#!/usr/bin/python
"""
This script emits four linked files that visualize 5, 10, 15, and 20-year
cancer survival rates as shown in Tufte, Beautiful Evidence, p. 176.
It uses part of Joe Gregorio's sparkline plotter, adds data labels, and
sorts each result view by descending survival rate.
"""
# sparkline plotting courtesy of joe gregorio
import sys, os, re
import Image, ImageDraw, ImageFont
import StringIO
def plot_sparkline_smooth(results, args):
step = int(args.get('step', '2'))
height = int(args.get('height', '20'))
if height < 6: height = 6
(dmin, dmax) = [int(x) for x in args.get('limits', '0,100').split(',')]
im = Image.new("RGB", ((len(results)-1)*step+4 + 8, height + 8 ), 'white')
draw = ImageDraw.Draw(im)
coords = zip(range(1,len(results)*step+1, step),
[height + 3 - (y-dmin)/(float(dmax - dmin + 1)/(height-4))
for y in results])
# add data labels
for i in range(len(coords)):
x,y = coords[i]
draw.text ( (x,y-10), str(results[i]), fill="black")
draw.line(coords, fill="#000000")
del draw
f = StringIO.StringIO()
im.save(f, "PNG")
return f.getvalue()
# data from tufte, beautiful evidence, p 176
datadict = {
'Prostate' : (99,95,87,81),
'Thyroid' : (96,96,94,95),
'Testis' : (95,94,91,88),
'Melanomas' : (89,87,84,83),
'Breast' : (86,78,71,65),
"Hodgkin's" : (85,80,74,67),
'Corpus uteri,uterus' : (84,83,81,79),
'Urinary,bladder' : (82,76,70,68),
'Cervix,uteri' : (71,64,63,60),
'Larynx' : (69,57,46,38),
'Rectum' : (63,55,52,49),
'Kidney,renal pelvis' : (62,54,50,47),
'Colon' : (62,55,54,52),
"Non-Hodgkin's" : (58,46,38,34),
'Oral cavity,pharynx' : (57,44,38,33),
'Ovary' : (55,49,50,50),
'Leukemia' : (43,32,30,26),
'Brain,nervous system' : (32,29,28,26),
'Multiple myeloma' : (30,13, 7 ,5),
'Stomach' : (24,19,19,15),
'Lung,bronchus' : (15,11, 8, 6),
'Esophagus' : (14, 8, 8, 5),
'Liver,bile duct' : ( 8, 6, 8, 8),
'Pancreas' : ( 4, 3, 3, 3),
}
keys = datadict.keys()
vertMultiple = 1.5
def normalize(s):
s = re.sub("[, ]",'_',s)
return s.replace("'","")
for fname in keys:
data = datadict[fname]
mn = min(data)
mx = max(data)
diff = mx-mn
args = {'step':100,'height':diff*vertMultiple,'limits':'%d,%d' % (mn,mx)}
image_data = plot_sparkline_smooth(data, args)
f = open("%s.png" % normalize(fname),'wb')
f.write(image_data)
f.close
rates = (
( '5 year', 0),
('10 year', 1),
('15 year', 2),
('20 year', 3)
)
pagetemplate = """
Estimates of percentage survival rates
Estimates of percentage survival rates
explanation
"""
rowtemplate = """
| %s | %s |
"""
for rate in (0,1,2,3):
fname = normalize(rates[rate][0])+'.html'
print fname
links = ''
for link in (0,1,2,3):
label = rates[link][0]
href = normalize(label)+'.html'
if rate==link:
links += '%s | ' % label
else:
links += '%s | ' % ( href, label)
tabledata = ''
keys.sort(lambda x, y: cmp(datadict[y][rate], datadict[x][rate]))
for key in keys:
tabledata += rowtemplate % ( key, '
' % normalize(key) )
page = pagetemplate % (datadict,links,tabledata)
f = open(fname,'w')
f.write(page)
f.close()