From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thomas Petazzoni Date: Thu, 20 Jun 2019 08:31:10 +0200 Subject: [Buildroot] [PATCH 1/4] package: remove 'v' prefix from github-fetched packages In-Reply-To: <35a7118e-cdf1-b29e-bceb-6c02ecef187e@mind.be> References: <20190612064209.23619-1-victor.huesca@bootlin.com> <20190612064209.23619-2-victor.huesca@bootlin.com> <35a7118e-cdf1-b29e-bceb-6c02ecef187e@mind.be> Message-ID: <20190620083110.4986e5fa@windsurf> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: buildroot@busybox.net On Wed, 19 Jun 2019 22:50:01 +0200 Arnout Vandecappelle wrote: > Note: I double-checked: apparently, release-monitoring really *never* has a v > in front of the version. There are a few other things though, e.g. DIRECTFB_ I'm not sure what you referring to regarding DIRECTFB_. However, release-monitoring doesn't overall has something that enforces keeping or not keeping a version prefix. It is the definition of each package on release-monitoring that details how to extract the version, typically stripping any prefix. Because this is the norm on release-monitoring (stripping the prefix), any package definition that does not comply with this should probably be fixed on release-monitoring. My point is that if there is a bogus package definition on release-monitoring, we should not work it around in Buildroot, but fix it on release-monitoring itself. > It would also be nice if we had some formal way of checking that new packages > do this correctly. Actually, it would be good if pkg-stats could generate a > report in a human-readable format so it could be used as part of the review flow > to check if the mapping with release-monitoring is OK (plus all the other things > pkg-stats evaluates). Hm, yes, could be an interesting idea. Just a new pkg-stats option ? Thomas -- Thomas Petazzoni, CTO, Bootlin Embedded Linux and Kernel engineering https://bootlin.com