def learn_receipts():
common_words = pickle.load(open(common_word_path, 'r'))
fieldnames = []
for key in common_words.keys():
fieldnames.append(key)
for token in lexer.get_lexer().lextokens_all:
fieldnames.append(token)
training_csv_file = open(training_set_path, 'w')
training_csv_writer = csv.DictWriter(training_csv_file, fieldnames=fieldnames)
labels_csv_file = open(labels_path, 'w')
labels_csv_writer = csv.writer(labels_csv_file)
good_folder = glob.glob(aldelo_path+'/LEARN/GOOD')
bad_folder = glob.glob(aldelo_path+'/LEARN/BAD')
start_learning(good_folder, 1, training_csv_writer, labels_csv_writer)
start_learning(bad_folder, 0, training_csv_writer, labels_csv_writer)
training_csv_file.close()
labels_csv_file.close()
pickle.dump(fieldnames, open(fieldnames_array_path,'wb'))
def parse_text(text):
# If you're making changes to the parser, you need to run the the code with
# SHOULD_REBUILD_PARSER=1 in order to update it.
should_rebuild_parser = int(os.getenv('SHOULD_REBUILD_PARSER', '0'))
if should_rebuild_parser:
parser = yacc.yacc()
else:
parser = yacc.yacc(debug=0, write_tables=0)
return parser.parse(text, lexer=lexer.get_lexer())
def parse_text(text):
# If you're making changes to the parser, you need to run the the code with
# SHOULD_REBUILD_PARSER=1 in order to update it.
should_rebuild_parser = int(os.getenv('SHOULD_REBUILD_PARSER', '0'))
if should_rebuild_parser:
parser = yacc.yacc()
else:
import parsetab
parser = yacc.yacc(debug=0, write_tables=0, tabmodule=parsetab)
return parser.parse(text, lexer=lexer.get_lexer())
def __init__(self, path):
self.path = path
with self.open() as receipt_file:
self.text = receipt_file.read()
self.lexer = get_lexer()
self.lexer.input(self.text)
self.tokens = []
for idx, token in enumerate(self.lexer):
# .lexpos refers to the lexer's *current* position, not the token's.
token._position = token.lexpos
token._match = self.lexer.lexmatch
token._index = idx
self.tokens.append(token)
self.by_position = {t._position: t for t in self.tokens}
def p_ordinary(p):
"""
ordinary : NUM
| CHAR
"""
p[0] = Node(content=p[1])
def get_parser():
return ply.yacc.yacc()
if __name__ == "__main__":
parser = get_parser()
latex_string = "\sqrt{2+3}"
custom_lexer = lexer.get_lexer()
cv = parser.parse(latex_string, custom_lexer) #,debug=1)
print(interpreter(cv))
while True:
try:
try:
s = raw_input()
except NameError: # Python3
s = input('spi> ')
cv_s = parser.parse(s, custom_lexer)
print(interpreter(cv_s))
except EOFError:
break
def __init__(self, formula):
lexer = get_lexer()
lexer.input(formula)
self.formula = formula
self.tokens = list(lexer)
from lexer import get_lexer
import os
import io
import glob
import operator
import pickle
lexer = get_lexer()
curr_dir = os.path.dirname(__file__)
data_path = os.path.join(curr_dir, 'Receipt data/data')
aldelo_path = os.path.join(data_path, 'Aldelo')
common_words_path = 'common_words.pkl'
dict = {}
aldelo_stores = glob.glob(aldelo_path + '/*')
store_dicts = []
for store in aldelo_stores:
store_dict = {}
receipts = glob.glob(store + '/*.txt')
for receipt in receipts:
f = io.open(receipt, 'r', encoding='utf-8')
try:
lexer.input(f.read())
except:
continue
while True:
tok = lexer.token()
def __init__(self, formula):
lexer = get_lexer()
lexer.input(formula)
self.formula = formula
self.tokens = list(lexer)
def __init__(self, string):
self.lex = lexer.get_lexer(string)
self.current = next(self.lex)
self.last_error = None
def tokenize(file):
lexer = get_lexer()
lexer.input(file.read())
return lexer
