def test_analyzer():
"""
Analyzer tests
"""
language_data = text_analyzer.language_data_loader(GRAMMAR_PATH,
COUNTER_GRAMMAR_PATH,
START_WORDS_PATH,
STOP_WORDS_PATH)
message = "This is a new medicine for hyperthyroidism"
# \
# "and helps fight against diabetes, cancer + heart disease. " \
# "https://t.co/sw6mvsslg"
analysis = text_analyzer.analyzer(message, language_data['start_words'],
language_data['grammar'],
language_data['counter_grammar'],
language_data['stop_words'],
language_data['magic_bullet_grammar'])
assert analysis[1] == 'hyperthyroidism'
assert analysis[0] == 'a new medicine'
message = "some unrelated message that only talks about watching tv"
analysis = text_analyzer.analyzer(message, language_data['start_words'],
language_data['grammar'],
language_data['counter_grammar'],
language_data['stop_words'],
language_data['magic_bullet_grammar'])
assert analysis[0] == '<nothing_found>'
message = "#hyperthyroidism for hyperthyroidism"
analysis = text_analyzer.analyzer(message, language_data['start_words'],
language_data['grammar'],
language_data['counter_grammar'],
language_data['stop_words'],
language_data['magic_bullet_grammar'])
assert analysis[0] == '<nothing_found>'
assert analysis[1] == 'hyperthyroidism'
def test_counter_analyzer():
"""
Counter analyzer tests
"""
language_data = text_analyzer.language_data_loader(GRAMMAR_PATH,
COUNTER_GRAMMAR_PATH,
START_WORDS_PATH,
STOP_WORDS_PATH)
message = "A new medicine for obesity"
analysis = text_analyzer.counter_analyzer(message, 'obesity',
language_data['counter_grammar'])
assert analysis is False
def test_language_data_loader():
"""
Language data loader tests
"""
language_data = text_analyzer.language_data_loader(GRAMMAR_PATH,
COUNTER_GRAMMAR_PATH,
START_WORDS_PATH,
STOP_WORDS_PATH)
assert 'eczema' in language_data['start_words'].keys()
assert '^@\\w+$' in language_data['stop_words']
assert '[s] \\w+ed to (healthier|better)( \\S+){0,7} [p]' in language_data[
'grammar']
assert 'chances for ( \S+){0,5} [p]' in language_data['counter_grammar']
assert r'[s] effective( \w+){0,2} (in|for|to)( \w+){0,5} [p]' in language_data[
'grammar']
from datetime import datetime
from analyzer.engines import user_analyzer
from analyzer.engines import text_analyzer
## Initialization ##
# User analysis
DICTIONARY = user_analyzer.dictionary_parser(
'./language_data/user_dictionary.txt')
LEXICON = user_analyzer.lexicon_generator('./language_data/user_grammar.txt',
DICTIONARY)
STRING_TWITTER_QUERIES = user_analyzer.string_twitter_queriesParser(
'./language_data/string_twitter_queries.txt')
# Text analysis
LANGUAGE_DATA = text_analyzer.language_data_loader(
'./language_data/grammar.txt', './language_data/counter_grammar.txt',
'./language_data/start_words.txt', './language_data/stop_words.txt')
def nlp_analysis(job_json):
"""
It takes a job as an input and returns an analysis.
"""
analysis = dict()
# Get 'profile' and 'health_related'
user_analysis = user_analyzer.user_analyzer(job_json['user_name'],
job_json['user_description'],
STRING_TWITTER_QUERIES,
LEXICON)
analysis['profile'] = user_analysis[1]