Thanks for the proposals, I've added newest=false in the SetUp, as well as ellipsis. This seems ok for the 0.6 branch, but not enough for the trunk.
Christophe Benji York a écrit : > On Mon, Aug 11, 2008 at 3:27 PM, Jim Fulton <[EMAIL PROTECTED]> wrote: >> On Aug 11, 2008, at 5:17 AM, Chris Withers wrote: >> >>> Christophe Combelles wrote: >>>> Is this change ok? >>>> It seems to solve the previous failures in the 3.4 kgs, and speeds >>>> up the test. >>> I don't have a problem with the change, but this damages the >>> usefulness of README.txt as documentation. >>> >>> As someone who's never used zope3recipes, I'm going to read that and >>> wonder why offline = true is in there all the time as there are no >>> comments or explanation. >>> >>> Myself, I'd try and find another way of getting offline=true to >>> happen so that the docs can actually remain actual docs without test >>> cruft strewn throughout them. >> >> I agree. >> >> A sneaky way to do this would be to change the setup code to set a >> HOME environment variable to a temporary directory >> containing .buildout/default.cfg with: > > I'll throw out another option: you can add doctest examples in reST > comments to enable offline mode. Something like this: > > """ > .. Set offline mode so the tests run correctly (or whatever). > > >>> DO THING HERE > > Now we foo the bar with the baz > > >>> baz.foo(bar) > """ > > When rendered to HTML (or any other target) the first part won't be > visible, but the second part will be. > > If you have to do this much a global approach (like the one Jim > suggested) would likely be better, but I wanted to demonstrate this > approach because it can be useful when you don't want to pollute > otherwise good documentation. _______________________________________________ Zope-Dev maillist - Zope-Dev@zope.org http://mail.zope.org/mailman/listinfo/zope-dev ** No cross posts or HTML encoding! ** (Related lists - http://mail.zope.org/mailman/listinfo/zope-announce http://mail.zope.org/mailman/listinfo/zope )