wangkev

I would like to follow the same vanilla `beancount` patterns: `identify` `extract` `archive` and `generate` `test`. The benefit of the pattern is that there is determinism in the process. With...

Ultimately, I would like to use `beancount-import` normally through the web UI. The question is more to do with how to marry the two similar, but also sufficiently different, workflows...

There are a few use cases. For example, to implement an adder or shifter function. ```python def add(field: pd.Series, n: int): return field + n ``` ```python def shift(field: pd.Series,...

Yes, exactly. I have settled on something like the former you mentioned for now, just to have something that works. The preferred, imo, is definitely the latter, which I find...

Absolutely - happy to test out and give feedback! I also kind of like explicitly specifying a node vs. a literal. Although, I do find the expression API of `polars`...

Hi -- thanks for the update, this is neat! The overall functionality seems to be exactly what we discussed. I have a few thoughts from playing around: 1. `parameterize_values` and...

> Re: (1) the idea was to have it be a more speciifc API, but its hard to say it shouldn't accept them? E.G. why would it not allows upstream...

Another option is field vs value.

The use case to write the business logic once and have different input/outputs types (e.g. `dict` or `pd.Series`). This can be achieved quite easily with `numpy` broadcasting. For example, the...

The branch fixes the issue when all functions have the same signature. > This still means downstream nodes need to have the same type signature -- this doesn't change any...