From lrr@cray.com Mon Jan 27 21:24:04 1997 Received: from timbuk.cray.com (root@timbuk.cray.com [128.162.19.7]) by dkuug.dk (8.6.12/8.6.12) with ESMTP id VAA22694 for ; Mon, 27 Jan 1997 21:23:42 +0100 Received: from ironwood.cray.com (ironwood-fddi.cray.com [128.162.21.36]) by timbuk.cray.com (8.8.4/CRI-gate-8-2.11) with ESMTP id OAA13145 for ; Mon, 27 Jan 1997 14:23:37 -0600 (CST) Received: from poplar409.cray.com (lrr@poplar409 [128.162.149.9]) by ironwood.cray.com (8.8.4/CRI-ccm_serv-8-2.8) with SMTP id OAA18910 for ; Mon, 27 Jan 1997 14:23:14 -0600 (CST) From: Larry Rolison Received: by poplar409.cray.com (8.6.12/btd-b3) id OAA08491; Mon, 27 Jan 1997 14:23:32 -0600 Message-Id: <199701272023.OAA08491@poplar409.cray.com> Subject: Informal Review of Source Processing Draft Standard To: sc22wg5@dkuug.dk Date: Mon, 27 Jan 1997 14:23:31 -0600 (CST) X-Mailer: ELM [version 2.4 PL24-CRI-b] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Much of this was derived from the informal ballot announcement that Miles crafted and sent out regarding N1243 (the "Fortran-like" conditional compilation paper). This call for an informal ballot is for N1247 (the "cpp-like" source processing paper). I have sent a properly formatted version of the current version of the "cpp-like" source processing paper to both Miles (N1247) and to Mallory (97-112). I would like to discuss this document at the special meeting of the Development Body on Wednesday evening in Las Vegas, when I shall expect the DB to provide recommendations to the full WG in the form of draft resolutions as to what should be done next. These draft resolutions will be discussed by the full WG during preliminary resolution processing on Thursday. In order to expedite the discussions on the Wednesday evening, I would like to conduct an informal review of the document by electronic mail, so that everyone, including those who cannot attend the Las Vegas meeting, will be able to contribute their opinions. As a (very) brief reminder of the recent history of this topic, WG5 has discussed the issue of "conditional compilation" at a number of meetings, as has X3J3. At the last opportunity for input from individual countries, the U.S. stated that it does not feel the current "Fortran-like" draft should go forward as a draft standard because the "cpp-like" version has not been given sufficient consideration. In Las Vegas, WG5 needs to decide if the "Fortran-like" draft or the "cpp-like" draft is the preferred model of source processing and then needs to decide if whether the draft of the chosen model is ready for submission to SC22 for its first CD ballot, whether the chosen draft requires more work, or whether some other action is required. As I stated in my reply to the ballot on the "Fortran-like" conditional compilation paper (which I sent only to Miles), the "Fortran-like" model suffers from not one but two fatal flaws: * It does not standardize common practice (the form of conditional compilation generally accomplished through the use of cpp or cpp-like source processors). * It does not contain contain any facilities whatsoever for macro definition and expansion. Thus, it is my contention that the "Fortran-like" model can not succeed in the Fortran marketplace. Fortran application writers WILL NOT use the "Fortran-like" model because it does not match anything that they have now, and they CAN NOT use the "Fortran-like" model because it does not contain any function-like macro expansion capabilities. These are the main reasons why I continue to pursue the "cpp-like" model. Please read the draft (N1247), and then reply to *me* (not Miles) by email as follows NO LATER THAN WEDNESDAY 5TH FEBRUARY. (Note that whether you copy your informal ballot to the whole WG5 list is up to you. It might be helpful, but you may wish to keep your coments conficential - at least until after the ballot closes.) ---------------------------------------------------------------------------- Informal Ballot on N1247 (Draft Standard for Source Processing in Fortran) ____________ I believe that the current draft should supplant | YES | NO | the current "Fortran-like" model in all further | | | considerations of source processing ("conditional ------------ compilation") If NO, please provide details of what changes you wish to see made in order that the document will be suitable for substitution for the "Fortran-like" model. NOTE: I am NOT interested in NO votes based on either of * WG5 already voted to use the "Fortran-like" model (the U.S. has voted that the "cpp-like" model has not been given sufficient consideration) * It is too late to switch models now (it is NEVER too late in language design or in life to do the right thing) I am only interested in TECHNICAL reasons why you believe N1247 is insufficient as a source processing model for Fortran. I AM interested in responses to questions that exist in the current version of the paper (but not as reasons for voting NO). (signed) ........................................ (date) ........................................ Return to lrr@cray.com to arrive no later than midnite in your time zone on Wednesday 5th February 1997. ------------------------------------------------------------------------------ Larry Rolison lrr@cray.com Cray Research - A Silicon Graphics Company 655F Lone Oak Drive Eagan, MN 55121 ------------------------------------------------------------------------------