From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: X-Spam-Checker-Version: SpamAssassin 3.3.2 (2011-06-06) on dcvr.yhbt.net X-Spam-Level: X-Spam-ASN: AS16276 94.23.0.0/16 X-Spam-Status: No, score=-1.1 required=3.0 tests=AWL,BAYES_00,BODY_8BITS, RCVD_IN_DNSWL_LOW,URIBL_BLOCKED shortcircuit=no autolearn=unavailable version=3.3.2 X-Original-To: unicorn-public@bogomips.org Received: from perdizione.investici.org (perdizione.investici.org [94.23.50.208]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by dcvr.yhbt.net (Postfix) with ESMTPS id 1786D1FE1C for ; Tue, 29 Sep 2015 08:00:44 +0000 (UTC) Received: from [94.23.50.208] (perdizione [94.23.50.208]) (Authenticated sender: praveen@autistici.org) by localhost (Postfix) with ESMTPSA id 9A247120FAA; Tue, 29 Sep 2015 08:00:41 +0000 (UTC) User-Agent: K-9 Mail for Android In-Reply-To: <20150929073650.GA7434@dcvr.yhbt.net> References: <560A31F1.3060608@debian.org> <20150929073650.GA7434@dcvr.yhbt.net> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Subject: Re: Request to follow SemVer/mention it in homepage From: Pirate Praveen Date: Tue, 29 Sep 2015 13:30:36 +0530 To: Eric Wong CC: unicorn-public@bogomips.org Message-ID: <40CB6BDA-920D-44E1-B364-D7228248C747@debian.org> List-Id: On २९ सप्टेंबर, २०१५ १:०६:५० [PM] IST, Eric Wong wrote: >Pirate Praveen wrote: >> Do you follow Semantic Versioning as defined by semver.org? If yes, >> can you mention it on the homepage? If not, can you consider it? > >Maybe we follow semver... > >No, I'd rather not mention it on the homepage or be formally/officially >bound to it (see below) > >I have considered it in the past, but decided against it... > >unicorn is on the verge of becoming 5.x: > >This is for internal changes which some weirdo projects (notably >Rainbows!) rely on; but was never considered public API or marketed >as such. I absolutely do not want people building server-specific >apps when Rack exists. Can you consider SemVer from 5.x? >Rack will also hit 2.x soon, but regardless of incompatible changes >in Rack, unicorn must remain compatible with both 1.x and 2.x for >practical reasons. > >> It will help us a lot in maintain rails applications in debian. We >> like to keep only one version of any app/lib in debian and if you can >> guarantee Semantic Versioning rails apps using unicorn could declare >a >> looser dependency rather than upto the patch version. Now gitlab >> defines unicorn ~> 4.8.3 and diaspora defines ~> 4.9.0 If you are >> following Semantic Versioning they could change it to ~> 4.8 and ~> >> 4.9 and 4.9.0 will satisfy both. Currently debian has 4.9.0 and it >> does not match ~> 4.8.3 > >Tying a Rack app to unicorn is totally, completely wrong and defeats >the >point of Rack. > >Honestly, I could understand tying an app to thin with it's EM-specific >API, but unicorn(!?!?) Even if an app isn't thread-safe, it should >still work in passenger, thin, or puma/webrick configured for a >single-thread... > > >Regardless of a formalities such as semver, I'll work to ensure unicorn >can be compatible[1] with any Rack apps in Debian main. >Heck, perhaps Rack 0.9.x still works (assuming the installed Ruby >version supports it). > >Just let us know what real problems you find, they will be fixed. If you can convince gitlab to declare ~> 4.8 and assure it will work with 4.9, that will solve the issue for now. See https://gitlab.com/gitlab-org/gitlab-ce/issues/2820 Also if you can help convinse diaspora upstream to declare a looser dependency, that will also help. >Fwiw, Debian is my preferred platform and I can stay 100% within email >with the Debian BTS + lists. Glad to know. > >[1] I would never say unicorn is the best choice for all apps, > but it should at least limp along with sufficiently loose > definitions of "working", perhaps with hundreds of inefficient > processes in some cases. -- Sent from my Android device with K-9 Mail. Please excuse my brevity.