如何使用之间将字符串值转换为DateTimeIndex以进行熊猫范围选择？

Question

I have a pandas read excel method from where I get a dataframe and the parse_dates parameter is false by default. For visualization with Dash, there are two drop downs for start time and end time and I need to filter the dataframe within this time range based on the value selected in the drop down, but somehow I am stuck here and it gives an exception

"TypeError: Index must be DatetimeIndex"

df = pd.read_excel(datasource)

html.Div([
        html.Label("Please select a day", style={'textAlign': "center"}),
        dcc.Dropdown(
            id='daydropdown',
            options=[{'label': i, 'value': i} for i in dropdownDate],
            multi=False
        ),
        html.Label("Please select a start time", style={'textAlign': "center"}),
        dcc.Dropdown(
            id='starttimedropdown',
            options=[{'label': i, 'value': i} for i in dropdownStartTime],
            multi=False
        ),
        html.Label("Please select an end time", style={'textAlign': "center"}),
        dcc.Dropdown(
            id='endtimedropdown',
            options=[{'label': i, 'value': i} for i in dropdownEndTime],
            multi=False
        ),
])


@app.callback(
    Output('availability graph', 'figure'),
    [Input('daydropdown', 'value'),
     Input('starttimedropdown', 'value'),
     Input('endtimedropdown', 'value'),
     Input('vehicletype', 'value')]
)
def update_graph(selectedday, selectedstarttime, selectedendtime, selectedvehicletype):
    # get all the parking spaces for the selected day and for the selected vehicle type
    df1 = df[(df["Datum"] == selectedday) & (df["Klassifikation"] == selectedvehicletype)]
    # filter parking spaces by the time range selected
    start = pd.to_datetime(selectedstarttime)
    end = pd.to_datetime(selectedendtime)
    df2 = df1[df1['Uhrzeit'].between_time(start, end)]

I expect the df2 to be filtered according to the selected start and end time, but the exception is thrown in this line, df2 = df1[df1['Uhrzeit'].between_time(start, end)]

The format of the selected time is HH:MM:SS

Answer 1

Your format for Uhrzeit is not a datetime but just an object (it misses the date). Even so formatting the time strings selectedstarttime and selectedendtime is only possible because pandas put the date of today in front of the time section:

pd.to_datetime('08:22:22') = Timestamp('2019-08-06 08:22:22')

If you want to use between_time make sure that the object you are checking is a datetime object. The arguments in the function can be written as HH:MM, and dont have to be datetime:

df2['Urhzeit'].between_time('0:15', '0:45')