Loading ...
Sorry, an error occurred while loading the content.

Re: НА: [eiffel_software] Eiffel 7.3 rep orting incorrect VDRDs

Expand Messages
  • Thomas Beale
    ... Some good news finally... the gobo changes above have worked. My parser code now compiles as is, bar a couple of places where the production rule allowed
    Message 1 of 6 , Sep 1, 2013
    • 0 Attachment
      On 31/08/2013 18:38, Eric Bezault wrote:
      > On 8/31/2013 2:14 PM, Thomas Beale wrote:
      >> On 31/08/2013 02:57, Alexander Kogtenkov wrote:
      >>> Would adding a postcondition "False" to the routine abort_with_error
      >>> do the trick as well?
      >>
      >> you mean because the compiler will see it and no longer require any
      >> extant variables to be attached after that point in the routine
      >> execution? That sounds worth trying. I'll report back ...
      >
      > Whatever the solution adopted (False postcondition or
      > "check aborted: False then end"), the version of geyacc
      > with the rescue clause has been checked-in in Git.
      >

      Some good news finally...

      the gobo changes above have worked. My parser code now compiles as is,
      bar a couple of places where the production rule allowed an empty value,
      and therefore really did have to be detachable. After a couple of hours
      updating other bits of my code, I have a full compilation under 7.3
      (note: with latest gobo Git tool generated parsers, and libs, not the
      $ISE_EIFFEL ones).

      I'm testing it now, but superficially, things appear to be working
      normally in the application.

      Now we are making some real progress...!

      - thomas
    • Peter Gummer
      ... Without adding any False postconditions or checks! I don t understand how simply adding a rescue clause has allowed this to compile. - Peter Gummer
      Message 2 of 6 , Sep 1, 2013
      • 0 Attachment
        Thomas Beale <thomas.beale@...> wrote:

        > Some good news finally...
        >
        > the gobo changes above have worked. My parser code now compiles as is,
        > bar a couple of places where the production rule allowed an empty value,
        > and therefore really did have to be detachable.


        Without adding any False postconditions or checks!

        I don't understand how simply adding a rescue clause has allowed this to compile.

        - Peter Gummer
      • Thomas Beale
        ... I did add the False post-condition to abort procedures. - thomas
        Message 3 of 6 , Sep 2, 2013
        • 0 Attachment
          On 02/09/2013 04:26, Peter Gummer wrote:
          > Thomas Beale <thomas.beale@...> wrote:
          >
          >> Some good news finally...
          >>
          >> the gobo changes above have worked. My parser code now compiles as is,
          >> bar a couple of places where the production rule allowed an empty value,
          >> and therefore really did have to be detachable.
          >
          > Without adding any False postconditions or checks!

          I did add the False post-condition to abort procedures.

          - thomas
        Your message has been successfully submitted and would be delivered to recipients shortly.