Xref: utzoo news.software.b:3965 comp.sources.bugs:2179 Newsgroups: news.software.b,comp.sources.bugs Path: utzoo!henry From: henry@utzoo.uucp (Henry Spencer) Subject: Re: C News patch of 16-Jan-1990 Message-ID: <1990Jan17.180516.13414@utzoo.uucp> Organization: U of Toronto Zoology References: <1990Jan16.230037.15898@utzoo.uucp> Date: Wed, 17 Jan 90 18:05:16 GMT In article tale@cs.rpi.edu (David C Lawrence) writes: >What is the preferred method of getting C News re-installed after >patching? What I've been doing is running doit.bin -i, doit.bin, >doit.news and then again.root. The introduction for this patch, with >talk of build being revamped to use previous answers, makes me wonder >whether I was suppose to run build each time. (Well, things _seem_ to >be working ...) Is there anything added to build that require it >being used again? In general, re-build is seldom or never necessary. I should look at some way to inform people about this automatically, though. There are some possible changes in the queue that might require it. Actually, a re-build is probably in order now, because dbmclose() is an issue that affects the doit files. The doit.* sequence you suggest is reasonable in general. >By the way, how come it does -Dvoid=int? ... Um, where? The only instances of this I can find in our makefiles are commented out. There is one -DVOID=int, but that's a slightly different story -- the problem there is turkeys who have gratuitously changed system library functions that used to return (meaningless) int values to return void instead. Unfortunately, ANSI C has even blessed some of these gratuitous incompatibilities, so we can't just blame them on broken systems. This is one of those cases where you just can't win; the -DVOID=int is needed for most current systems but is technically wrong. -- 1972: Saturn V #15 flight-ready| Henry Spencer at U of Toronto Zoology 1990: birds nesting in engines | uunet!attcan!utzoo!henry henry@zoo.toronto.edu