Mahmoud Hashemi

tracked it down and the big one coming from face was importing hashlib + socket builtin modules in boltons iterutils.

bumped from 20.7.0 until there's a need for a datastructure with booleans built from M expressions

This looks like it folds into a "breadcrumbs" protocol, probably not a blocker for 20.7.0. The data trace will have to suffice for now.

In theory we could allow [Merge()](https://glom.readthedocs.io/en/latest/grouping.html#glom.Merge) to update an existing dict.

I think either 1 or 2 are probably most realistic imo, esp if it's going to affect the existing tuple behavior.

Hi @Granitosaurus, usually in cases like this, I like to ask for the intended output. But thanks to your fairly thorough illustration, I assume the intention here is a multi-delete?...

Great! Looking forward to the prototype. I've always suspected putting a return type on a glom call would be quite challenging, but I'm very open to pleasant surprises :)

Just saw the PR, thanks, @sobolevn! Really clarifies some things. @kurtbrose Yeah I can see that. Sounds like another vote for `glompile()`! (to be clear, @sobolevn, Kurt and I have...

Stopping by to note that [Python's PEP 505](https://www.python.org/dev/peps/pep-0505/) on this got deferred. To me that's notable. Glom has an easy way of getting a default out of a long line...

Hey Brian! Glad you're enjoying glom :) The scenario you describe is a common one and one we definitely want to highlight. In general, we try to make glom as...