"},{"WHERE FP != 'S' "}}  //!= depends on your DB it might only accepts <>
{"AUXILIAR3",{"TIPODOC"},{"WHERE TIPODOC != NULL"}}
{"CARTEIRA3",{"FECHA"},{"WHERE COBRO = 0 "}}  }
 

Code: Select all  Expand view

STATIC FUNCTION ADO_CREATE( nWA, aOpenInfo )


   TRY
      IF Lower( Right( cDataBase, 4 ) ) == ".fdb"
         oConnection:Execute( "CREATE TABLE " + cTableName + " (" + StrTran( StrTran( aWAData[ WA_SQLSTRUCT ], "[", '"' ), "]", '"' ) + ")" )
      ELSE

         //oConnection:Execute( "CREATE TABLE [" + cTableName + "] (" + aWAData[ WA_SQLSTRUCT ] + ")" )

         N := FW_AdoCreateTableSQL( cTableName, aWAData[ WA_SQLSTRUCT ], oConnection, .T. )
         oConnection:Execute( N )

      ENDIF
   CATCH
      oError := ErrorNew()
      oError:GenCode := EG_CREATE
      oError:SubCode := 1004
      oError:Description := hb_langErrMsg( EG_CREATE ) + " (" + ;
         hb_langErrMsg( EG_UNSUPPORTED ) + ")"
      oError:FileName := aOpenInfo[ UR_OI_NAME ]
      oError:CanDefault := .T.

      FOR n := 0 TO oConnection:Errors:Count - 1
         oError:Description += oConnection:Errors( n ):Description
      NEXT

      UR_SUPER_ERROR( nWA, oError )
   END

   oConnection:Close()

   RETURN HB_SUCCESS



STATIC FUNCTION ADO_CREATEFIELDS( nWA, aStruct )

   LOCAL aWAData := USRRDD_AREADATA( nWA )
   LOCAL n

aWAData[ WA_SQLSTRUCT ] := aStruct


/*
   aWAData[ WA_SQLSTRUCT ] := ""

   FOR n := 1 TO Len( aStruct )
      IF n > 1
         aWAData[ WA_SQLSTRUCT ] += ", "
      ENDIF
      aWAData[ WA_SQLSTRUCT ] += "[" + aStruct[ n ][ DBS_NAME ] + "]"
      DO CASE
      CASE aStruct[ n ][ DBS_TYPE ] $ "C,Character"
         aWAData[ WA_SQLSTRUCT ] += " CHAR(" + str( aStruct[ n ][ DBS_LEN ] ) + ") NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "V"
         aWAData[ WA_SQLSTRUCT ] += " VARCHAR(" + str( aStruct[ n ][ DBS_LEN ] ) + ") NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "B"
         aWAData[ WA_SQLSTRUCT ] += " DOUBLE NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "Y"
         aWAData[ WA_SQLSTRUCT ] += " SMALLINT NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "I"
         aWAData[ WA_SQLSTRUCT ] += " MEDIUMINT NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "D"
         aWAData[ WA_SQLSTRUCT ] += " DATE NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "T"
         aWAData[ WA_SQLSTRUCT ] += " DATETIME NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "@"
         aWAData[ WA_SQLSTRUCT ] += " TIMESTAMP NULL DEFAULT CURRENT_TIMESTAMP"

      CASE aStruct[ n ][ DBS_TYPE ] == "M"
         aWAData[ WA_SQLSTRUCT ] += " TEXT NULL"

      CASE aStruct[ n ][ DBS_TYPE ] == "N"
         aWAData[ WA_SQLSTRUCT ] += " NUMERIC(" + str( aStruct[ n ][ DBS_LEN ] ) + ")"

      CASE aStruct[ n ][ DBS_TYPE ] == "L"
         aWAData[ WA_SQLSTRUCT ] += " LOGICAL"
      ENDCASE
   NEXT


*/


   RETURN HB_SUCCESS


 

Code: Select all  Expand view

function Main()

   local aArray := {}

/*
   DbCreate( "test2.mdb;table1", { { "FIRST",   "C", 30, 0 },;
                                   { "LAST",    "C", 30, 0 },;
                                   { "AGE",     "N",  8, 0 } }, "ADORDD" )

*/
   USE test2.mdb VIA "ADORDD" TABLE "table1" NEW

browse()


   APPEND BLANK
   test2->First   := "HOMER si no Homer"
   test2->Last    := "Simpson"
   test2->Age     := 45

   APPEND BLANK
   test2->First   := "aaa Lara"
   test2->Last    := "Croft si no"
   test2->Age     := 32


   GO TOP

   xBrowse()
   DbCloseAll()




return nil

 

Have you looked at arrayrdd source, SQLRDD rdd source or Mediator SQL RDD from www.otc.pl?.

It may help.

Also, if you use find for seek?. I think in ADO indexes are fine for showing the data ordered by a field, but they are not so important for searching and speed as they are in DBF.

DbSeek()
Searches a value in the controlling index.
Syntax
DbSeek( <xValue>, [<lSoftSeek>], [<lFindLast>] ) --> lFound

Arguments
<xValue>
The value to search for. Its data type must match the data type of the index expression of the controlling index.
<lSoftSeek>
This optional value defaults to .F. (false) causing the DbSeek() function to position the record pointer at Eof() if <xValue> is not found in the index. When .T. (true) is passed for <lSoftSeek> and <xValue> is not found in the index, the record pointer is positioned on the record with the next higher index value.
<lFindLast>
<lFindLast> is only relevant when the database contains multiple records having identical index values. It defaults to .F. (false) causing the DbSeek() function to position the record pointer on the first record found. .T. (true) instructs DbSeek() to position the record pointer on the last of multiple records having the same index value. Return
DbSeek() returns .T. (true) if <xValue> is found, otherwise .F. (false).
Description
The DbSeek() function is used to perform fast searches in databases. To accomplish this, the database must be indexed, since DbSeek() searches the value <xValue> in the controlling index, rather than in the database. It operates in the current work area, unless it is used in an aliased expression.
When DbSeek() finds <xValue> in the controlling index, it returns .T. (true) and positions the record pointer to the corresponding record. The parameter <lFindLast> optionally specifies which record to find if there are multiple records having the same index value. By default, the first record is found. If <lFindLast> is .T. (true), DbSeek() positions the record pointer on the last of the records having identical index values.
After a successful search, the function Found() returns .T. (true) until the record pointer is moved again. In addition, both functions, BoF() and EoF() return .F. (false).
If the searched value is not found, DbSeek() positions the record pointer on the "ghost record" (Lastrec()+