[Python-ideas] The stdlib++ user experience

[Donald Stufft]

> On Sep 19, 2014, at 11:39 AM, Andrew Barnert <abarnert at yahoo.com.dmarc.invalid> wrote:
> 
> On Sep 19, 2014, at 5:42, Paul Moore <p.f.moore at gmail.com> wrote:
> 
>> Maven may be a central repository - I've never used it myself as the
>> complexity has always scared me off (you could say that about most of
>> Java, though ;-))
> 
> Don't you just go to the authority factory, ask it to give you an authority, then ask the authority for repository recommender factor, and so on through 8 more steps until you get a general package object that you can cast to the type you need? And the beauty is that, thanks to the magic of static type checking with a rigid but weak type system, if it turns out not to be able to do what you want, instead of a TypeError: BeautifulSoup has no SGML parser that you've never seen before? you get the same convenient NullPointerException from your ISGMLParser as in every other bug you've had so far. :)
> 
> But seriously, I think it is worth looking at what other languages' package systems have that we might want to be jealous of as far as discovery, especially those (Cargo, go pkg, Node, gems, CocoaPods, etc.) that were designed after CPAN and the Cheeseshop and the related communities existed. I think Java, C++, and C may be the _least_ relevant places to look (although I could definitely be wrong about that).
> 
> For example, some of the newer languages' systems make it easier for you to fork an existing package, and to tie packaging to DVCS repos, with the consequence that if abarnert/spamify falls by the wayside, it might be more likely pmoore/spamify that replaces it than in Python, where it would more likely be a punny PIL->Pillow type of thing. That seems like a nice feature. Do we want that? Can we get from here to there without radical changes?
> 
> _______________________________________________
> Python-ideas mailing list
> Python-ideas at python.org
> https://mail.python.org/mailman/listinfo/python-ideas
> Code of Conduct: http://python.org/psf/codeofconduct/


I think at the point we’re well into the territory of what should move to distutils-sig.

---
Donald Stufft
PGP: 7C6B 7C5D 5E2B 6356 A926 F04F 6E3C BCE9 3372 DCFA

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/python-ideas/attachments/20140919/fd22fc78/attachment.html>