[Scons-users] env.Dictionary bug? doc problem?

[Bill Deegan]

Any internal non-test impact?

On Mon, Jul 23, 2018 at 7:46 AM, Mats Wichmann <mats at wichmann.us> wrote:

> On 07/23/2018 07:51 AM, Mats Wichmann wrote:
> > On 07/22/2018 12:26 PM, Bill Deegan wrote:
> >> Can you check if any internal usage depends on this slicing behavior?
> >>
> >> -Bill
> >
> > Plenty of impacts.
> >
> > A bunch of tests are written to the idea that Dictionary(key) returns
> > key's value, not a subset of the dictionary.
>
> looks like there's also an expectation that Dictionary() returns a view,
> not a new dictionary, such that modifying something through a Dictionary
> call "sticks". See the below test code - the modification of env2
> through its Dictionary() - this will not work if Dictionary returns a
> newly constructed dict, as I had imagined.
>
>         # Ensure that lists and dictionaries are
>         # deep copied, but not instances.
>         class TestA(object):
>             pass
>         env1 = self.TestEnvironment(XXX=TestA(), YYY = [ 1, 2, 3 ],
>                            ZZZ = { 1:2, 3:4 })
>         env2=env1.Clone()
>         env2.Dictionary('YYY').append(4)
>         env2.Dictionary('ZZZ')[5] = 6
>         assert env1.Dictionary('XXX') is env2.Dictionary('XXX')
>         assert 4 in env2.Dictionary('YYY')
>         assert not 4 in env1.Dictionary('YYY')
>         assert 5 in env2.Dictionary('ZZZ')
>         assert 5 not in env1.Dictionary('ZZZ')
>
> Not really sure what to do with this.
> _______________________________________________
> Scons-users mailing list
> Scons-users at scons.org
> https://pairlist4.pair.net/mailman/listinfo/scons-users
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://pairlist4.pair.net/pipermail/scons-users/attachments/20180723/885cc01f/attachment.html>