def __init__(self):
self.index_binary = None
self.files = dict()
self.aodp = DbHandler()
# Создаем временную папку, если ее нет
if not os.path.exists(Folders.temp):
os.makedirs(Folders.temp)
# оздаем 3 папки для Сфинкса
if not os.path.exists(SphinxConfig.var_dir+ '/run'):
os.makedirs(SphinxConfig.var_dir + '/run')
if not os.path.exists(SphinxConfig.var_dir+ '/log'):
os.makedirs(SphinxConfig.var_dir + '/log')
if not os.path.exists(SphinxConfig.var_dir+ '/data'):
os.makedirs(SphinxConfig.var_dir + '/data')
def __init__(self, source="http"):
self.db_handler = DbHandler()
self.source = source
self.updalist_generator = None
self.tablelist_generator = None
self.allowed_tables = None
class Updater:
# Source: "http", directory (as a full path to unpacked xmls)
def __init__(self, source="http"):
self.db_handler = DbHandler()
self.source = source
self.updalist_generator = None
self.tablelist_generator = None
self.allowed_tables = None
def __get_entries_from_folder(self, path_to_xmls):
for (dirpath, dirnames, filenames) in walk(path_to_xmls):
for filename in filenames:
if filename.endswith(".XML"):
xmltable = AoXmlTableEntry.from_dir(filename, dirpath.replace("\\", "/") + "/")
if xmltable.table_name in allowed_tables:
yield xmltable
break
@classmethod
def get_current_fias_version(cls):
db = None
try:
db = DBImpl(psycopg2, DatabaseConfig)
rows = db.get_rows('SELECT version FROM "CONFIG" WHERE id=0', True)
assert len(rows) > 0, "Cannot get a version"
return rows[0]['version']
except:
return 0
finally:
if db:
db.close()
@classmethod
def __set__update_version(cls, updver=0):
db = DBImpl(psycopg2, DatabaseConfig)
try:
assert isinstance(updver, int), "Update version must be of int type."
db.execute('UPDATE "CONFIG" SET version={} WHERE id=0'.format(updver))
finally:
db.close()
# Получает верию ФИАС с клавиатуры (если мы берем базу из папки или локального архива и не можем определить,
# что это за версия
@staticmethod
def __get_update_version_from_console():
mode = None
while not mode:
try:
mode = int(raw_input('Enter FIAS update version (3 digit):'))
except ValueError:
print "Not a valid fias version, try again."
return mode
def __get_updates_from_folder(self, foldername):
fias_db_version = self.__get_update_version_from_console()
yield dict(intver=fias_db_version,
textver="Version {}".format(fias_db_version),
delta_url=foldername,
complete_url=foldername)
@staticmethod
def __get_updates_from_rar(url):
aorar = AoRar()
if url.startswith("http://") or url.startswith("https://"):
aorar.download(url)
if url.endswith(".rar") and path.isfile(url):
aorar.local(url)
for table_entry in aorar.get_table_entries(allowed_tables):
yield table_entry
def __init_update_entries(self, updates_generator):
if self.source == "http":
assert updates_generator, "No generator"
self.tablelist_generator = self.__get_updates_from_rar
self.updalist_generator = updates_generator
return
if self.source.endswith(".rar"):
self.tablelist_generator = self.__get_updates_from_rar
self.updalist_generator = self.__get_updates_from_folder(self.source)
return
if path.isdir(self.source):
self.tablelist_generator = self.__get_entries_from_folder
self.updalist_generator = self.__get_updates_from_folder(self.source)
assert self.tablelist_generator, "No valid source."
def process_single_entry(self, operation_type, table_xmlentry, chunck_size=50000):
aoparser = AoDataParser(table_xmlentry, chunck_size)
aoparser.parse(lambda x, y: self.db_handler.bulk_csv(operation_type, table_xmlentry.table_name, x, y))
def create(self, updates_generator):
self.__init_update_entries(updates_generator)
self.db_handler.create_structure()
for update_entry in self.updalist_generator:
logging.info("Processing DB #%d", update_entry['intver'])
for table_entry in self.tablelist_generator(update_entry['complete_url']):
if table_entry.operation_type == AoXmlTableEntry.OperationType.update:
table_entry.operation_type = AoXmlTableEntry.OperationType.create
self.process_single_entry(table_entry.operation_type, table_entry)
Updater.__set__update_version(update_entry['intver'])
else:
logging.info("No updates more.")
self.db_handler.create_indexes(db_shemas.keys())
logging.info("Create success")
def update(self, updates_generator):
self.__init_update_entries(updates_generator)
# Drop all indexes if updates needed
indexes_dropped = False
for update_entry in self.updalist_generator:
if not indexes_dropped:
self.db_handler.drop_indexes(allowed_tables)
indexes_dropped = True
logging.info("Processing update #%d", update_entry['intver'])
for table_entry in self.tablelist_generator(update_entry['delta_url']):
self.process_single_entry(table_entry.operation_type, table_entry)
Updater.__set__update_version(update_entry['intver'])
else:
logging.info("No updates more.")
# Re-create all indexes (if dropped)
if indexes_dropped:
self.db_handler.create_indexes(allowed_tables)
logging.info("Update success")
class SphinxHelper:
def __init__(self):
self.index_binary = None
self.files = dict()
self.aodp = DbHandler()
# Создаем временную папку, если ее нет
if not os.path.exists(Folders.temp):
os.makedirs(Folders.temp)
# оздаем 3 папки для Сфинкса
if not os.path.exists(SphinxConfig.var_dir+ '/run'):
os.makedirs(SphinxConfig.var_dir + '/run')
if not os.path.exists(SphinxConfig.var_dir+ '/log'):
os.makedirs(SphinxConfig.var_dir + '/log')
if not os.path.exists(SphinxConfig.var_dir+ '/data'):
os.makedirs(SphinxConfig.var_dir + '/data')
def configure_indexer(self, indexer_binary, config_filename):
logging.info("Start configuring Sphinx...")
self.index_binary = indexer_binary
# Create ADDROBJ config
self.files['addrobj.conf'] = self.__create_ao_index_config()
# Produce dict file
self.files['dict.txt'] = self.__create_suggestion_dict()
# Put dict into db
self.files['dict.csv'] = self.__dbexport_sugg_dict()
# Create SUGGEST config
self.files['suggest.conf'] = self.__create_sugg_index_config()
# Create main config (sphinx.conf)
out_fname = self.__create_main_config(config_filename)
# Indexing both configs
run_index_cmd = "{} -c {} --all --rotate".format(self.index_binary, out_fname)
logging.info("Indexing main (%s)...", out_fname)
os.system(run_index_cmd)
logging.info("All indexes were created.")
# remove temp files
for fname, fpath in self.files.iteritems():
try:
os.remove(fpath)
except:
logging.warning("Cannot delete %s. Not accessible.", fpath)
logging.info("Temporary files removed.")
logging.info("Successfully configured. Please restart searchd.")
def __create_sugg_index_config(self):
fname = os.path.abspath(Folders.temp + "/suggest.conf")
logging.info("Creating config %s", fname)
conf_data = template('aore/templates/sphinx/idx_suggest.conf', db_host=DatabaseConfig.host,
db_user=DatabaseConfig.user,
db_password=DatabaseConfig.password,
db_name=DatabaseConfig.database, db_port=DatabaseConfig.port,
index_name=SphinxConfig.index_sugg,
sphinx_var_path=SphinxConfig.var_dir)
f = open(fname, "w")
f.write(conf_data)
f.close()
logging.info("Done.")
return fname
def __dbexport_sugg_dict(self):
logging.info("Place suggestion dict to DB %s...", self.files['dict.txt'])
dict_dat_fname = os.path.abspath(Folders.temp + "/suggdict.csv")
csv_counter = 0
with open(self.files['dict.txt'], "r") as dict_file, open(dict_dat_fname, "w") as exit_file:
line = None
while line != '':
nodes = []
line = dict_file.readline()
if line == '':
break
csv_counter += 1
splitting_seq = line.split(' ')
keyword = splitting_seq[0]
freq = splitting_seq[1].rstrip('\n')
assert keyword and freq, "Cannot process {}".format(self.files['dict.txt'])
nodes.append(keyword)
nodes.append(trigram(keyword))
nodes.append(freq)
exit_file.write("\t".join(nodes) + "\n")
try:
dict_file.close()
exit_file.close()
except:
pass
self.aodp.bulk_csv(AoXmlTableEntry.OperationType.update, "AOTRIG", csv_counter, dict_dat_fname)
logging.info("Done.")
def __create_ao_index_config(self):
fname = os.path.abspath(Folders.temp + "/addrobj.conf")
logging.info("Creating config %s", fname)
conf_data = template('aore/templates/sphinx/idx_addrobj.conf', db_host=DatabaseConfig.host,
db_user=DatabaseConfig.user,
db_password=DatabaseConfig.password,
db_name=DatabaseConfig.database, db_port=DatabaseConfig.port,
sql_query=template('aore/templates/postgre/sphinx_query.sql').replace("\n", " \\\n"),
index_name=SphinxConfig.index_addjobj,
sphinx_var_path=SphinxConfig.var_dir,
min_length_to_star=SphinxConfig.min_length_to_star)
f = open(fname, "w")
f.write(conf_data)
f.close()
logging.info("Done.")
return fname
def __create_suggestion_dict(self):
fname = os.path.abspath(Folders.temp + "/suggdict.txt")
logging.info("Make suggestion dict (%s)...", fname)
run_builddict_cmd = "{} {} -c {} --buildstops {} 200000 --buildfreqs".format(self.index_binary,
SphinxConfig.index_addjobj,
self.files['addrobj.conf'], fname)
os.system(run_builddict_cmd)
logging.info("Done.")
return fname
def __create_main_config(self, config_fname):
out_filename = os.path.abspath(config_fname)
logging.info("Creating main config %s...", out_filename)
conf_data = template('aore/templates/sphinx/sphinx.conf',
sphinx_listen=SphinxConfig.listen.replace("unix://", ''),
sphinx_var_path=SphinxConfig.var_dir)
f = open(out_filename, "w")
for fname, fpath in self.files.iteritems():
if ".conf" in fname:
with open(fpath, "r") as conff:
for line in conff:
f.write(line)
f.write('\n')
f.write(conf_data)
f.close()
logging.info("Done.")
return out_filename
