Hi, with 2.5.1 on the horizon, a look at the remaining bugs on Savannah (https://savannah.gnu.org/support/?group=bison):
- "Testcase #50 failing non-deterministically for bison 2.4": I assume this was fixed by 47fa57; at least lib/subpipe.c is gone anyway :-). - "Updating to bison 2.4 breaks WebKit": This seems to have been a duplicate of http://permalink.gmane.org/gmane.comp.parsers.bison.bugs/3078 and solved on WebKit's side per https://bugs.webkit.org/show_bug.cgi?id=22205. - "YYPOPSTACK used inconsistently": The user infers from yacc.c to "garbage collected environments". The code in lalr1.cc looks fine. - "--no-parser and %no-parser option is broken": Probably "Wont Do". - "ambiguity problem with bison GLR parser": The reporter initially misread the manual. - "Possible bug in overflow handling": After staring for half an hour at yacc.c's code paths, I think this is not a bug (and hope that saving a few bytes is really worth the very "special" memory management Bison uses :-)). - "parser stack overflow handling + documentation on it": The bug's title should be "Document yyoverflow()". - "C++ template makes it very difficult to override func- tions": This is probably a matter of taste. Tim
