5 Mar 2008 5 Mar '08
If I do :
$ easy_install -U somepackage
The script will scan the PyPI simple page to look for an egg.
But if it fails on an url page that is starting with https, and if the user is behind a proxy that blocks it, it will block there before any egg has a chance to be downloaded.
I guess a good practice is to avoid https URLs in any metadata, I guess the user can also call easy_install with the direct egg url in that case,
but maybe easy_install could do something particular on https links to avoid
this issue, if doable ? Maybe by ordering the reading, to put https ones at the end ? (I don't know how https failures works when a proxy blocks them)
-- Tarek Ziadé | Association AfPy | www.afpy.org Blog FR | http://programmation-python.org Blog EN | http://tarekziade.wordpress.com/
Age (days ago)
Last active (days ago)