Making multi-level indexing optional #26
Conversation
This modification addresses issue #25 by adding an additional parameter multi_index to some functions. When multi_index=False the output will be a dataframe with a single-level index (datetime) independently of the number of sites being queried.
dataretrieval/nwis.py
Outdated
| @@ -24,9 +24,12 @@ | |||
| WATERDATA_SERVICES = ['qwdata', 'measurements', 'peaks', 'pmcodes', 'water_use', 'ratings'] | |||
|
|
|||
|
|
|||
| def format_response(df, service=None): | |||
| def format_response(df, multi_index=None, service=None): | |||
There was a problem hiding this comment.
Parameter order for functions matter, so it is often best to put new parameters last.
There was a problem hiding this comment.
Yes, service is only used in 1 function, so it is more convinient to add it as second. If third, every call to multi_index will need to be named as service=None and it's not included most of the times. Do you think it is still better to add multi_index last?
dataretrieval/nwis.py
Outdated
| """Setup index for response from query. | ||
| """ | ||
| if multi_index: | ||
| multi_index=None |
There was a problem hiding this comment.
These two lines guarantee multi_Index will be None, it doesn't matter what value.
| @@ -71,6 +75,8 @@ def get_qwdata(datetime_index=True, wide_format=True, sites=None, start=None, en | |||
| If the qwdata parameter begin_date is supplied, it will overwrite the start parameter | |||
| end: string | |||
| If the qwdata parameter end_date is supplied, it will overwrite the end parameter | |||
| multi_index: boolean | |||
There was a problem hiding this comment.
ah, multi_index is a boolean. Set the default to a boolean instead of None. That'll clean up the code in the format_response function above.
dataretrieval/nwis.py
Outdated
| @@ -155,7 +163,7 @@ def _discharge_measurements(**kwargs): | |||
| return _read_rdb(response.text), _set_metadata(response, **kwargs) | |||
|
|
|||
|
|
|||
| def get_discharge_peaks(sites=None, start=None, end=None, **kwargs): | |||
| def get_discharge_peaks(sites=None, start=None, end=None, multi_index=None, **kwargs): | |||
There was a problem hiding this comment.
You can clean up a lot of this by updating def format_response to accept kwargs and then pop multiindex only there instead of popping it higher up and cascading through functions that only pass it along.
This modification addresses issue #25 by adding an additional parameter multi_index to some functions. When multi_index=False the output will be a dataframe with a single-level index (datetime) independently of the number of sites being queried.