Re: Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2013-09-05 Thread Bastien Nocera
Hey,

Em Wed, 2013-02-13 às 15:18 +, Simon McVittie escreveu:
> On 21/12/12 15:16, Bastien Nocera wrote:
> > So this is what I ended up with:
> > http://people.freedesktop.org/~hadess/idle-inhibition-spec/
> 
> > Idle inhibition is achieved by ... calling an Inhibit function ... on
> > a well-known D-Bus name.
> 
> s/function/method/ - D-Bus doesn't have functions.

Fixed.

> It seems pretty strange that the "API documentation" chapter is so vague
> - you have to read the "API reference" to know what's actually going on.
> Rename to "API overview" maybe, and/or add more hyperlinks?

Renamed.

> > API notes
> 
> "Design notes" or "Rationale"?

Renamed.

> > org.freedesktop.ScreenSaver — The Idle Inhibition Service manages the
> > inhibition requests.
> 
> There's nothing here to say that the process with the o.f.S well-known
> name must implement the o.f.S interface. There's also nothing here to
> say which object-path in that process must have that interface.

Sigh. That'd be because I didn't read the HTML output at all, other than
looking that it did generate, and the input does contain that
information:
$ git grep '

>  By
> inspecting the GNOME implementation, it appears to be
> /org/freedesktop/ScreenSaver.
> 
> I suggest re-wording to:
> 
> The Idle Inhibition service manages inhibition requests.
> Implementations of this well-known bus name must have an object
> /org/freedesktop/ScreenSaver which implements the
> _org.freedesktop.ScreenSaver_ interface.
> 
> where _..._ is a hyperlink.

I've put this in the API overview, otherwise this gets pulled in to the
TOC as the label for the link.

> Oswald wrote:
> > the spec is ridiculously over-structured for its contents.
> 
> I think it would seem less ridiculous if it was (only) published as a
> single HTML page with the same content, which I think is just a matter
> of using different XSLT? The single page / multi-page split can happen
> if it grows more methods later.

I haven't figured out how to do that, and after spending 10 minutes
trying to make it output what I wanted, I think it's quite enough work
for the handful of people that will ever be reading this.

Note that I found out (when VLC wasn't actually inhibiting playback)
that KDE doesn't implement the API on the /org/freedesktop/ScreenSaver
but on /ScreenSaver. I've made a mention of that in the spec as well.

Cheers

___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg


Re: Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2013-02-13 Thread Simon McVittie
On 21/12/12 15:16, Bastien Nocera wrote:
> So this is what I ended up with:
> http://people.freedesktop.org/~hadess/idle-inhibition-spec/

> Idle inhibition is achieved by ... calling an Inhibit function ... on
> a well-known D-Bus name.

s/function/method/ - D-Bus doesn't have functions.

It seems pretty strange that the "API documentation" chapter is so vague
- you have to read the "API reference" to know what's actually going on.
Rename to "API overview" maybe, and/or add more hyperlinks?

> API notes

"Design notes" or "Rationale"?

> org.freedesktop.ScreenSaver — The Idle Inhibition Service manages the
> inhibition requests.

There's nothing here to say that the process with the o.f.S well-known
name must implement the o.f.S interface. There's also nothing here to
say which object-path in that process must have that interface. By
inspecting the GNOME implementation, it appears to be
/org/freedesktop/ScreenSaver.

I suggest re-wording to:

The Idle Inhibition service manages inhibition requests.
Implementations of this well-known bus name must have an object
/org/freedesktop/ScreenSaver which implements the
_org.freedesktop.ScreenSaver_ interface.

where _..._ is a hyperlink.

Oswald wrote:
> the spec is ridiculously over-structured for its contents.

I think it would seem less ridiculous if it was (only) published as a
single HTML page with the same content, which I think is just a matter
of using different XSLT? The single page / multi-page split can happen
if it grows more methods later.

   S
___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg


Re: Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2013-02-13 Thread Bastien Nocera
On Wed, 2013-02-13 at 08:45 +0100, Oswald Buddenhagen wrote:
> On Tue, Feb 12, 2013 at 06:45:38PM +0100, Bastien Nocera wrote:
> > On Fri, 2012-12-21 at 16:16 +0100, Bastien Nocera wrote:
> > > Resend, as the original mail was never moderated through (the attached
> > > patch was too big).
> > 
> > And the draft is now live at:
> > http://specifications.freedesktop.org/idle-inhibit-spec/0.1/
> > 
> ch01.html
> "anothe[r]"
> "to avoid the idle" -- huh?

Fixed.

> ch04.html
> "The D-Bus service name is [what?] intentionally,"

Fixed.

> the spec is ridiculously over-structured for its contents.

It has the same structure as other specs. It would also not be so
over-structured if it included everything I thought should have been in
the API, but I guess I'm fighting a losing battle here.

> i expect an in-depth discussion of the purpose of this spec in the
> preface, specifically addressing *all* questions raised in the mails we
> exchanged with lubos.

I would have taken this in my stride 2 months ago, when I first re-sent
the draft, but now it's really too late. I'm happy to review patches,
otherwise it'll have to wait for the next re-visit.

Cheers

___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg


Re: Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2013-02-13 Thread Oswald Buddenhagen
On Tue, Feb 12, 2013 at 06:45:38PM +0100, Bastien Nocera wrote:
> On Fri, 2012-12-21 at 16:16 +0100, Bastien Nocera wrote:
> > Resend, as the original mail was never moderated through (the attached
> > patch was too big).
> 
> And the draft is now live at:
> http://specifications.freedesktop.org/idle-inhibit-spec/0.1/
> 
ch01.html
"anothe[r]"
"to avoid the idle" -- huh?

ch04.html
"The D-Bus service name is [what?] intentionally,"

the spec is ridiculously over-structured for its contents.

i expect an in-depth discussion of the purpose of this spec in the
preface, specifically addressing *all* questions raised in the mails we
exchanged with lubos.
___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg


Re: Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2013-02-12 Thread Bastien Nocera
On Fri, 2012-12-21 at 16:16 +0100, Bastien Nocera wrote:
> Resend, as the original mail was never moderated through (the attached
> patch was too big).

And the draft is now live at:
http://specifications.freedesktop.org/idle-inhibit-spec/0.1/

Thanks Vincent!

___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg


Spec draft (was Re: Rework of the org.freedesktop.ScreenSaver API?)

2012-12-21 Thread Bastien Nocera
Resend, as the original mail was never moderated through (the attached
patch was too big).

So this is what I ended up with:
http://people.freedesktop.org/~hadess/idle-inhibition-spec/

Patch for xdg-specs available at:
http://people.freedesktop.org/~hadess/0001-Add-idle-inhibit-D-Bus-service-spec.patch

Patches welcome.

___
xdg mailing list
xdg@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/xdg