Dixi quod…
>Ian Jackson dixit:
>
>>Thorsten Glaser writes ("Re: Bug#858769: "cvs init" creates CVSROOT/history
>>and val-tags world-writeable"):
>>> 0664, actually; CVS operates group-writable.
>>
>>Does this not depend on the umask ?
>
>To some degree, yes. On the other hand, there’s an explicit chmod
>in there, and normally you operate with a CVSUMASK of 002 precisely
>to work in groups.
On further reflection, I removed the chmod() calls, making the
permission bits of both files depend only on ${CVSUMASK:-002}
as is customary for other files in the repository (including
updating the documentation accordingly).
The regular Unix umask is used for files in the working copy
(the checkout). This is to distinguish between these two, as
people will normally work with umask 022 and CVSUMASK 002 or
possibly 007 to facilitate sharing a repository among a Unix
user group.
Of course, this means I’ll have to restart running the insane
testsuite of this… but hey, progress! In the meanwhile, if you
wish to inspect the whole patchset (I’ll have to split off the
parts I can justify getting into stretch despite the freeze):
https://anonscm.debian.org/cgit/collab-maint/cvs.git/log/?h=mirbsd
bye,
//mirabilos
--
18:47⎜<mirabilos:#!/bin/mksh> well channels… you see, I see everything in the
same window anyway 18:48⎜<xpt:#!/bin/mksh> i know, you have some kind of
telnet with automatic pong 18:48⎜<mirabilos:#!/bin/mksh> haha, yes :D
18:49⎜<mirabilos:#!/bin/mksh> though that's more tinyirc – sirc is more comfy
