def post_processing()

in data_parsing.py [0:0]

• 27 lines of code
• 2 McCabe index (conditional complexity)

def post_processing(senses):
	quotes = []
	examples = []

	#for each word, pos pair, generate idxs
	word_keys = set([generate_word_key(s) for s in senses])
	word_idxs = {}
	for w in word_keys:
		word_idxs[w] = 0

	#pull out examples, quotations
	for s in senses:
		#generate sense id
		word_k = generate_word_key(s)
		s_id = '{}.{}'.format(word_k, word_idxs[word_k])
		word_idxs[word_k] += 1

		#label with sense_id
		s['sense_id'] = s_id
		#pull out any quotes
		if len(s['quotations']) > 0:
			for x, attrib in s['quotations']:
				sent = clean_text(x, match_sense=s_id)
				if len(sent) > 0:
					q = (sent, s_id, attrib)
					quotes.append(q)

		#pull out any examples
		if len(s['examples']) > 0:
			for x in s['examples']:
				sent = clean_text(x, match_sense=s_id)
				if len(sent) > 0:
					e = (sent, s_id)
					examples.append(e) 
		#remove quotes, examples lists from sense obj
		s.pop('quotations')
		s.pop('examples')

	return senses, quotes, examples