Forum Discussion

jhonathanramos's avatar
jhonathanramos
Occasional Contributor
7 years ago

Trying to get SQL Exception using "ADO.CreateADOCommand"

Hello there;

 

This is my first topic of doubt. I have been researching for some time without a solution and so I decided to ask, because even with some research inside and outside the Smartbear community I could not find a solution. :smileysad:

 

I am trying to execute an SQL statement in SQLServer through the existing ADO.Command in TestComplete, expecting the return of an exception.

 

The database only returns if there is a syntax error. However, if the syntax is correct, but the operation can not be performed, the database is not returning the exception. Just with the return of the exception I can handle the situation in my code.

Test execution needs to stop if the database fails to execute the statement I submitted to it.

 

Exemplifying:

I tried to use "try-except" in my delphiscript, but the database is not returning the exception forcing it to drop directly on the except line. (Neither using a division-by-zero statement)

 

I've also tried to create a T-SQL with a "try-except" forcing a raiserror, but even then, I'm not able to catch the exception.

 

I have looked at the properties in the ADO.CreateADOCommand and there is apparently no property to flag the exception.

 

I am believing that the database suppresses this information and it is not returning to the application.

Or maybe I'm failing at something I have not yet observed.

 

Could anyone tell me what might be happening? And how could I solve this to get the exception I hope so much?
Thank you. :smileyhappy:

  • tristaanogre's avatar
    tristaanogre
    Esteemed Contributor
    7 years ago

    From your post, I'm assuming you're trying to get the SQL database queries to raise an exception on the SQL server side and then have your Delphi Script in TestComplete handle that exception, is that correct?

     

    If so... I don't believe that's possible.  Anything returned by ADO is not returned as raised exception but encapsulated in the ADO objects.  You'll need to retrieve any error codes, conditions, etc., from your result set and then use that information to raise an exception within your delphi script code.

    • jhonathanramos's avatar
      jhonathanramos
      Occasional Contributor
      7 years ago

      Hello, Robert,
      Thanks for your response. :smileyhappy:

       

      I'm not sending a query to the database expecting a set.

       

      Imagine that Delphiscript is sending an SQL statement as a "Restore Database TC from Database_Snapshot" or an "insert". Assuming that in 1st case the path to the backup file to be restored is not available at that time and the 2nd "insert" is prevented by integrity constraint.

       

      The delphiscript would only wait for the completion of the execution of the SQL statement that was sent, if execution is prevented for some reason. Should this reason not be raised as an exception?
      If this is not how it works, how will I know if it has failed or not? 

       

      Thank you. :smileyvery-happy:

      • tristaanogre's avatar
        tristaanogre
        Esteemed Contributor
        7 years ago

        Well, result set was just an example.  I don't know, off hand, the specifics of every SQL command that can be executed via ADO.  But the principle that I mentioned is still the same: exceptions raised in the SQL will not re-raise as exceptions in your DelphiScript.  So, yes, on the SQL server side, it would raise as an exception if your SQL is coded to do so.  But the ADO COmmand object "handles" that exception so that it is not reraised. In order for you to determine success/failure of the SQL you'll need to either examine whatever results you DO get back (could be numeric error code, status code, etc) OR, run a subsquent query to determine the state of the database after the command to see if it completed successfully.  

         

         I could be wrong in this, but that's been my experience.