Re: [sv-ec] class scope resolution operator


Subject: Re: [sv-ec] class scope resolution operator
From: Arturo Salz (Arturo.Salz@synopsys.com)
Date: Thu Mar 13 2003 - 13:17:58 PST


Francoise,

My comments are interspersed below in blue.

    Arturo
  ----- Original Message -----
  From: Francoise Martinolle
  To: Arturo Salz ; sv-ec@eda.org
  Cc: David W. Smith
  Sent: Thursday, March 13, 2003 9:20 AM
  Subject: Re: [sv-ec] class scope resolution operator

  Comments and questions:

  Section 11.8.1:

  static task foo(); ... endtask // static class method with automatic variable lifetime
  task static bar(); ... endtask // non-static class method with static variable lifetime

  In the declaration for bar, does the static keyword applies to the variables defined in the
  task bar? Is it specific to tasks and functions declared in classes?

  The declaration for bar is an extension of v2k, which allows
      task automatic bar() ... endtask

  The enhancement is the use (and existence) of the static keyword
  to denote the lifetime of arguments and variables.

  This is not specific to class methods, what's specific to class methods
  is the declaration of foo.

  Section of 11.20:

  I would like to replace the :: with the normal . used in Verilog for scope delimitor.

  See my earlier resposne to Jay.

  "Class-scope resolved expressions
  can be read (in expressions), written (in assignments or subroutines calls) or triggered off (in event expressions)."
  I think you may want to replace "expressions" with "names".

  No. I don't think so. I meant expressions.
  .
  The following sentence is unclear in its meaning:
  "Type declarations nested inside a class scope are public and can be accessed outside
  the class."

  What that means is that "type decarations inside a class cannot be declared local or protected".
  This might have made more sense when the public keyword existed.

  Yes they may be public but they have to be accessed through the parent class scope.
  Adding something like that would be more clear, otherwise, it seems that nested class
  types can be accessed directly without scope nesting. I suggest to add " can be accessed
  outside the class through the appropriate nested scope path.

  Francoise
         '

    03:19 AM 3/12/2003 -0800, Arturo Salz wrote:

    Greetings,

    Attached is the proposal for the class scope resolution operator :: (AI-56)
     
        Arturo



This archive was generated by hypermail 2b28 : Thu Mar 13 2003 - 13:16:19 PST