import math
def jsfile(dp, fout, title):
htmltext = """
"""
htmltext = htmltext % (dp, title)
f = open(fout, 'w')
f.write(htmltext)
f.close()
#print htmltext
def jsfile2(dp, fout, title):
htmltext = """
"""
htmltext = htmltext % (dp, title)
f = open(fout, 'w')
f.write(htmltext)
f.close()
#print htmltext
def jsfile3(fout, jschunk, htmldiv):
htmltext = """
%s
"""
htmltext = htmltext % (jschunk, htmldiv)
f = open(fout, 'w')
f.write(htmltext)
f.close()
#print htmltext
def format_weekly_data_points3(infile, outfile, title):
jstext = """
var data%s = google.visualization.arrayToDataTable(
%s
);
var options = {
title: '%s',
legend: { position: 'top', maxLines: 2 },
colors: ['#5C3292', '#1A8763', '#871B47', '#999999', '#5CFF92'],
interpolateNulls: false,
};
var chart%s = new google.visualization.Histogram(document.getElementById('chart_div%s'));
chart%s.draw(data%s, options);
"""
f = open(infile)
points_series = {}
cnt = 0
for l in f.readlines():
l = l.strip("\n").split(',')
# to limit the data points to a partcular year range, add the condition
# l[2][0:4] not in ['2014', '2015', '2013']:
if math.fabs(float(l[1])) > 0.035 or l[2][0:4] not in ['2014', '2015', '2013']:
continue
else:
if l[2][5:7] not in points_series:
points_series[l[2][5:7]]= []
if l[0] == '1':
points_series[l[2][5:7]].append( "[%s, null,null,null,null]," % l[1])
elif l[0] == '2':
points_series[l[2][5:7]].append( "[null, %s, null,null,null]," % l[1])
elif l[0] == '3':
points_series[l[2][5:7]].append( "[null,null,%s, null,null]," % l[1])
elif l[0] == '4':
points_series[l[2][5:7]].append( "[null,null,null,%s, null]," % l[1])
elif l[0] == '5':
points_series[l[2][5:7]].append( "[null,null,null,null,%s ]," % l[1])
# cnt = cnt + 1
# if cnt > 50:
# break
s = '[["M", "T", "W", "R", "F"],'
jstexts = {}
jschunk = ''
htmldiv = ''
for month, magnitudes in points_series.iteritems():
s = '[["M", "T", "W", "R", "F"],'
s = s + ''.join(e for e in magnitudes)
s = s[:len(s)-1] + ']'
jstexts[month] = jstext
jstexts[month] = jstexts[month] % (month, s, "HKCE Open/Previous day fluctuation by month:" + month, month, month, month, month)
print month, len(magnitudes) -1
jschunk = jschunk + jstexts[month]
for month in sorted(points_series.keys()):
htmldiv = htmldiv + '' % month
#print jschunk
print htmldiv
jsfile3(outfile, jschunk, htmldiv)
def format_weekly_data_points2(infile, outfile, title):
# this routine formats data into a tabular format of 5 columns
# each column represents a series of a weekday with monday starts at 1
# and tuesday starts at 2
# the value at each column is the % change in HSI on the open vs previous close
# the objective is to see whether there exists a pattern with mondays
# experience more fluctations when market opens
f = open(infile)
points = []
for l in f.readlines():
l = l.strip("\n").split(',')
if math.fabs(float(l[1])) > 0.035 or l[2][5:7] <> '10':
continue
else:
if l[0] == '1':
points.append( "[%s, null,null,null,null]," % l[1])
elif l[0] == '2':
points.append( "[null, %s, null,null,null]," % l[1])
elif l[0] == '3':
points.append( "[null,null,%s, null,null]," % l[1])
elif l[0] == '4':
points.append( "[null,null,null,%s, null]," % l[1])
elif l[0] == '5':
points.append( "[null,null,null,null,%s ]," % l[1])
s = '[["M", "T", "W", "R", "F"],'
s = s + ''.join(e for e in points)
s = s[:len(s)-1] + ']'
print len(points) -1
jsfile(s, outfile, title)
def format_data_points(infile, outfile, title):
f = open(infile)
points = []
for l in f.readlines():
l = l.strip("\n").split(',')
points.append( "['%s', %s]," % (l[0], l[1]))
s = '[["daily change", "Num Days"],'
s = s + ''.join(e for e in points)
s = s[:len(s)-1] + ']'
print s
print len(points) -1
jsfile(s, outfile, title)
if __name__ == '__main__':
#format_data_points("data/hsi-monday-change.csv", "data/hsi-monday-change.html", "monday open change magnitude")
#format_weekly_data_points2("data/hsi-weekday-change.csv", "data/hsi-weekday-change.html", "weekday open change magnitude")
#format_weekly_data_points2("data/hsi-weekday-change-over-5-years.csv", "data/hsi-weekday-change-over-5-years.html", "weekday open change magnitude")
#format_weekly_data_points3("data/hsi-weekday-change-over-5-years.csv", "data/hsi-weekday-change-over-5-years.html", "weekday open change magnitude")
format_weekly_data_points3("data/hkce-weekday-change-over-5-years.csv", "data/hkce-weekday-change-over-5-years.html", "weekday open change magnitude")