Brian Dowling

Entities return pd.Timestamp for time related entries, but calls that accept time don't accept that as a valid format. So doing something like: ``` orders = api.list_orders(after='2018-01-01', status='all', direction='asc') orders.extend(...

Adding additional timeframe buckets, is a bit of a process.

1

I was fooled by the config into thinking I could just enter another entry in the list, but realized this does not actually handle the mappings from these tags. To...

RemoveTimeBucket access to delete errant data loads

5

There does not appear to be a way to access RemoveTImeBucket in the RPC API, which is probably good from a security standpoint. Perhaps there should be a CLI way...

Allow algo's to influence client_order_id for algo grouping

1

It is useful to use client_order_id to help group different algos. If there was a way to define a order_prefix for example it could be used in alpaca.Backend._new_order_id for users...

Add fetch_link to make it much easier to follow linked URLs within an API

5

This is an attempt at answering the question I had in #352 By simply adding `x-operationId: [linkedOperation]` in your spec to URLs that are returned from the API that are...

Handling next and previous pagination of results

4

I'm looking for an example of dealing with pagination in the case where the API provides a next/previous URL for you to follow. It is not clear to me if...

Initial load speed improvements, caching best practices?

1

I'm curious if there are any best practices for improving the initial load/parsing process. e.g. I was wondering if anyone has attempted to use pickle or shelve at any stages...

Support for additional account types

6

It would be great to offer additional account types, such as.. - Joint - IRAs, - Roth IRAs - Custodial Roth IRAs - Cash

Fetching from refs/pulls instead

1

I'm curious why not just pull from refs/pulls/[PR#]/head instead of trying to fetch from the original contributor's branch? This fails in cases when the remote branch was deleted or doesn't...

Pandas serialization with pyarrow

1

Would it make sense to switch the pandas serialization to use pyarrow, performance could be significant on larger datasets. https://github.com/pandas-dev/pandas/pull/28494/files#diff-03b380f521c43cf003207b0711bac67fR2527