From owner-freebsd-chat Sat Jun 15 08:59:45 1996 Return-Path: owner-chat Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id IAA01375 for chat-outgoing; Sat, 15 Jun 1996 08:59:45 -0700 (PDT) Received: from haywire.DIALix.COM (root@haywire.DIALix.COM [192.203.228.65]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id IAA01368 for ; Sat, 15 Jun 1996 08:59:39 -0700 (PDT) Received: (from news@localhost) by haywire.DIALix.COM (8.7.5/8.7.3) id XAA04359 for freebsd-chat@freebsd.org; Sat, 15 Jun 1996 23:59:34 +0800 (WST) X-Authentication-Warning: haywire.DIALix.COM: news set sender to usenet-request@haywire.dialix.com using -f Received: from GATEWAY by haywire.DIALix.COM with netnews for freebsd-chat@freebsd.org (problems to: usenet@haywire.dialix.com) To: freebsd-chat@freebsd.org Date: 15 Jun 1996 15:59:34 GMT From: peter@spinner.DIALix.COM (Peter Wemm) Message-ID: <4puml6$3f5$1@haywire.DIALix.COM> Organization: DIALix Services, Perth, Australia. References: <199606031449.PAA02874@cadair.elsevier.co.uk> Subject: Re: Indentation styles Sender: owner-chat@freebsd.org X-Loop: FreeBSD.org Precedence: bulk In article <199606031724.TAA19354@uriah.heep.sax.de>, j@uriah.heep.sax.de (J Wunsch) writes: > As Paul Richards wrote: > >> J> The last set of cosmetical ``fixes'' of this type was the Great >> J> ``Remove trailing whitespace'' saga, and it has been considered a >> J> big mistake afterwards. It breaks all third-party software (and >> J> remember, many things of us _are_ third-party, since they've got >> J> bsd_44_lite vendor-branch tags). >> >> Wasn't this because they weren't on a separate vendor branch? If they >> had been the new release would have patched cleanly against the vendor >> branch and then merged into the main branch without any problems. > > Well, they are no ``outside source'' patches, but internally > maintained. Thus, they don't really belong into a vendor branch. > > Anyway, CVS is a nice tool, but everything with vendor-branching seems > to be messy. Ask Peter Wemm for more detailed stories... The outcome > of the Greate Whitespace Saga was ``Never again''. It was a rather > useless operation that costed many hours of conflict resolution later. Heh, more like "hundreds of hours" so far, and still counting. I bet poor Jeffrey Hsu is silently cursing it still, since he's working on the kernel source as a whole with the Lite-2 merge he's working on. The basic problem is that cvs (and rcs) is just plain dumb, and cannot tell the difference between "just whitespace conflicts" and "real conflicts". When it comes to bigger things like taylor-uucp, bind, xntpd, doing a 3-way merge of the new version is rather complicated for no real benefit. Indeed, on quite a few occasions, we've reverted and put *back* the whitespace so that we can get cleaner diffs relative to the vendor version.... Never again.... (same goes for indent... Only indent only home-grown stuff that "belongs" to you or suffer the wrath of the other developers... Cosmetic changes for nothing more than personal preference is highly frowned apon as it can destroy parallel work others may have that has not been committed yet.) Cheers, -Peter > -- > cheers, J"org > > joerg_wunsch@uriah.heep.sax.de -- http://www.sax.de/~joerg/ -- NIC: JW11-RIPE > Never trust an operating system you don't have sources for. ;-)