[sane-devel] sane-backends release 1.0.26 schedule

[Louis Lagendijk]

On Sat, 2017-05-06 at 15:55 +0200, Wilhelm wrote:
> Am 06.05.2017 um 14:27 schrieb m. allan noah:
> > On Sat, May 6, 2017 at 5:52 AM, Louis Lagendijk <louis at fazant.net>
> > wrote:
> > > On Fri, 2017-04-28 at 08:04 -0400, m. allan noah wrote:
> > > > Ok folks, it's time to get another sane-backends release out
> > > > the
> > > > door.
> > > > 
> > > > Olaf has done a good job of cleaning up our contributors list
> > > > and
> > > > curating the bug tracker. However, there are a handful of
> > > > patches in
> > > > the bug tracker that could still be applied, once they are
> > > > reviewed.
> > > > Also, quite a number of backends that are now unmaintained. So,
> > > > this
> > > > is a good time to get involved with sane. If you benefit from
> > > > this
> > > > project, and have some programming experience, we could use the
> > > > help.
> > > > 
> > > > Schedule:
> > > > 
> > > > May 7: Feature freeze (only fix bugs and update docs after this
> > > > date)
> > > > May 14: Code freeze (only update docs after this date)
> > > > May 21: Release
> > > > 
> > > 
> > > Hi,
> > > Yesterday when I had a look at our bug tracker for any issues in
> > > my
> > > code I found https://alioth.debian.org/tracker/?func=detail&group
> > > _id=30
> > > 186&aid=315004&atid=410366
> > > This is an issue for scanbd integration that requires more
> > > flexibility
> > > for configuration of dll-loading: when scanbd is used users need
> > > to use
> > >  the net backend only, but scanbd/saned need to be fed with the
> > > "normal" list of backends.
> > > I made a patch to dll.c where
> > > - It used the dll.conf with the name pointed out by env. var
> > > SANE_CONFIG_FILE if defined, if not
> > > - it tries to load a dll2.conf if it exists. This is meant to be
> > > a file
> > > dropped in thre sane config dir by scanbd. If that does not exist
> > > - it follows the existing code path.
> > > 
> > > I added a #include statement in the config file so dll2.conf can
> > > include dll.conf if so required.
> > > 
> > > I am in the process of testing and cleaningup. but my question
> > > is:
> > > should I commit this change so close to the freeze date?
> > > Documentation
> > > is still to be done, but I would still have 2 weeks for that.
> > > Alan, what do you think?
> > > 
> > 
> > I don't now recall the entire discussion around the guts of
> > scanbd's
> > implementation, but you description sounds a little odd to me.
> > 
> > 1. If dll2.conf is created by scanbd, and scanbd is not running,
> > and
> > the user uses scanimage or another frontend, he will unknowingly
> > load
> > dll2.conf first. Even if dll2.conf #includes dll.conf, it is still
> > a
> > behavior change.
> > 2. Do you not also need changes to saned to make this work? That
> > seemed to be the case in the earlier discussion.
> > 3. I think I would prefer a more clear name than dll2.conf, but I
> > cannot think of one :)
> 
> From a scanbd point of view it would be suffcient to have either:
> 
> 1) an env-var e.g. SANE_CONFIG_FILE which is normally unset and all
> sane-applications use the compile-time setting. scanbd can set this
> to
> an alternative (e.g. scanbd-dll.conf) file before starting saned.
> 
> 2) give saned a -c <file> option
> 
> I would prefer 1)
> 
Well 2 is not an option in my opinion as there is no way for saned to
pass an argument to the rest of saned. That is why I implemented 1).

The change to an alternative dll.file was driven by the fact that when
packaging scanbd you want to minimize the amount of configuration to be
done by the user. If the scanbd package drops a dll2.conf file in
SANE_CONFIG_DIR there is no need for manual configuration.

I am now implenmenting a #pidfile directive for in dll2.conf that will
check for the pidfile and fail processing of dll2.conf and fallback to
dll.conf. 

The way I envisage this working is:
The package drops a dll2.conf file containing:

#pidfile=/var/run/scanbd.pid
net

so sane users see only the net backend if scanbd is active.

The default dll.conf remains unchanged, but net should be commented out

Scanbd uses dll.conf (sets SCAN_CONFIG_FILE=dll.conf)


The only configuration required from the user is:
- comment net from dll.conf if not already done
- populate net.conf with localhost

I changed the name of the alternative config file to dll-override.conf
(but remain open for better alternatives).

I hope that these changes (dll-override.conf as name + check for
pidfile) address concerns people may have so I can commit the changes
today and be in time for 1.0.26.

Comments anybody?

BR, Louis