J3/01-373R2 Date: 2001 December 4 To: J3 From: Subgroup C Subject: Chapter 2 comments 1)Note 2.1. Replace with: "Constraints and other information related to rules that do not begin with R2 appear in the appropriate Section." 2)C201. No action: constraints don't always follow the rule. 3)2.2. Deferred to subgroup B. 4)2.2.3. Delete "(9.5.4.7)" 5)2.2.1, line 5. "any order" -> "a processor-dependent order" Note 2.2. Insert space to indicate beginning of second paragraph. [ Unrelated problem: program text and procedures are confused.] 6)2.2.4. Delete 3rd sentence line 6. "may request access to" -> "may access" 7)2.3, lines 3-4. "and certain . . . constructs." -> "and not all executable statements may appear in all contexts." 8)2.3.2. No action. 9)Withdrawn. 10)Table 2-2. No action. 11)Table 2-2. Delete Note 2. 12) 2.4.1. No action. 13) 2.4.1.2, line 6. "structure constructors are" -> "a structure constructor is" 14) 2.4.2, line 4. Replace with "in 4.4. The" 2.4.2, line 6 "of the intrinsic types." -> "of its components." [ Related problem: The definition/description of "ultimate component" doesn't work for a derived type that has a component that is a pointer to that same type. ] 15)2.4.3.1. No action. Part of a complex can be defined by equivalencing a real to the imaginary part (e.g.) of the complex and defining the real. 16) Withdrawn 17) 2.4.3.1, line 1-. Add "also" before "may". 18)No action. Assignment is not an operation. 19)2.4.6. No action. There are other ways to cause a procedure to be executed, but they are not references. 20)2.5.5. Add at end of 2nd paragraph: "An occurrence of derived-type input/output (10.6.5) or derived-type finalization (4.5.10) also is a procedure reference. 21)Deferred to subgroup B. ------------------------------------------------------------------------------------------------------------- [ 2.5.4. Unrelated problem: We seem to have lost the distinction between becoming defined and being defined. ]