MapParse.py

import numpy
import re
import spacy

nlp = spacy.load('en_core_web_sm')


scene_Sentence_Map = open("sceneSentenceMap.txt", "r")
script = open("script.txt", "r")
scenes = []
sceneMap = []
for line in scene_Sentence_Map:
    mapLine = line.split(",")
    scene = mapLine[0].split("\t")
    scene = scene[0]
    scenes.append(scene)
    mapLine[0] = mapLine[0][len(scene):]
    for i in range(0, len(mapLine) ):
        mapLine[i] = mapLine[i].replace("[","")
        mapLine[i] = mapLine[i].replace("\t","")
        mapLine[i] = mapLine[i].replace(" ", "")
        mapLine[i] = mapLine[i].replace("]", "")
        mapLine[i] = mapLine[i].replace("\n", "")
    sceneMap.append(mapLine)

scenesText = [""] * len(scenes)

for i in range(len(scenes)):
	for j in range(len(sceneMap[i])):
		scenesText[i] = scenesText[i] + script.readline()

print (scenesText[0])
		
summary = open("wikiplot.txt", "r")
summaryLines = []
for line in summary:
	summaryLines.append(line)



sceneTermFreq = []
for i in range(len(scenes)):
	sceneTermFreq.append(dict())
	words = re.split("\W", scenesText[i])
	for word in words:
		if word != "":
			if word in sceneTermFreq[i]:
				sceneTermFreq[i][word] = sceneTermFreq[i][word] + 1
			else:
				sceneTermFreq[i][word] = 1
				

t1 = nlp(summaryLines[0])
t2 = nlp(scenesText[0])

#similarity = t1.similarity(t2)

#print (t1, "\n", t2, "\n", similarity)

"""

doc = nlp.make_doc(scenesText[0])
nlp.pipeline
for name, proc in nlp.pipeline:
	doc = proc(doc)
	print([(ent.text, ent.label_) for ent in doc.ents])
#for entity in t2.ents:
#	print(entity.text, entity.label_)
"""

"""
for s in sceneTermFreq[0]:
	term = nlp(s)
	for entity in term.ents:
		print(entity.text, entity.label_)	
		
		"""