Duplicated records

Question

im lost now... i need to create a program that recive a Table 1... run it searching for duplicated records... copy them (duplicated records) to table 2 and the others(that dont have any duplicated) copy to table 3... im exausted to think about that... i need help...

what kind of tables i have:
T1= messy full of duplicated records
T2= T1 without the duplicated records
remember pls: what i need is a table with the list of whichs records are duplicated
(before running)
T1                             T2                    TABLE3
A                                A          
B                                B
C                                C
A                               
B                               
C                                    

(after running)
TABLE 1                             TABLE2                    TABLE3
A                                        A                                 A
B                                        B                                 B
C                                        C                                 C
A                                        D
B                                        E                             (notice that E is not duplicated in t1 so it cant appear on t3)
C
D
D
E
B                                       

Answer 1

Leonardo,

I think nobody would understand what you are asking. Try elaborating your question. Also define what is a duplicate for you.

 

 

Answer 2

ok.. here it is... first a duplicated record is 2 records that has the same key.. the exctaly same value...

and as an answer i was hoping for something like... try this code here... i gave alot of time to it and all i came up was this:


CLOSE DATABASES
CREATE TABLE c:\barrados (LIVRO N(6), TIPO C(6), FOLHA n(6), TERMO n(6),NOME c(68), MAE c(68),cartorio c(15), mode n(1))
table1 = GETFILE('DBF', 'Tabela 1','SET', 1, 'Analisar')
USE (table1)
FOR i =1 TO RECCOUNT()
  GO RECORD i
  var1=livro
  var2=tipo
  var3=folha
  var4=termo
  var5=nome
  var6=mae
  var7=cartorio
     FOR c1 = i+1 TO RECCOUNT()
      GO RECORD c1
      IF var1==livro AND var3==folha AND var4==termo AND var5==nome AND var6==mae   
      INSERT INTO barrados (livro,tipo,folha,termo,nome,mae,cartorio,mode) ;
        VALUES (var1,var2,var3,var4,var5,var6,var7,1)
    ENDIF   
    ENDFOR   
ENDFOR        
CLOSE DATABASES


it works... slow.. but works... i set table1... and in "barrados" i have all registers that are repeated in table1...

important to notice: it works kind like this... it "take" record i and compare with all bellow him... i think i cant use scan or select cause they will compare the record X with all the table... and eventually u will compare X with itself and will acuse the X is duplicated when it really isnt... get it? its complicated... if u need even futher explanation just tell me...

Answer 3

*-- With this SELECT you'll get unique records in a cursor

SELECT * FROM T1 GROUP BY Field1 ORDER BY Field1 INTO CURSOR T2

*-- With this SELECT you'll get records not duplicated

SELECT * FROM T1 GROUP BY Field1 HAVING COUNT(*) = 1 ORDER BY Field1 INTO CURSOR T3

 

I only use "Field1" as group criteria because you do not tell nothing about fields; you must put in GROUP BY clause all the fields that let you find a duplicated record.

 

Best regards

 

dD

Answer 4

Code Snippet

SELECT livro,tipo,folha,termo,nome,mae,cartorio,mode ;

 FROM (m.table1) ;

 HAVING COUNT(*) > 1 ;

 GROUP BY livro,tipo,folha,termo,nome,mae,cartorio,mode ;

 INTO TABLE ('c:\barrados')

 

You could also use scan...endscan (and it would be fast) but Select-SQL is easier to code and portable to other backends.

 

PS: It is not complicated at all but you need to explain your problem well.

Answer 5

tks alot!!! thats perfect!!!