From: SourceForge.net <no...@so...> - 2010-09-12 11:49:57
|
Bugs item #1103722, was opened at 2005-01-17 09:19 Message generated for change (Comment added) made by crategus You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1103722&group_id=4933 Please note that this message will contain a full copy of the comment thread, including the initial issue submission, for this request, not just the latest update. Category: Lisp Core - Translator Group: None >Status: Pending >Resolution: Wont Fix Priority: 5 Private: No Submitted By: Robert Dodier (robert_dodier) Assigned to: Nobody/Anonymous (nobody) Summary: compile_file claims compilation failed when it succeeded Initial Comment: compile_file uses the tertiary return value of COMPILE-FILE to tell whether the compilation succeeded. At least one lisp (Clisp) returns a positive value if there are warnings, so compile_file thinks the compilation failed when actually there weren't errors. A comment in the code (src/transs.lisp) says ;; If the compiler encountered errors, don't set bin-file to ;; indicate that we found errors. Is this what we want? I think the answer is "no". ---------------------------------------------------------------------- >Comment By: Dieter Kaiser (crategus) Date: 2010-09-12 13:49 Message: >From the postings of this bug report, there seems to be no consent what should be the correct and save behavior of compile_file. Furthermore, we have no example which shows a problem with the current behavior of the function compile_file. Because, this is a very old bug report, there seems to be no really problems. I suggest to close this bug report as 'won't fix'. Setting the status to pending. Dieter Kaiser ---------------------------------------------------------------------- Comment By: Nikodemus Siivola (demoss) Date: 2005-03-30 15:15 Message: Logged In: YES user_id=283348 The third return value of COMPILE-FILE indicates presence of errors or non-style-warning warnings; non-style-warning warnings are an indication of real problems and undefined behaviour, and should be treated as such: eg. SBCL emits full warnings for unbound variables and type-errors detected at compile-time. In other words, the current behaviour is IMO good and safe. ---------------------------------------------------------------------- You can respond by visiting: https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1103722&group_id=4933 |