Python bokeh figure quad categorical x-axis does not plot if strings present -

i have dataframe 2 columns: x_att1 , y_att1. y_att1 contains int or float. if x_att1 contains numbers string can plot y_att1. if x_att1 contains strings nothing plotted. don't know why. bug or feature? use bokeh 0.10.0

code 1 numbers string @ x_att1 (works):

from bokeh.plotting import figure, output_file, show bokeh.models import linearaxis, range1d import pandas pd  output_file("bars.html")  df = pd.dataframe() df['y_att1'] = [0.70,0.68,0.50,0.48,0.30,0.28] df['x_att1'] = ['0','4','2','7','3','1']  p = figure(title="twin y-axis bar example", x_range=df['x_att1'].values.tolist())  p.quad(bottom=0, top=df['y_att1'], left=df['x_att1'].values.tolist()     , right=df['x_att1'].values.tolist(), line_width=7, line_color='red')  p.yaxis.axis_label = 'left y axis' p.yaxis.axis_label_text_color = 'red'  p.xaxis.axis_label = 'x axis' p.xaxis.axis_label_standoff = -5  show(p) 

code 2 strings @ x_att1 (does not work):

from bokeh.plotting import figure, output_file, show bokeh.models import linearaxis, range1d import pandas pd  output_file("bars.html")  df = pd.dataframe() df['y_att1'] = [0.70,0.68,0.50,0.48,0.30,0.28] df['x_att1'] = ['b','d','a','h','f','e'] #changed line  p = figure(title="twin y-axis bar example", x_range=df['x_att1'].values.tolist())  p.quad(bottom=0, top=df['y_att1'], left=df['x_att1'].values.tolist()     , right=df['x_att1'].values.tolist(), line_width=7, line_color='red')  p.yaxis.axis_label = 'left y axis' p.yaxis.axis_label_text_color = 'red'  p.xaxis.axis_label = 'x axis' p.xaxis.axis_label_standoff = -5  show(p) 

code 3 strings @ x_att1 , trick (works):

from bokeh.plotting import figure, output_file, show bokeh.models import linearaxis, range1d import pandas pd  output_file("bars.html")  df = pd.dataframe() df['y_att1'] = [0.70,0.68,0.50,0.48,0.30,0.28] df['x_att1'] = ['b','d','a','h','f','e']  p = figure(title="twin y-axis bar example", x_range=df['x_att1'].values.tolist())  p.quad(bottom=0, top=df['y_att1'], left=df['x_att1'].values.tolist()     , right=df['x_att1'].values.tolist(), line_width=7, line_color='red')  p.yaxis.axis_label = 'left y axis' p.yaxis.axis_label_text_color = 'red'  #added these 2 lines p.y_range.start = 0 p.y_range.end = df['y_att1'].max()*1.1  p.xaxis.axis_label = 'x axis' p.xaxis.axis_label_standoff = -5  show(p) 

nice day!