Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.lang.python > #19703 > unrolled thread

Re: configobj

Back to article view | Back to comp.lang.python

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: configobj Andrea Crotti <andrea.crotti.0@gmail.com> - 2012-02-01 10:34 +0000

#19703 — Re: configobj

On 02/01/2012 12:21 AM, Terry Reedy wrote:
> On 1/31/2012 11:06 AM, Andrea Crotti wrote:
>> I have a couple of questions about configobj, which I'm happily trying
>> to use for this project.
>
> When asking about 3rd party modules, please include a url, so we can 
> be sure of what you mean and even take a look. Is
> www.voidspace.org.uk/python/configobj.html
> what you mean?

Yes I meant that sorry.

>
>> which looked less complete and more cumbersome (to me at least)
>
> Does ConfigParser have the same problems you found with ConfigObj.
>

Well no, but also because it just gets raw strings, and doesn't do any 
validation.
With ConfigObj I can do simply a spec file

skip_pesky_pyc_paths = string_list
include_extra_paths = string_list
use_real_dependencies = bool(default=False)
compute_dependencies_recursively = bool(default=False)

And the options which are not declared will default automatically to 
that value.
And also I never really liked the ConfigParser API..

Anyway I solved just leaving these long lists somewhere else, but otherwise
I think a better solution would be YAML in general (which doesn't have 
much validation either
apparently).

[toc] | [standalone]

Back to top | Article view | comp.lang.python

csiph-web