Search code examples
pythonplotlyplotly-python

Plotly Weekly & Monthly Range selector buttons not working on time series data

I am having last 1 year time series financial data. I removed weekends as no data is present.

I created plotly viz, 2w,1m range selector buttons are not showing correct time range. While YTD,6m,all showing correct data.

2w button showing 2 week forward days but there is no data, i want to see last two week and 1m data.

Reproducable Code-

# !pip install investpy
import pandas as pd
import investpy
import datetime
import plotly.graph_objects as go
import plotly.figure_factory as ff
import plotly.express as px

today = datetime.now()  #Today Datetime
today_fut = today.strftime("%Y,%m,%d") #Converting today date to string format "%Y,%m,%d"
today_fut = datetime. strptime(today_fut, '%Y,%m,%d').date() #Converting today's date to "date" format for NSE Tools library
today = today.strftime("%d/%m/%Y") #Converting today date to string format "%d/%m/%Y"
one_year= datetime.today() - timedelta(days=370) #Subrtracting todays date with 370 Days
one_year = one_year.strftime("%d/%m/%Y") #Converting into  string format "%d/%m/%Y"

df = investpy.get_index_historical_data(index="Nifty 50",country="India",from_date=str(one_year),to_date= str(today))
print("Investpy NF Dataframe",df.tail())

bnf = investpy.get_index_historical_data(index="Nifty Bank",country="India",from_date=str(one_year),to_date= str(today))
print("Investpy BNF Dataframe",bnf.tail())

fig = go.Figure(data = [ go.Scatter(x = df.index,y = df['Close'],line=dict(color = 'Steelblue',width=2),mode='lines+markers',name = 'NIFTY'),
                        go.Scatter(x = bnf.index,y = bnf['Close'],line=dict(color = 'yellowgreen',width=2),mode='lines+markers',name = 'BANK NIFTY'),
                        ])
fig.update_layout(
    title='NF and BNF',template = 'plotly_dark',xaxis_tickformat = ' %d %B (%a)<br> %Y',
    yaxis_title='NF & BNF',yaxis_tickformat= "000",yaxis_side = 'right',xaxis_title='Date',legend = dict(bgcolor = 'rgba(0,0,0,0)'))

layout = go.Layout(showlegend=True)

##https://plotly.com/python/legend/
fig.update_layout(legend=dict(
    yanchor="top",
    y=0.99,
    xanchor="left",
    x=0.01  ))
#hide weekends
fig.update_xaxes( rangeslider_visible=True, rangebreaks=[
        dict(bounds=["sat", "mon"]) ])
##https://plotly.com/python/legend/
fig.update_layout(legend=dict(
    orientation="h",
    yanchor="bottom",
    y=1.02,
    xanchor="right",
    x=1
))


config = dict({'scrollZoom': False})

fig.update_layout(
    xaxis=dict(rangeselector=dict(buttons=list([
               dict(count=14,label="2w",step="day",stepmode="todate"),
               dict(count=1,label="1m",step="month",stepmode="backward"),
               dict(count=3,label="3m",step="month",stepmode="backward"),
               dict(count=6,label="6m",step="month",stepmode="backward"),
               dict(count=1,label="YTD",step="year",stepmode="todate"),
               dict(step="all") ])),rangeslider=dict(visible=True),type="date"))
fig.update_layout(
                  xaxis_rangeselector_font_color='white',
                  xaxis_rangeselector_activecolor='red',
                  xaxis_rangeselector_bgcolor='green',
                 )
 
fig.show()
pio.write_html(fig, file='wrong_range_selecor_for-2w&1m_button.html', auto_open=True)

Pls let me know , if there is any bug in plotly range selector or I am doing something wrong in xaxis=dict(rangeselector=dict(buttons=list([xaxis=dict(rangeselector=dict(buttons=list([ code.

Snap- When I click on 2w but it show future dates, where no data is present in dataframe.

2w button range slider 1m button snap- enter image description here

Solution

  • Yes it seems like a bug with the range selector (type='date'). It occurs when setting the scatter plot mode to 'markers' or 'lines+markers' : in fact, using markers leads to the addition of an extra range added to the xaxis, extending the default date range from the input data.

    So when you click on 2w, the slider goes from that out-of-range date (in the future) back to 2w before that date, the same happens with the other ranges actually, not only 1m, but it is less noticeable for bigger ones.

    You could set a fixed range to circumvent the problem using rangeslider_range and range :

    fig.update_xaxes(
    rangeslider_visible=True,
    rangebreaks=[dict(bounds=["sat", "mon"])],
    range=[df.index[0], df.index[-1]],
    rangeslider_range=[df.index[0], df.index[-1]]
)

    ...until you click on "all". The problem is that the range selector button step='all' will always reset the layout to use the extended range.

    This does not happen when you set the scatter mode to just mode='lines'. In this case the default/'all' slider range fits the data.

    So I guess the quick fix is to remove markers, the long one is to use Dash callbacks to have a full control of what the buttons do.