def quiz():
video_id = request.args.get('v')
# get summary
summary = summarizer.get_summary(
str(redis_utilities.get_dict(common.summary_dictionary, video_id)))
# Add space after full stop
rx = r"\.(?=\S)"
summary = re.sub(rx, ". ", summary)
re.sub(".", ". ", summary)
json_sentences = json.loads(
ibm_speech_to_text.get_timestamped_data(video_id))
prevJumpToTime = 0
questions = question_generator.generate_quiz(summary, video_id)
for question in questions:
correct_index = question['correctIndex']
correct_answer = question['answers'][correct_index]
for json_sentence in json_sentences:
(json_sentence_time) = utilities.get_sec(json_sentence['time'])
if json_sentence_time >= prevJumpToTime and json_sentence[
'line'].find(correct_answer) >= 0:
question['jumpToTime'] = json_sentence_time
prevJumpToTime = json_sentence_time
break
print(json.dumps(questions))
redis_utilities.add_json(video_id, json.dumps(questions))
def updateVariable(self):
for i, word in enumerate(self.words[:-1]):
if word == 'variable': #Find the reactor variables, tempanpassung and rotation
self.init_vars_dict[self.words[i + 1]] = self.words[i + 3]
elif word[-1] == ':': #Find the initial variables, TMGa_1 etc.
if word[0] == 'T' or word[0:2] == 'Cp':
word = word.replace(':', '')
var = word
self.initial_flows.append(var)
# After that point, time and changing variables on that time are found
elif isTimeFormat(word):
self.change_times.append(get_sec(word))
elif word == 'to' and self.words[i + 1].isdigit():
var = self.words[i - 1]
self.gas_flows.append(var)
elif word == '=':
'''When iterating valves, since the Martins input files give some weird valves (e.g. for, Prepare)
if '.' structure is added.'''
if self.words[i + 1] == 'open' or self.words[i + 1] == 'close':
if '.' in self.words[i - 1]:
var = self.words[i - 1]
self.valves.append(var)
elif self.words[i - 1].startswith('Reactor'):
self.reactor_variables.append(self.words[i - 1])
elif word == 'ReactorTemp' or word == 'ReactorPress':
self.reactor_variables.append(word)
self.reactor_variables = self.reactor_variables + [
gas for gas in set(self.gas_flows)
if not ('.' in gas or 'Run' in gas or 'Push' in gas)
]
self.gas_flows = list(
sorted(set(self.gas_flows) - set(self.reactor_variables)))
self.valves = list(sorted(set(self.valves)))
self.initial_flows = list(sorted(set(self.initial_flows)))
self.reactor_variables = list(sorted(set(self.reactor_variables)))
try:
del self.reactor_variables['up']
except:
KeyError
# for gas in gas_temp:
# #Create reactor variables list
# if '.' in gas or 'Run' in gas:
# continue
# else:
# reactor_variables_temp.add(gas)
self.change_times.insert(0, 0)
# print(len(set(gas_flows)-set(reactor_variables)))
#Subtract common elements from gas lists
return self.gas_flows, self.valves, self.initial_flows, self.reactor_variables, \
self.valves, self.initial_flows, self.change_times, self.init_vars_dict
def notes():
'''video_id = request.args.get('v')
if not video_id:
return {"Issue":"Please pass video ID v in query params"}
text_file_path = text_folder_path+video_id+'.txt'
json_file_path = json_folder_path+video_id+'.json'
url = 'https://www.youtube.com/watch?v=' + video_id'''
v = ''
if video_id not in common.summary_dictionary:
utilities.transcribe(v, video_id)
'''if not utilities.check_if_video_exists(video_id):
utilities.get_vtt_file(url)
vtt_to_txt.convert_to_text(video_id)
'''
text = str(redis_utilities.get_dict(common.summary_dictionary, video_id))
notes = summarizer.get_summary_with_ratio(text, 0.5)
summary = summarizer.get_summary_with_word_count(text, 100)
#result_list = get_key_phrases(notes, 'en')
#global_keywords = []
prevJumpToTime = 0
json_sentences = json.loads(
ibm_speech_to_text.get_timestamped_data(video_id))
sentence = str(notes[out_index])
note = {}
note['sentence'] = sentence
#note['keywords'] = keyword_list
print(keyword_list)
# need to improve
if len(keyword_list) > 0:
keyword = keyword_list[0]['keyword']
for json_sentence in json_sentences:
json_sentence_time = utilities.get_sec(json_sentence['time'])
if json_sentence_time >= prevJumpToTime and json_sentence[
'line'].find(keyword) >= 0:
note['jumpToTime'] = json_sentence_time
prevJumpToTime = json_sentence_time
break
else:
note['jumpToTime'] = prevJumpToTime
response = {"notes": notes, "summary": summary}
return json.dumps(response)
def home():
video_id = request.args.get('v')
if not video_id:
return {"Issue": "Please pass video ID v in query params"}
url = 'https://www.youtube.com/watch?v=' + video_id
#get audio
# video_id = utilities.get_audio_file(url)
text_file_path = text_folder_path + video_id + '.txt'
json_file_path = json_folder_path + video_id + '.json'
if not utilities.check_if_video_exists(video_id):
utilities.get_vtt_file(url)
vtt_to_txt.convert_to_text(video_id)
# get summary
summary = summarizer.get_summary(open(text_file_path, 'r').read())
# Add space after full stop
rx = r"\.(?=\S)"
summary = re.sub(rx, ". ", summary)
re.sub(".", ". ", summary)
# get questions and return
json_sentences = json.loads(open(json_file_path, 'r').read())
questions = question_generator.generate_trivia(summary)
prevJumpToTime = 0
for question in questions:
correct_index = question['correctIndex']
correct_answer = question['answers'][correct_index]
for json_sentence in json_sentences:
json_sentence_time = utilities.get_sec(json_sentence['time'])
if json_sentence_time >= prevJumpToTime and json_sentence[
'line'].find(correct_answer) >= 0:
question['jumpToTime'] = json_sentence_time
prevJumpToTime = json_sentence_time
break
# print(questions)
return json.dumps(questions)
def notes():
video_id = request.args.get('v')
if not video_id:
return {"Issue": "Please pass video ID v in query params"}
text_file_path = text_folder_path + video_id + '.txt'
json_file_path = json_folder_path + video_id + '.json'
url = 'https://www.youtube.com/watch?v=' + video_id
if not utilities.check_if_video_exists(video_id):
utilities.get_vtt_file(url)
vtt_to_txt.convert_to_text(video_id)
text = open(text_file_path, 'r').read()
notes = summarizer.get_summary_with_ratio(text, 0.5)
summary = summarizer.get_summary_with_word_count(text, 100)
result_list = get_key_phrases(notes, 'en')
global_keywords = []
prevJumpToTime = 0
for out_index, cur_item in enumerate(result_list):
keyword_list = []
for current_phrase in cur_item['KeyPhrases']:
# if current_phrase['Score'] >= target_phrase['Score'] and word_count(current_phrase['Text']) < word_count(target_phrase['Text']):
if utilities.isGoodScore(current_phrase):
keyword_list.append(current_phrase['Text'].strip())
if len(keyword_list) > 0:
keyword_list = utilities.filter_nouns(keyword_list, 'en')
# remove similar words
for keyword in keyword_list:
if keyword not in global_keywords:
global_keywords.append(keyword.strip())
else:
keyword_list.remove(keyword)
keyword_list = [x for x in keyword_list if x in keyword_list]
for keyword in keyword_list:
try:
print(keyword)
syns = wordnet.synsets(keyword)
if len(syns) > 0:
keyword_obj = {}
keyword_obj['keyword'] = keyword
keyword_obj['definition'] = syns[0].definition()
keyword_obj['examples'] = syns[0].examples()
keyword_list[keyword_list.index(keyword)] = keyword_obj
except:
print('error', keyword)
keyword_list = [x for x in keyword_list if type(x) is not str]
json_sentences = json.loads(open(json_file_path, 'r').read())
sentence = str(notes[out_index])
note = {}
note['sentence'] = sentence
note['keywords'] = keyword_list
print(keyword_list)
# need to improve
if len(keyword_list) > 0:
keyword = keyword_list[0]['keyword']
for json_sentence in json_sentences:
json_sentence_time = utilities.get_sec(json_sentence['time'])
if json_sentence_time >= prevJumpToTime and json_sentence[
'line'].find(keyword) >= 0:
note['jumpToTime'] = json_sentence_time
prevJumpToTime = json_sentence_time
break
else:
note['jumpToTime'] = prevJumpToTime
notes[out_index] = note
response = {"notes": notes, "summary": summary}
return json.dumps(response)