[Numpy-discussion] Final push for NumPy 1.0.5 (I need your help!)

[Charles R Harris]

On Sat, Apr 5, 2008 at 4:10 PM, Anne Archibald <peridot.faceted at gmail.com>
wrote:

> On 05/04/2008, James Philbin <philbinj at gmail.com> wrote:
> > I've posted patches for:
> >
> >  #630: If float('123.45') works, so should numpy.float32('123.45')
> >  #581: random.set_state does not reset state of random.standard_normal
>
> Patches for #601, #622, #692, #696, #717 now in trac; I'd like to do
> something about #720 (variance of complex arrays needs docs and
> tests), but any patch for it is necessarily going to tread on the toes
> of my patch for #696 (ddof parameter to var needs tests). I'm not
> quite sure what the best way to handle this sort of thing is.
>
> More generally, my local working copy is now rater divergent from the
> upstream. What's the recommended way to deal with this? Make sure I
> have all the patches submitted to trac, then just revert to raw SVN?
> So far I've just been editing the output of svn diff down to only the
> bit that's relevant to each patch, but it's getting pretty long.
>

Do you have commit access now? You could pull a fresh copy of svn, apply
your patches one by one, and commit each time. That way you can avoid
conflicts, which can be a pain when your local copy has diverged a lot from
upstream. Lots of small commented changes are also preferable to one big
patch.

Chuck
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/numpy-discussion/attachments/20080405/7629b971/attachment.html>