From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <e@80x24.org>
X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on dcvr.yhbt.net
X-Spam-Level: 
X-Spam-ASN:  
X-Spam-Status: No, score=-4.0 required=3.0 tests=ALL_TRUSTED,BAYES_00
	shortcircuit=no autolearn=ham autolearn_force=no version=3.4.2
Received: from localhost (dcvr.yhbt.net [127.0.0.1])
	by dcvr.yhbt.net (Postfix) with ESMTP id AA9E71F5AE;
	Tue, 20 Jul 2021 21:07:24 +0000 (UTC)
Date: Tue, 20 Jul 2021 21:07:24 +0000
From: Eric Wong <e@80x24.org>
To: Konstantin Ryabitsev <konstantin@linuxfoundation.org>
Cc: meta@public-inbox.org
Subject: Re: Restarting daemons on config file change
Message-ID: <20210720210724.GA25323@dcvr>
References: <20210719200318.wo4ydjonc3u3ibtu@nitro.local>
 <20210719204935.GA31835@dcvr>
 <20210720170019.p5d23tjb4v2mrekn@nitro.local>
 <20210720203433.GA28287@dcvr>
 <20210720204925.7xkk6ktihohxnfo3@nitro.local>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
In-Reply-To: <20210720204925.7xkk6ktihohxnfo3@nitro.local>
List-Id: <meta.public-inbox.org>

Konstantin Ryabitsev <konstantin@linuxfoundation.org> wrote:
> On Tue, Jul 20, 2021 at 08:34:33PM +0000, Eric Wong wrote:
> > > Okay, let me see what I can come up with. Looks like the best course of action
> > > is to:
> > > 
> > > 1. use a global blocking lock
> > > 2. copy the config file to a new location
> > > 3. make the necessary changes to the temporary config file
> > > 4. move the temp config file to overwrite the current config
> > > 5. release the lock
> > 
> > public-inbox-init already does all of that.  It doesn't set
> > all possible keys, yet (e.g. infourl, maybe some others).
> 
> I figured as much, but we do want to set extra keys *and* write the config in
> a certain order (e.g. prioritizing some sources over others by listing them
> first). I currently do this via a list-id globbing match (e.g.
> --listid-priority=*.linux.dev,*.kernel.org,*).

Ah, that gets tricky, since git-config doesn't seem to provide a
mechanism for ordering.  Not sure, perhaps serializing -init
invocations somehow is the way to go?  I'm not sure how to
expose that ordering, though...

> Apropos, question -- will public-inbox-init do anything to the config file if
> it finds that all the settings are already in place, or will it just accept
> them?

It only checks conflicting addresses, at the moment.