Parse Tohoku Univ ecei local page's weight table

parseWeightTable.py

# -*- coding: utf-8 -*-

import lxml.html
import sys

def getText(element):
	# this space is Zenkaku-space
	return element.text_content().encode('utf-8').strip().replace('　','')

# localpages html (weighttable1x.html)
f = open(sys.argv[1])
lines = f.read()
f.close()

root = lxml.html.fromstring(lines)
tables = root.cssselect('table')

# only numbers
tableHeader=[]
moduleWeight={}
meanWeight={}
labName=[]
name = ''

for table in [tb for i, tb in enumerate(tables) if i < len(tables) - 1]:
	for i, tr in enumerate(table.cssselect('tr')):
		if i == 0 or len(tableHeader) == 0:
			tableHeader = [getText(text) for text in tr.cssselect('td')]
		else:
			if i % 2 == 1:
				name = getText(tr.cssselect('td')[0])
				labName.append(name)
				moduleWeight[name] = [getText(text) for i, text in enumerate(tr.cssselect('td')) if i != 0]
			else:
				meanWeight[name] =  [getText(text) for text in tr.cssselect('td')]

'''
for name in labName:
	print moduleWeight[name]
	print meanWeight[name]
'''

# subjects, must solve II and Ⅱ, I and Ⅰ...
subjectList = tables[len(tables) - 1]
subjectHeader = []
subjectSet = ''
subjectNameAndNumber = {}
subjectSetList = []

for i, tr in enumerate(subjectList.cssselect('tr')):
	if i == 0:
		subjectHeader = [getText(text) for text in tr.cssselect('td')]
	else:
		subjectSet = getText(tr.cssselect('td')[0])
		subjectSetList.append(subjectSet)
		subjectNameAndNumber[subjectSet] = [text.strip() for text in getText(tr.cssselect('td')[1]).split()]
'''
for subjectSetName in subjectSetList:
	print subjectNameAndNumber[subjectSetName]
'''