def get_repeating_lines(start, increment, lines, maxPercentDif=.34):
"Returns either None or a tuple for an xrange of the repeating line numbers."
head = lines[start]
count = 0
for line in islice(lines,start+increment,None,increment):
if percent_dif(head, line) < maxPercentDif:
count+=1
else:
break
if count <= 1:
return None
else:
return (start, start+increment*(count+1), increment)
def is_probable_lle(group, lines, outlierFactor=1):
"Returns true if the group is a probable error in a repetative group."
start, end, inc = group
percent_difs = []
for i in xrange(start, end, inc):
for j in xrange(i+inc, end, inc):
percent_difs.append(percent_dif(lines[i], lines[j]))
avg = mean(percent_difs)
sigma = std(percent_difs)
for dif in percent_difs:
if abs(dif-avg)>outlierFactor*sigma:
return True
return False
