[Distutils] PEP 241 draft
M.-A. Lemburg
mal@lemburg.com
Wed Mar 14 15:06:01 2001
Amos Latteier wrote:
>
> Andrew Kuchling wrote:
> > I've checked in a first draft of PEP 241. Here it is; please offer
> > comments and changes.
>
> Looks good.
Agreed :-)
> > Authors might expect to include a METADATA file of their own that
> > would be used instead of the generated file, but this will not be
> > permitted.
>
> Important point!
Can't we have arbitrary extra fields starting with e.g. 'X-MyHeader'
which are then application defined ? (Mail headers work this way.)
> > XXX are we sure RFC-822 is enough?
> > The METADATA file format is a single set of RFC-822 headers
> > parseable by the rfc822.py module. The field names listed in the
> > following section are used as the header names.
>
> I agree that if meta-data is a flat list of names and values then rfc822
> is probably fine.
Agreed.
About the format you indicate: I don't think that the quotes
around the values are necessary for RFC-822 (but could be wrong).
At least mail headers don't have quotes, so it's probably ok to
omit them.
> > Name
> >
> > The name of the package. XXX what's the set of legal characters?
>
> Why should the legal characters be restricted? OK, no newlines, but
> otherwise what do we care?
Doesn't really matter as long as it's 7-bit, e.g. ASCII or UTF-7.
> > Author (optional)
> > Author-email
>
> Should we allow more than one contact email address? For example, there
> may be multiple authors, or there may be some other relevant contact
> email addresses.
We could probably allow the value to be a comma separated list !?
> > Keywords
> >
> > A list of additional keywords to be used to assist searching
> > for this package in a larger catalog.
> >
> > XXX Keywords should probably be constrained to be from a fixed
> > list, but I don't think this can be enforced by the 'sdist'
> > command. (The list might be large, and it could only be updated
> > with new Distutils releases, which seems too inflexible.)
> >
> > Example: 'dog puppy voting election'
>
> It might be more useful to have categories like CPAN's categories
>
> http://cpan.org/modules/by-category/
>
> or Source Forge's 'Trove' categories
>
> http://sourceforge.net/softwaremap/trove_list.php
>
> of the Vaults of Parnassus's categories
>
> http://www.vex.net/parnassus/apyllo.py?a=tree
>
> Your example keywords could probably be extracted from the description
> and long description and used by search engines when indexing. I'm not
> sure keywords are needed if we have full text description and long
> description.
>
> Categories, on the other hand, help not with searching so much as
> browsing.
Why not have both ?
--
Marc-Andre Lemburg
______________________________________________________________________
Company & Consulting: http://www.egenix.com/
Python Pages: http://www.lemburg.com/python/