class Group(object):
# Represent a group of data
# Groups can be compared to eachother
# Groups try to combine common data stores in more than 1 data file
_headings = []
_headingsType = []
_currRow = 0
def __init__(self, name, fileTypes):
self._name = name
self._fileTypes = fileTypes
# @param filenameList: List of filenames that contains data.
# @summary:
# Uses the file parsers to parse each data file and combine common data
def createGroup(self, filenameList):
# Get all the possible headers in the files
for filename in filenameList:
if filename.endswith('.csv'):
self._parser = CSVParser(filename)
self._parser.parse()
for heading in self._parser.getHeadings():
if self._headings.count(heading) == 0:
self._headings.append(heading)
self._headingsType.append(self._parser.getHeadingType(heading))
db = SQL("C:\\PerGraph\\" + self._name + ".db")
db.connect()
db.createTable("data",self._headings, self._headingsType)
db.commit()
db.close()
def sort(self, heading):
print 'BEFORE'
for item in self._allData:
print item
try:
idx = self._headings.index(heading)
self._allData.sort(lambda x, y: cmp(x[idx], y[idx]))
except ValueError:
print 'Header %s does not exist' % heading
print 'AFTER'
for item in self._allData:
print item
def createGroup(self, filenameList):
# Get all the possible headers in the files
for filename in filenameList:
if filename.endswith('.csv'):
self._parser = CSVParser(filename)
self._parser.parse()
for heading in self._parser.getHeadings():
if self._headings.count(heading) == 0:
self._headings.append(heading)
self._headingsType.append(self._parser.getHeadingType(heading))
db = SQL("C:\\PerGraph\\" + self._name + ".db")
db.connect()
db.createTable("data",self._headings, self._headingsType)
db.commit()
db.close()
path_csv = './csvs/words_gre.csv'
path_sname = 'wgre2'
path_addb = './{}.addb'.format(path_sname)
path_wf = './{}.wf'.format(path_sname)
num_gwp = 1
if __name__ == '__main__':
addb = load_addb(path_addb)
if not addb: addb = AdapDictDB()
iqueue = mp.Queue(maxsize=3000)
oqueue = mp.Queue(maxsize=3000)
mlock = mp.Lock()
time_parsecsv = time.time()
csvparser = CSVParser(path_csv)
cands_word = csvparser.process_file()
time_parsecsv = time.time() - time_parsecsv
print("{:.3f} seconds passed for parsing {} words from {}".format(
time_parsecsv, len(cands_word), path_csv))
parsers = list()
for i in range(num_gwp):
parsers.append(GetWordProcess(mlock, iqueue, oqueue))
for i in range(num_gwp):
parsers[i].start()
time_getwords = time.time()
words_in_addb = addb.words.keys()
words_failed = list()
def parse_csv(self, name_csv='default'):
path_csv = './csvs/{}.csv'.format(name_csv)
time_parsecsv = time.time()
csvparser = CSVParser(path_csv)
cands_word = csvparser.process_file()
time_parsecsv = time.time() - time_parsecsv
print("{:.3f} seconds passed for parsing {} words from {}".format(
time_parsecsv, len(cands_word), path_csv))
parsers = list()
for i in range(self.num_gwp):
parsers.append(GetWordProcess(self.mlock, self.iqueue,
self.oqueue))
for i in range(self.num_gwp):
parsers[i].start()
time_getwords = time.time()
words_failed = list()
num_finished_words = 0
num_words_existed = 0
num_test = 0
num_twords = len(cands_word.keys())
num_to_print = max(int(num_twords / 50), 1)
num_words_per_iteration = 100
cws = list(cands_word.keys())
num_finished = 0
while num_finished < self.num_gwp:
if len(cws) > 0 and self.oqueue.qsize() < num_words_per_iteration:
for cw in cws[:num_words_per_iteration]:
if not self.addb.exist_word(cw): self.iqueue.put(cw)
else: num_words_existed += 1
# cws.remove(cw)
time.sleep(0.001)
cws = cws[num_words_per_iteration:]
if len(cws) == 0 and self.iqueue.qsize() == 0:
for i in range(len(parsers)):
self.iqueue.put('FINISHED')
if self.oqueue.qsize() > 0:
# num_words_got = 0
rw = self.oqueue.get()
if rw:
if isinstance(rw, list) and rw[0] == 'FINISHED':
num_finished += 1
words_failed += rw[1]
else:
try:
if not self.addb.add_word(rw):
if rw.name not in cands_word.keys():
self.addb.words[rw.name].set_impf(4)
num_twords += 1
else:
self.addb.words[rw.name].set_impf(
cands_word[rw.name][0])
self.addb.words[rw.name].add_smean(
cands_word[rw.name][1])
# print("new word {} added with impf {}".format(rw.name, addb.words[rw.name].get_impf()))
num_finished_words += 1
# rw.print_part()
else:
print("{} is already in addb?".format(rw.name))
except Exception as e:
print(
"[ERROR | ParsingManager/parse_csv] {}".format(
str(e)))
if num_finished_words % num_to_print == 0:
print(
"num_finished_words : {:5d} / {:5d} ({:.2f}%)".
format(num_finished_words, num_twords,
num_finished_words / num_twords * 100))
else:
time.sleep(0.05)
for i in range(self.num_gwp):
parsers[i].terminate()
parsers[i].join()
time_getwords = time.time() - time_getwords
print("\n\n\n{:.3f} seconds passed for processing {} words".format(
time_getwords, num_twords))
print("{} of them already existed".format(num_words_existed))
print("words_failed: {}".format(words_failed))
save_addb(self.addb, self.path_addb)
self.save_wordsfailed(self.path_wf, words_failed)
print("\n")
for cw in cands_word:
if cw not in self.addb.words.keys():
print("\"{}\" is not in addb!".format(cw))
# self.addb.view_words()
return num_finished_words
