processGoldAlignment.py

import re
import sentenceSceneMap


summaryScriptAlignment = dict()

def openFiles(path):
	scriptProcess = sentenceSceneMap.ScriptSentenceMapExtractor(path)
	sentSceneMap = scriptProcess.extractSentenceMap()

	alignmentFile = open(path + 'al_wiki_script.txt', 'r')
	sceneSentenceMap = open(path + 'sceneSentenceMap.txt', 'r')
	return alignmentFile, sceneSentenceMap, sentSceneMap

def addScriptSentenceNum(summarySentenceNum, scriptSentenceNum, sentSceneMap):
	if not (summarySentenceNum in summaryScriptAlignment):
		summaryScriptAlignment[summarySentenceNum] = set()
	summaryScriptAlignment[summarySentenceNum].add(scriptSentenceNum)
	

def processAlignment(path):
	alignmentFile, sceneSentenceMap, sentSceneMap = openFiles(path)
	for line in alignmentFile:
		aligned = re.search('\t([0-9]*)', line)
		summarySentenceNum = re.search('([0-9]*)/p_', line)
		scriptSentenceNum = re.search('-([0-9]*)/', line)
	
		if aligned and aligned.group(1) != '0':
			addScriptSentenceNum(int(summarySentenceNum.group(1)), scriptSentenceNum.group(1), sentSceneMap)

	"""
	alignment = []
	count = 0
	for key in summaryScriptAlignment:
		alignment.append([])
		for element in summaryScriptAlignment[key]:
			alignment[count].append(element)
		count += 1
	"""
	

	#print(summaryScriptAlignment)
	return summaryScriptAlignment, sentSceneMap
	#print(alignment)