From owner-freebsd-ports Fri Jan 5 4:30:24 2001 From owner-freebsd-ports@FreeBSD.ORG Fri Jan 5 04:30:21 2001 Return-Path: Delivered-To: freebsd-ports@freebsd.org Received: from blizzard.sabbo.net (ns.sabbo.net [193.193.218.18]) by hub.freebsd.org (Postfix) with ESMTP id 6C85937B400 for ; Fri, 5 Jan 2001 04:30:18 -0800 (PST) Received: from vic.sabbo.net (root@vic.sabbo.net [193.193.218.109]) by blizzard.sabbo.net (8.10.1/8.10.1) with ESMTP id f05CUBM21448; Fri, 5 Jan 2001 14:30:13 +0200 Received: from notebook.FreeBSD.org (notebook.vega.com [192.168.1.100]) by vic.sabbo.net (8.11.1/8.9.3) with ESMTP id f05CUCX12594; Fri, 5 Jan 2001 14:30:12 +0200 (EET) (envelope-from sobomax@FreeBSD.org) Message-Id: <5.0.2.1.2.20010105142031.024c3c10@192.168.1.50> X-Sender: max@192.168.1.50 X-Mailer: QUALCOMM Windows Eudora Version 5.0.2 Date: Fri, 05 Jan 2001 14:29:55 +0200 To: Peter Pentchev From: Maxim Sobolev Subject: Re: Is there any reason for FETCH_BEFORE_ARGS? Cc: Ilya Martynov , freebsd-ports@FreeBSD.org In-Reply-To: <20010105141009.G10329@ringworld.oblivion.bg> References: <5.0.2.1.2.20010105135621.024d0550@192.168.1.50> <20010105133600.E10329@ringworld.oblivion.bg> <20010105124725.D10329@ringworld.oblivion.bg> <20010105133600.E10329@ringworld.oblivion.bg> <20010105133919.F10329@ringworld.oblivion.bg> <5.0.2.1.2.20010105135621.024d0550@192.168.1.50> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii"; format=flowed Sender: owner-freebsd-ports@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.org At 14:10 05.01.01, Peter Pentchev wrote: >On Fri, Jan 05, 2001 at 02:01:12PM +0200, Maxim Sobolev wrote: > > At 13:39 05.01.01, Peter Pentchev wrote: > > >On Fri, Jan 05, 2001 at 01:36:00PM +0200, Peter Pentchev wrote: > > >[snip] > > > > > > > > Hmm this is an interesting idea.. something along the lines of.. > > > > > > > > FETCH_BEFORE_ARGS=fetchprogram,hostname,args > fetchprogram,hostname,args ... > > > > > > > > e.g. > > > > > > > > FETCH_BEFORE_ARGS=fetch,http://server.needing.b/path/,-b \ > > > > > >err.. I meant.. FETCH_BEFORE_ARGS=fetch,server.needing.b,-b \ > > > > > > > fetch,server.needing.t,-t > > > > > > > > Then the do-fetch target should, for each URL, scan FETCH_BEFORE_ARGS, > > > > and use the arguments if and only if it finds a match on *both* the > fetch > > > > program and the hostname (not URL) from which it's currently trying > > > > to fetch. > > > > > > > > Something to think about.. > > > > Needless complexity, especially considering that `-b' is compatibility > > option which is likely to be deleted soon (IMHO, support for 3-STABLE now > > is very close to Attic). > >What you mean is, the whole concept of FETCH_BEFORE_ARGS should go? Of course no. Actually I was saying that per-URL fetch option is PITA and doesn't really necessary. >Or just that the ports that are setting -b shall stop setting it? >If it's the latter, then any ports setting any fetch(1)-specific options >shall still fail when another FETCH_CMD is being used :( Yes, but I don't see how this situation could be improved - neither POSIX nor SUSv2 says nothing about standard options for downloaders. :( We don't (and will not, IMO) support anything but fetch(1). -Maxim To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-ports" in the body of the message