01-389r1 JOR Response to 01-357r1 - Comments on Section 9 To: J3 From: Craig Dedo Date: December 4, 2001 Subject: JOR Response to 01-357r1 - Comments on Section 9 JOR reviewed the recommendations in 01-357r1 and recommends as follows. References are to both the 01-007r3 and 01-007r4. 1 Edits r3: [168:16] r4: [170:25] Yes. This is fixed in 01-007r4. r3: [192:3] r4: [194:30] Yes. Delete the extraneous period (".") between "available" and "in a". r4: [195:1] Add a period at the end of the sentence. r3: [204:6] r4: [207:19] No. We use this construct in similar circumstances in several other places, most notably in 9.9.3 and 9.9.5. Remove the period at the end of the line in 01-007r4. This period does not exist in 01-007r3. 2 Combine Sections 9.5.3 and 9.9 This is a complex issue. Postpone until later. 3 Miscellaneous r3: [169:19] r4: [171:13] Delete "permanently". r3: [194:21+] r4: [197:19+] Add the following note. The EOR= specifier is ignored if the pending data transfer operation is not a non-advancing READ. The END= specifier is ignored if the pending data transfer operation is not a READ. r3: [179:27-29] r4: [182:3-5] Do nothing. r3: [184:7-8] r4: [186:29-31] Wait until JOR decides what to do about the recommendations in section 2. r3: [197:49+] r4: [200:3-4] No. The answer to this question is at the beginning of section 9.8.1. r3: [205:2,5] r4: [208:18, 21] Change "variable" to "do-variable" in two places. 4 MTE Proposal JOR decided for a variety of reasons that it is best to leave the rules about backspacing over records as they are. In practice, most Fortran compilers will allow the program to backspace over records written using list-directed or namelist formatting since there is no way to tell if a particular record in a file was written using either of these methods or regular format expressions. References 01-007r4, Fortran 2000 Draft [End of J3 / 01-389r1]