jglynn / segmund Goto Github PK

1. When running on IBM Cloud perhaps we should run on waitress or something similar?

2. Force HTTPS

Strava API limitations will result in small window we'll have access to the data -- consider persisting the results instead.

Current rate limit:
Rate Limits (?) | 100 requests every 15 minutes, 1000 daily

Need to determine interval/event to fetch data from Strava. Could start off as manual and then evolve to a scheduled job.

eg.
Unable to set attribute average_grade_adjusted_speed on entity <Split split=3 distance=707.60 m elapsed_time=0:08:02

Quiet down Stravalib, we don't care bout those attributes.

Replace strava.py with common strava lib (even though most look old and unmanaged)
Improve refresh token usage

Only include riders results who have completed The HOP segment
Only include rides posted for the given date

Use the segmund system token instead of user john token for requests to Strava API that are not made on behalf of a user

https://knanne.github.io/notebooks/visualize_strava_data_in_python.html
https://altair-viz.github.io/getting_started/starting.html
https://pandas.pydata.org/pandas-docs/stable/getting_started/dsintro.html

Inspiration? segment-challenge-dml

Following that model and tweaking for our own use -- it could look something like the following (assuming we want to design something simple that also works for challenges other than the HOP)

• Challenges (ie. The Overall HOP Challenge)
• Stages (ie. 6/4 HOP)
• Stage_Segments (ie. Segments for Given stage: Bren Hill, etc. plus a Mystery Segment)
• Stage_Segment_Efforts (ie. User Segment Effort data)

We should persist the new one if a user re-registers. If the API generates new tokens, we should use those ones.

Originally posted by @rocconnick in #10