From mboxrd@z Thu Jan 1 00:00:00 1970 From: =?UTF-8?Q?G=C3=A1bor?= Boskovits Subject: bug#30113: SVN checkouts without descriptive file names Date: Sun, 14 Jan 2018 18:43:55 +0100 Message-ID: References: <20180114160253.GA17578@jasmine.lan> <04cea8f4-3bee-9992-2e9e-b5c2e25ca5a8@tobias.gr> <20180114175311.3d463379@scratchpost.org> Mime-Version: 1.0 Content-Type: multipart/alternative; boundary="001a113c00d6b464070562c00a60" Return-path: Received: from eggs.gnu.org ([2001:4830:134:3::10]:45354) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1eamLC-0000YL-5J for bug-guix@gnu.org; Sun, 14 Jan 2018 12:45:07 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1eamL8-00070l-52 for bug-guix@gnu.org; Sun, 14 Jan 2018 12:45:06 -0500 Received: from debbugs.gnu.org ([208.118.235.43]:48377) by eggs.gnu.org with esmtps (TLS1.0:RSA_AES_128_CBC_SHA1:16) (Exim 4.71) (envelope-from ) id 1eamL8-00070X-0k for bug-guix@gnu.org; Sun, 14 Jan 2018 12:45:02 -0500 Received: from Debian-debbugs by debbugs.gnu.org with local (Exim 4.84_2) (envelope-from ) id 1eamL7-00016Z-Q7 for bug-guix@gnu.org; Sun, 14 Jan 2018 12:45:01 -0500 Sender: "Debbugs-submit" Resent-Message-ID: In-Reply-To: <20180114175311.3d463379@scratchpost.org> List-Id: Bug reports for GNU Guix List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: bug-guix-bounces+gcggb-bug-guix=m.gmane.org@gnu.org Sender: "bug-Guix" To: Danny Milosavljevic Cc: 30113@debbugs.gnu.org --001a113c00d6b464070562c00a60 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable Ok, I started a guix lint on 0.14.0.1183-1b321. Will attach log here asap. 2018-01-14 17:53 GMT+01:00 Danny Milosavljevic : > On Sun, 14 Jan 2018 17:43:27 +0100 > Tobias Geerinckx-Rice wrote: > > > G=C3=A1bor Boskovits wrote on 14/01/18 at 17:13: > > > Maybe we could use guix to check for these, and some > > > other things could also be spotted. > > > WDYT? > > > > Agreed, I think. > > Yeah, +1 > > > We should be able to improve the quality of these guesses: the > > repository URI is about as likely to be foo://bar/... as a > > regular tarball URI. > > > > Or we make a file-name mandatory for certain methods. > > I agree that some heuristics to figure out the file-name from > (svn-reference-url ref) would be nice. > > It's not that important that the store filenames are meaningful to humans= , > it's just nice-to-have. So I'd say heuristic it and be done with it. > > On second thought, maybe even default file-name to the Guix package name > (if possible at this point in Guix). The hash value will make it unique > regardless (for example if there are multiple svn-download blocks in the > same package) > --001a113c00d6b464070562c00a60 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
Ok, I started a guix lint on=C2=A00.14.0.1183-1b321.
W= ill attach log here asap.


2018-01-14 17:53 GMT+01:00 Danny Milosav= ljevic <dannym@scratchpost.org>:
On Sun, 14 Jan 2018 17:43:27 +0100
Tobias Geerinckx-Rice <me@tobias.gr&= gt; wrote:

> G=C3=A1bor Boskovits wrote on 14/01/18 at 17:13:
> > Maybe we could use guix to check for these, and some
> > other things could also be spotted.
> > WDYT?
>
> Agreed, I think.

Yeah, +1

> We should be able to improve the quality of these guesses: the
> repository URI is about as likely to be foo://bar/<package>... a= s a
> regular tarball URI.
>
> Or we make a file-name mandatory for certain methods.

I agree that some heuristics to figure out the file-name from (svn-r= eference-url ref) would be nice.

It's not that important that the store filenames are meaningful to huma= ns, it's just nice-to-have.=C2=A0 So I'd say heuristic it and be do= ne with it.

On second thought, maybe even default file-name to the Guix package name (i= f possible at this point in Guix).=C2=A0 The hash value will make it unique= regardless (for example if there are multiple svn-download blocks in the s= ame package)

--001a113c00d6b464070562c00a60--