On Oct 7, 2013, at 1:42 PM, Gabriel Dos Reis wrote:

> At the evening session (Thursday September 26) at the Chicago meeting,
> there was a consensus expressed that erroneous preprocessor constructs
> should not be ground for undefined behavior.
>
> The attached PDF file is my current attempt at implementing that consensus.
> Please read and comment. Suggestions for improvements, corrections welcome.


sec 3 para 1 typos:
  "the real" -> "the realm"
  "prescription" -> "proscription"

sec 3 para 1 suggested rephrasing:
  "… a proscription (on the effect of certain potentially evaluated string literal expressions) that is outside …"

sec 2.2/6 and 2.3 suggestion:
  make the literal more recognizable by stating it as a power of two minus one

sec 3 para 2 suggested rephrasing:
  "… limit on the line-number explicitly recognizes an implicit constraint (on programs) that is best diagnosed …"

sec 2 title: "Changes" -> "Proposed wording"

sec 1 para 2: "section §3" -> either "section 3" or "§3"

sec 1 para 1: "At that meeting, the attendance felt strongly that …" -> "Those attending that meeting felt strongly that …" or "That meeting's attendees felt strongly that …"

sec 2.3: "… in #line directive …" -> "… in a #line directive …"


Best,

-- WEB