From owner-freebsd-ports@freebsd.org Wed Dec 25 16:49:49 2019 Return-Path: Delivered-To: freebsd-ports@mailman.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mailman.nyi.freebsd.org (Postfix) with ESMTP id B68A91E08B1 for ; Wed, 25 Dec 2019 16:49:49 +0000 (UTC) (envelope-from franco@lastsummer.de) Received: from host64.shmhost.net (host64.shmhost.net [213.239.241.64]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) server-signature RSA-PSS (4096 bits)) (Client did not present a certificate) by mx1.freebsd.org (Postfix) with ESMTPS id 47jfFm3NS2z44HM for ; Wed, 25 Dec 2019 16:49:48 +0000 (UTC) (envelope-from franco@lastsummer.de) Received: from francos-mbp.fritz.box (ip9234d229.dynamic.kabel-deutschland.de [146.52.210.41]) by host64.shmhost.net (Postfix) with ESMTPSA id 47jfFj5Fs9zJl0d; Wed, 25 Dec 2019 17:49:45 +0100 (CET) Content-Type: text/plain; charset=us-ascii Mime-Version: 1.0 (Mac OS X Mail 13.0 \(3608.40.2.2.4\)) Subject: Re: XXX needs Python 3.4 at least, but 2.7 was specified From: Franco Fichtner In-Reply-To: <9b0c65a4-07ad-9743-687a-81ec0b1d36ee@druid.net> Date: Wed, 25 Dec 2019 17:49:44 +0100 Cc: FreeBSD Ports Content-Transfer-Encoding: 7bit Message-Id: References: <1fe096e1-30e8-7739-3cbd-3575e32fbe18@utanet.at> <9a97aecf-a50e-e16a-a222-ea7844245e28@utanet.at> <9b0c65a4-07ad-9743-687a-81ec0b1d36ee@druid.net> To: D'Arcy Cain X-Mailer: Apple Mail (2.3608.40.2.2.4) X-Virus-Scanned: clamav-milter 0.101.4 at host64.shmhost.net X-Virus-Status: Clean X-Rspamd-Queue-Id: 47jfFm3NS2z44HM X-Spamd-Bar: / Authentication-Results: mx1.freebsd.org; dkim=none; dmarc=none; spf=none (mx1.freebsd.org: domain of franco@lastsummer.de has no SPF policy when checking 213.239.241.64) smtp.mailfrom=franco@lastsummer.de X-Spamd-Result: default: False [-0.40 / 15.00]; ARC_NA(0.00)[]; RCVD_VIA_SMTP_AUTH(0.00)[]; RECEIVED_SPAMHAUS_PBL(0.00)[41.210.52.146.khpj7ygk5idzvmvt5x4ziurxhy.zen.dq.spamhaus.net : 127.0.0.11]; FROM_HAS_DN(0.00)[]; MV_CASE(0.50)[]; IP_SCORE(-0.15)[ip: (0.81), ipnet: 213.239.192.0/18(-0.02), asn: 24940(-1.55), country: DE(-0.02)]; MIME_GOOD(-0.10)[text/plain]; DMARC_NA(0.00)[lastsummer.de]; AUTH_NA(1.00)[]; NEURAL_HAM_LONG(-0.85)[-0.851,0]; TO_MATCH_ENVRCPT_SOME(0.00)[]; TO_DN_ALL(0.00)[]; RCPT_COUNT_TWO(0.00)[2]; RCVD_IN_DNSWL_NONE(0.00)[64.241.239.213.list.dnswl.org : 127.0.10.0]; NEURAL_HAM_MEDIUM(-0.80)[-0.797,0]; R_SPF_NA(0.00)[]; FROM_EQ_ENVFROM(0.00)[]; R_DKIM_NA(0.00)[]; MIME_TRACE(0.00)[0:+]; ASN(0.00)[asn:24940, ipnet:213.239.192.0/18, country:DE]; MID_RHS_MATCH_FROM(0.00)[]; RCVD_TLS_ALL(0.00)[]; RCVD_COUNT_TWO(0.00)[2] X-BeenThere: freebsd-ports@freebsd.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Porting software to FreeBSD List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 25 Dec 2019 16:49:49 -0000 > On 25. Dec 2019, at 12:59 PM, D'Arcy Cain wrote: > > On 12/24/19 5:56 PM, w.schwarzenfeld wrote: >> At least a workaround (maybe the sollutiion): >> >> >> https://bugs.freebsd.org/bugzilla/attachment.cgi?id=209881&action=diff > > Another workaround if you don't want to modify files in the repo is to > simply go into the problem folder, /usr/ports/graphics/graphene in the > OP's case, and run "make install" before returning to the original > build. You may have to do it multiple times for different packages though. While Poudriere is nice and all and does this automagically it is a major annoyance trending lately that you cannot build ports without these silly build errors. If USES= python:XXXX is used it doesn't mean ports should just barf all over users with these non-errors when they don't build deps with that particular version that one didn't even specify for the deps. It seems like USES enforces this recursively but should not as witnessed by this "workaround". There's still a relevant DEFAULT_VERSIONS entry but it is happily ignored. If all users need to use Poudriere just to have a chance at building ports we're at the point where Poudriere itself is becoming problematic for port framework quality, something that always seemed to linger in the background for the last couple of years and now it's finally here. And yes, lang/rust was one recent example that has python:3.3+ now and fails to build somewhere down the dependency chain. Cheers, Franco