Hi dp,
>Only problem is same symbol representing different operators and has
>different precedence
>depending on context.
>
>sere & bool & bool
>
>second '&' has higher precedence then first '&'.
It is true that "&" as a sequence operator has different precedence than
"&" as an HDL operator. However, note that as Erich has indicated, the
interpretations where the difference matters are not allowed by the
grammar. Thus, a yacc grammar can be coded by giving the "&" operator the
HDL precedence, and letting the grammar rules do the rest.
Regards,
Avigail
_______________________________________________________
Avigail Orni
Formal Methods Group
IBM Haifa Research Laboratory
Tel: 972-4-829-6396 ornia@il.ibm.com
dp
<dp@realintent.co To: Avigail Orni/Haifa/IBM@IBMIL
m> cc: erichm@cadence.com, owner-vfv@eda.org, vfv@eda.org
Subject: Re: question
01/07/2004 23:59
Avigail Orni wrote:
> The grammar in
>the LRM is an abstract grammar whose purpose is to define the language. It
>cannot be used as it is for building a yacc parser. However, it should be
>possible to build an _equivalent_ conflict-free yacc parser -- this is
>
Well if flex/yacc tools can't be used for generating conflict free
parser then, I think it is a serious limitation. It is just very-2
complicated for developers to impliment anything.
Only problem is same symbol representing different operators and has
different precedence
depending on context.
sere & bool & bool
second '&' has higher precedence then first '&'.
dp
Received on Mon Jul 5 04:17:27 2004
This archive was generated by hypermail 2.1.8 : Mon Jul 05 2004 - 04:18:08 PDT