Add NDBC simple web service #241
Conversation
siphon/simplewebservice/ndbc.py
Outdated
| super(NDBC, self).__init__('https://www.ndbc.noaa.gov/') | ||
|
|
||
| @classmethod | ||
| def station_observations(cls): |
There was a problem hiding this comment.
Left over from spelling internal debating 😄
siphon/simplewebservice/ndbc.py
Outdated
| return available_data | ||
|
|
||
| @classmethod | ||
| def raw_buoy_data(cls, buoy, type='txt'): # noqa: A002 |
There was a problem hiding this comment.
How about going with data_type instead of type so we can avoid overriding a built-in (and the flake8-plugin checking for it).
|
I've addressed the changes, switched to using pandas table parser vs fixed width, added several examples, and tied up a few loose ends here. |
examples/ndbc/buoy_met_request.py
Outdated
|
|
||
| #################################################### | ||
| # Let's make a simple time series plot to checkout what the data look like. | ||
| fig = plt.figure(figsize=(12, 10)) |
There was a problem hiding this comment.
I think I prefer this:
fig, ax = plt.subplots(3, 1, figsize=(12, 10))
ax2b = ax[2].twinx()
ax[1].plot(...)
...What do you think?
There was a problem hiding this comment.
I don't love that personally - it's CS pure, but not how a scientist would implement a plot. I often named my axes how they would be labeled in the figure (a, b, etc)
There was a problem hiding this comment.
I could be convinced, but it's minor
There was a problem hiding this comment.
If you want to label them that way, great. I have a pretty intense aversion to using numbers in variable names where an array will do—especially when you can get an array out of an API call. This is literally an example from my Pythonic plotting workshop unit.
You can still accomplish what you want with:
fig, (a, b, c) = plt.subplots(3, 1, figsize=(12, 10))How’s that read to you?
I’m not interested in how a “typical” scientist would do this—I’m interested in implementing best practice.
There was a problem hiding this comment.
Then we'll need to change how it's done elsewhere in our workshop materials :)
siphon/simplewebservice/ndbc.py
Outdated
| '3hr_pressure_tendency': 'hPa', | ||
| 'time': None} | ||
|
|
||
| df = pd.read_fwf(StringIO(content), skiprows=2, |
There was a problem hiding this comment.
Intentional that this is still using fwf?
|
What did the table parser get you? |
|
Table parser avoided the weirdness of the wind speed cutting off the tens digit unless there is 10 m/s or greater in the first 100 rows.... because the file is justified strangely. |
This simple web service provides access to the latest observations file for all buoys, as well as the realtime buoy data for 9 different data types.