On Friday 07 December 2001 02:45 pm, Tavis Rudd wrote:
> But Geoff suggested that it might be better to require an explicit
> arg to the start script to make damn sure it doesn't get them from
> the wrong place. I don't think it's that likely. Nonetheless the
> launcher script should loudly decla
On Friday 07 December 2001 13:34, Chuck Esterbrook wrote:
> On Friday 07 December 2001 12:43 pm, Tavis Rudd wrote:
> > The scheme we were talking about yesterday would allow the end
> > user to make the choice of whether to go for a single file or
> > split it up. To Webware and its components it
On Friday 07 December 2001 12:43 pm, Tavis Rudd wrote:
> The scheme we were talking about yesterday would allow the end user
> to make the choice of whether to go for a single file or split it up.
> To Webware and its components it would be transparent. I think
> option 2 or 3 should be the defau
On Friday 07 December 2001 13:15, Chuck Esterbrook wrote:
> On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> > The WebKit launcher would accept either a "Configs" directory, or
> > a
>
> We could also take the same route that Python and other apps follow
> when searching for files:
>
On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> The WebKit launcher would accept either a "Configs" directory, or a
We could also take the same route that Python and other apps follow
when searching for files:
- look in standard places
- WebKit/Configs
On Friday 07 December 2001 12:49 pm, Tavis Rudd wrote:
> Why not just go for only the assignment style and get rid of the old
> dictionary parsing method completely? Transitioning an app from one
> to the other is a matter of a few minutes. Keeping both around just
> makes the code more complex
On Friday 07 December 2001 11:04 am, Geoffrey Talvola wrote:
> The __init__.py was just so that the config files could be imported
> as a package. But if there are good reasons to stick with *.config,
> then obviously you can't just import the files as a package, so the
> reason for __init__.py g
On Friday 07 December 2001 11:26, Chuck Esterbrook wrote:
> On Friday 07 December 2001 11:04 am, Geoffrey Talvola wrote:
> > >The plan I was changing for Configurable was simply to do an
> > > exec, unless the first char was a {, in which case I would do
> > > an eval.
> >
> > Just make sure to ch
On Friday 07 December 2001 10:55, Chuck Esterbrook wrote:
> On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> > As far as naming conventions, these are valid Python
> > modules/packages, so .py would probably be best. But it would be
> > nice to be able to identify config files from
On Fri, Dec 07, 2001 at 10:55:20AM -0800, Chuck Esterbrook wrote:
> On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> > As far as naming conventions, these are valid Python
> > modules/packages, so .py would probably be best. ?But it would be
> > nice to be able to identify config fil
At 10:55 AM 12/7/01 -0800, Chuck Esterbrook wrote:
>On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> > As far as naming conventions, these are valid Python
> > modules/packages, so .py would probably be best. But it would be
> > nice to be able to identify config files from their fi
On Friday 07 December 2001 06:05 am, Geoffrey Talvola wrote:
> As far as naming conventions, these are valid Python
> modules/packages, so .py would probably be best. But it would be
> nice to be able to identify config files from their filename, so it
> could look like this for a multiple-file c
At 05:30 PM 12/6/01 -0800, Tavis Rudd wrote:
>I've attached a copy of the old .webkit_config_annotated file
>translated to the proposed format. It seems to work with all
>versions of Python, not just 2.1 and up. This file contains all the
>settings and is more complicated than the average user w
On Thursday 06 December 2001 15:40, Geoffrey Talvola wrote:
> At 03:34 PM 12/6/01 -0800, Tavis Rudd wrote:
> >The keyword 'class' might be a bit misleading as we're not
> > promissing 1-to-1 mappings to actual classes, but that's no big
> > deal either. In fact, we can use anything that supports
At 03:34 PM 12/6/01 -0800, Tavis Rudd wrote:
>you wrote:
> > - Why force all settings into a single file? I actually liked the
> > fact that settings for different components live in different
> > files, although I suppose I wouldn't mind the _option_ of combining
> > them into a single file. In
On Thursday 06 December 2001 15:04, you wrote:
> On Thu, Dec 06, 2001 at 03:34:29PM -0800, Tavis Rudd wrote:
> > If we just make WebKit and all other Webware components
> > """import webware_config""" to get their settings we can use a
> > single 'webware_config.py' module or 'webware_config/' pac
Geoff,
On Thursday 06 December 2001 06:32, Geoffrey Talvola wrote:
> Tavis,
>
> I'm slowly taking a look at your redesign code. The first big
> incompatibility I noticed is the configuration mechanism. My
> concerns are:
>
> - For Windows users, it's much nicer to have a specific extension
> ass
17 matches
Mail list logo
