Hi John. The answer to your question is indeed no. Only properties, sequences, and let have some distinct non-terminals (so far as I saw). So those constructs are already inconsistent with tasks and functions. Thus, the following comment in the SV-AC minutes is not really correct: "introducing a new non-terminal identical to sequence_actual_arg ... would also introduce an inconsistency between the BNF of sequences on the one hand, and the BNF of properties, functions, and tasks on the other hand" They are already inconsistent. Shalom ________________________________ From: Havlicek John-R8AAAU [mailto:john.havlicek@freescale.com] Sent: Tuesday, April 14, 2009 5:34 PM To: Bresticker, Shalom; Korchemny, Dmitry; sv-ac@server.eda.org Cc: Havlicek John-R8AAAU Subject: RE: [sv-ac] RE: feedback to P1800 WG on ballot issues Hi Shalom: You are focusing on the "actual" substring in the BNF non-terminal. I reasoned more abstractly. Do tasks and functions have distinct non-terminals in the positions of default argument and actual argument in an instance? I think that the answer is no. This is the precedent that Dmitry was referring to. Regardless of what the actual non-terminal is, are there distinct non-terminals in the productions for tasks and functions. J.H. ________________________________ From: owner-sv-ac@eda.org [mailto:owner-sv-ac@eda.org] On Behalf Of Bresticker, Shalom Sent: Tuesday, April 14, 2009 4:04 AM To: Korchemny, Dmitry; sv-ac@server.eda.org Subject: [sv-ac] RE: feedback to P1800 WG on ballot issues Sure. I personally agreed with the comment, but I would not vote 'no' because of it. The only use in the BNF of actual_argument is in assertion constructs, so you can't appeal to other parts of the BNF for precedent or consistency. Shalom ________________________________ From: Korchemny, Dmitry Sent: Tuesday, April 14, 2009 11:57 AM To: Bresticker, Shalom; sv-ac@server.eda.org Subject: RE: feedback to P1800 WG on ballot issues But the default argument is and actual argument - it is a default actual argument. If we introduce new terminals for all possible use cases, the BNF will become completely unmanageable. I think this issue should be discussed by all committees to elaborate the common methodology. I don't feel comfortable to such changes at the last minute. Regards, Dmitry From: Bresticker, Shalom Sent: Tuesday, April 14, 2009 9:20 AM To: Korchemny, Dmitry; sv-ac@server.eda.org Subject: RE: feedback to P1800 WG on ballot issues The point is that they don't use "actual_function_argument". The use of "actual" is confusing. By your argument, sequence_actual_arg is also inconsistent with tasks and functions, which use simply 'expression'. Note for example that there are many types of 'identifier' that all reduce to 'identifier'. For example, array_identifier ::= identifier block_identifier ::= identifier Each is used where the semantics match the name, even though they are syntactically identical. array_identifier is not used for blocks, and block_identifier is not used for arrays. On the other hand, 'identifier' and 'expression' are generic names with no semantic information about where they are used. They could be used anywhere they are syntactially correct. But if you use a name that has some semantic meaning, then you should not use it where the meaning does not match its use. Regards, Shalom ________________________________ From: Korchemny, Dmitry Sent: Tuesday, April 14, 2009 9:13 AM To: Bresticker, Shalom; sv-ac@server.eda.org Subject: RE: feedback to P1800 WG on ballot issues Hi Shalom, But the tasks and functions do not use default_function_argument, do they? Dmitry From: Bresticker, Shalom Sent: Tuesday, April 14, 2009 7:57 AM To: Korchemny, Dmitry; sv-ac@server.eda.org Subject: RE: feedback to P1800 WG on ballot issues Hi, 87 - Mantis 2649: sequence_actual_arg is used to represent the default argument SV-AC believes that there is no added value in introducing a new non-terminal identical to sequence_actual_arg. It would also introduce an inconsistency between the BNF of sequences on the one hand, and the BNF of properties, functions, and tasks on the other hand. Therefore SV-AC recommends to leave the text unchanged. [SB] Why would this be inconsistent with tasks and functions? Tasks and function BNFs use simply "expression" for default values. Module port defaults use "constant_expression". Neither uses "actual" for "default". As for properties, yes, the same problem exists in the property BNF. I personally agree with the comment, though I did not submit it. Shalom --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -- This message has been scanned for viruses and dangerous content by MailScanner<http://www.mailscanner.info/>, and is believed to be clean. --------------------------------------------------------------------- Intel Israel (74) Limited This e-mail and any attachments may contain confidential material for the sole use of the intended recipient(s). Any review or distribution by others is strictly prohibited. If you are not the intended recipient, please contact the sender and delete all copies. -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean.Received on Thu Apr 16 04:41:28 2009
This archive was generated by hypermail 2.1.8 : Thu Apr 16 2009 - 04:42:00 PDT