Date: Thu, 2 Nov 2000 20:51:29 +0800 From: Jing-Tang Keith Jang <keith@bsdvm.jtjang.idv.tw> To: freebsd-ports@FreeBSD.org Subject: Re: Something about ports/chinese Message-ID: <20001102205129.A7863@bsdvm.jtjang.idv.tw> In-Reply-To: <3A011826172.39E8FOXFAIR@intra.nextnode.com>; from foxfair@drago.fomokka.net on Thu, Nov 02, 2000 at 03:30:46PM %2B0800 References: <20001102144635.B5169@bsdvm.jtjang.idv.tw> <3A011826172.39E8FOXFAIR@intra.nextnode.com>
next in thread | previous in thread | raw e-mail | index | archive | help
On 11/02/00, Foxfair Hu wrote: > It's scheduled at my afternoon TODO list, and you all(keith and kevlo, > keichii -- why so many Ks? :P) make it faster, thanks! I see, thanks for committing. > Hmm... if outta-port is out of sync with your PR, or anyone's hot > codes at his own computer. I'll suggest that : We shouldn't maintain > two different ports tree(ports/chinese && outta-port in sinica) at the > same time. > Especially when someone wanna commit something, but he doesn't use the > source file from the PR's originator. If you had seen my post on Taiwan's FreeBSD mailing list a couple of weeks ago, you should know that I always sync outta-port with PR, and mark the outta-port as broken after the new port has sent in. So there is only one version that needs to be maintained. In netscape's case, we're lucky because adding XFree86-aoutlibs would fix it. But as for linux_base, not only both PR and outta-port are outdated, it's committed with a wrong name. I'm not sure how to fix this, since repo-copy cle_base to linux_base may be involved. Also see my comment below. > Actually I don't do "blink" commit at the normal time, but for some > reason I don't wanna test these ports(cle_base & netscape-*) in my > -current box. So I ask keichii to help me few hourse ago, to make sure > these ports are fine in installation procedure. I thought that's what jail(8) is for committers. :-) > And no, I didn't "ignore" these PRs keith sent, it is unfair when you > said it. If you have glanced the linux_base PR, you would know it's not cle_base. The PORTNAME in the Makefile is also linux_base. > That's why I called it "cle_base" rather then "linux_base", for the > later case : You should make a slave port or ask repo-copy first, and > send out your patch focus on Chinese. NOT make a new port everytime. > I mark the name different to do so, now, can you understand what I want > to do? :) It's clear to me now that you didn't even read the Makefile. If you have either read it or do something like `make fetch`, you would immediately know that it's impossible to make a slave port of emulators/linux_base, since the distfiles are entirely different. It's also impossible to do an "add-on" port, like japanese/linux_locale, because binaries like libX11 need to be replaced to support Big5. The only reason I'm wondering if a repo-copy is necessary, is that the overall structure is quite similar, just like www/netscape*, and it was indeed a modified version of marcel's linux_base port. Even though you thought it should be a new port, you didn't change the PORTNAME to cle_base. It's difficult to believe that you have checked the PR or the source carefully. Please, bygones, fix chinese/cle_base. Either repo-copy(may be slower) or move it to Attic and create a new chinese/linux_base would be fine. -- Keep it simple and stupid. To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20001102205129.A7863>