2012 server

[dtussman]

Which is better to convert a large legacy application to SQL: SQLRDD or ADORDD?

[Antonio Linares]

I would say that AdoRdd is the best way to go as there is full source code available and Antonio Ferreira is providing a great support for it

[dtussman]

Thanks Antonio,

Back in the 90's after much deliberation I chose Fivewin to migrate from DOS to Windows and never regretted it, so I will probably go with your advice again!

Regards,

David

[Antonio Linares]

David,

Thanks for trusting in my advice and for your support along all these years

Antonio Ferreira has confirmed that he has ported a large app to AdoRdd without any source code modifications!

Thats great

[AHF]

David,

I ported a large app with no code change.

In this case was easy because the app worked already with ADS so the indexes expressions with vars were all evaluated before sent to adordd.

This should be the only exception in the code of actual apps because vars are out of scope when sent to SQL so need to be evaluated before and sent the result as a string.

Lucas is also testing in his application and so far its seems its running quite good.

[dtussman]

No code changes? I was reading that ADDRDD Error thread and it seems like you have to do quite a bit of stuff with regard to files and indexes. Plus you can't use recno() anymore, which I use a lot. But if that's all there is I guess its not too bad.

[lucasdebeltran]

David,

You have only to indicate at the begining of your application a few settings and that´s all.

Yes, you can use Recno().

This is a sample:

Code: Select all  Expand view

function Main()

  RddRegister("ADORDD",1)
  RddSetDefault("ADORDD")

   SET DATE FORMAT TO "dd-mm-yyyy" // Formato Fecha: DD-MM-AAAA
   SET CENTURY ON
   SET EPOCH TO YEAR(DATE())-98


  SET ADO TABLES INDEX LIST TO {{ "TABLE1", {"TEST1", "FIRST"} ,;
                                            {"TEST2", "LAST"} ,;
                                            {"TEST3", "AGE"} },;
                                    {"LUCAS", {"LUCAS1", "NOMBRE"} ,;
                                              {"LUCAS3", "NOMBRE", "WHERE NOMBRE = 'LUCAS'"}     }     }


  SET ADODBF TABLES INDEX LIST TO { { "TABLE1", {"TEST1", "FIRST"} ,;
                                               {"TEST2", "LAST"} ,;
                                               {"TEST3", "AGE"} },;
                                    {"LUCAS", {"LUCAS1", "NOMBRE"} ,;
                                              {"LUCAS3", "NOMBRE", "NOMBRE = 'LUCAS'" }   } }




  SET ADO DEFAULT RECNO FIELD TO "HBRECNO"

  SET AUTOPEN OFF //ON //might be OFF if you wish

  //SET AUTORDER TO 1 // first index opened can be other

  SET ADO FORCE LOCK OFF


  SET ADO DEFAULT DATABASE TO "LUCASTABLE.MDB"  ENGINE TO "ACCESS" USER TO "" PASSWORD TO ""


 

That´s all.

The rest of your code is the SAME, including filters.

[dtussman]

Thanks Lucas,

That was helpful

David

[MarcoBoschi]

Dear driessen,
can you post again the link in wich is explained exactly the operations to do in server and clients?
I have a lot of problems

[hua]

This is a post on Harbour Users group from someone experiencing slowness on Windows 2012 too - https://groups.google.com/forum/#!topic ... -1OKaCvTxk

HTH