Ergonomic, hermetic, user-friendly software build system for Python, Go, Java, Kotlin, Protobuf, Scala, Shell. Pants lets you fearlessly scale up your codebase!

Pants

basically, the reason we clone `Value`s explicitly on the rust side is so that every one has a unique `Handle`

and then python handles the reference counting

(cffi makes the guarantee that all handles are unique)

<http://cffi.readthedocs.io/en/latest/ref.html#ffi-new-handle-ffi-from-handle>

```New in version 1.4: two calls to new_handle(x) are guaranteed to return cdata objects with different void * values, even with the same x. This is a useful feature that avoids issues with unexpected duplicates in the following trick: if you need to keep alive the “handle” until explicitly asked to free it, but don’t have a natural Python-side place to attach it to, then the easiest is to add() it to a global set. It can later be removed from the set by global_set.discard(p), with p any cdata object whose void * value compares equal.```

I’m getting an intermittent failure of test_changed_coverage_transitive_src_python_python_targets_test_library_py in the test changed integration tests

```Exception message: Unrecognized State type `178` on: &lt;cdata 'RawNode &amp;' 0x7fa664854890&gt;```

<@U066LC126> : hm. did you change how RawNodes are constructed?

I don’t think so, which is why it’s confusing

so: the RawNodes case is different from your Value case

because RawNodes actually needs to contain a rust allocation

so it actually needs to be allocated, and then later passed back to rust to be deallocated*

I changed the LocalScheduler, it’s possible the extraction is related

I changed something from a generator to a fn call triggering the nodes_destroy early because the nodes were no longer ref’d from python

there shouldn't be many roots, so a generator is probably overkill

does python’s ref counter work at fn boundaries?