kicad-developers team mailing list archive

[Chris Pavlina <pavlina.chris@xxxxxxxxx>]

What about using the date the change was made as a "version number"? Can
integerize it like 20160407 for example. This allows easy cross-referencing of
a format version with the revision that it was made in, and is guaranteed to
increase monotonically if you use a YMD format :)

On Thu, Apr 07, 2016 at 07:47:10PM +0200, jp charras wrote:
> Le 07/04/2016 18:42, Chris Pavlina a écrit :
> > On Thu, Apr 07, 2016 at 06:36:40PM +0200, jp charras wrote:
> >> Le 07/04/2016 17:52, Wayne Stambaugh a écrit :
> >>> On 4/7/2016 9:47 AM, Chris Pavlina wrote:
> >>>> Hi all,
> >>>>
> >>>> I'm targeting this email primarily at Wayne as versioning and release policy is
> >>>> involved.
> >>>>
> >>>> We've got a bit of a problem right now. We're currently adding features to the
> >>>> pcbnew format - JP just merged rounded-rect pads and has a patch in development
> >>>> for custom pads, and I'm looking at a patch to add angled fields. Problem is:
> >>>
> >>> JPs rounded rectangle commit is a problem.  I did not have a chance to
> >>> review JP's patch.  I would have recommended a file version bump.
> >>
> >> It can be done now. This is not too late.
> >> Just I need to know if the new version is 5 or 4.1 (adding rounded rect pads is a minor change,
> >> because the file format does not change when rounded rect pads are not used)
> >> For me, a major change is more when new features are always stored in file, and the file can be
> >> never read by older versions.
> > 
> > The format version isn't an integer? What's the point of minor versions of
> > _file formats_? They don't mean anything to end users, they're just for
> > comparison...
> > 
> 
> Because it is not yet used, it can be what we want (what we prefer).
> 
> We can use only an integer or use (like many other version id) or a notation like x.y
> Something like
> * when x changes, the file is no more readable by older versions
> * when y changes, the file can be readable by older versions if the new feature is not used
> 
> This is just for us, poor developers (and yes, also for comparisons), not for users.
> 
> 
> This is just a decision to take (and the corresponding code to write, because it does not exist).
> 
> -- 
> Jean-Pierre CHARRAS