Hi Frank, > Le 8 déc. 2019 à 04:18, Frank Heckenbach <f.heckenb...@fh-soft.de> a écrit : > > Hi Akim, > > I can't seem to apply your patches right now (e.g., they have > > -# define YYDPRINTF(Args) do {} while (yyfalse) > +# define YY_DPRINTF(Args) do {} while (yyfalse) > > while 3.4.91 has no such line, only some instances of > > # define YYDPRINTF(Args)
The patch I sent was against the current master. The commits have been pushed, you can get them from git. But 3.4.92 should be out RSN. > sometimes with a \ at the end), but it's not an urgent issue to me, > so I'll just wait for the next prerelease and test again then. But > don't bother making a prerelease just for that (unless you were > going to release 3.5 otherwise ;). I really would like to have feedback before the release :) >> Second, because I just could not find a means to have this work >> with between 4.8 and 5 included. With these compilers putting >> each call to YYFPRINTF between a pair of >> YY_IGNORE_USELESS_CAST_BEGIN and YY_IGNORE_USELESS_CAST_END works, >> but it's too painful and clutters the source. > > I agree that cluttering the source would be ugly, but if it's only > in the skeletons, couldn't you do it with m4? Yes, I can do that, but I meant to have something ok in the output too. Cheers!