sql = """
select K.a,K.b from (select H.b from (select G.c from (select F.d from
(select E.e from A, B, C, D, E), F), G), H), I, J, K order by 1,2;
"""
import sqlparse
from sqlparse.sql import IdentifierList, Identifier, Where
from sqlparse.tokens import Keyword, DML
def is_subselect(parsed):
if not parsed.is_group:
return False
for item in parsed.tokens:
if item.ttype is DML and item.value.upper() == 'SELECT':
return True
return False
def extract_from_part(parsed):
from_seen = False
for item in parsed.tokens:
if from_seen:
if is_subselect(item):
for x in extract_from_part(item):
yield x
elif item.ttype is Keyword:
raise StopIteration
else:
yield item
elif item.ttype is Keyword and item.value.upper() == 'FROM':
from_seen = True
def extract_table_identifiers(token_stream):
for item in token_stream:
if isinstance(item, IdentifierList):
for identifier in item.get_identifiers():
yield identifier.get_name()
elif isinstance(item, Identifier):
yield item.get_name()
# It's a bug to check for Keyword here, but in the example
# above some tables names are identified as keywords...
elif item.ttype is Keyword:
yield item.value
def extract_tables(sql):
stream = extract_from_part(sqlparse.parse(sql)[0])
return list(extract_table_identifiers(stream))
if __name__ == '__main__':
print 'Tables: %s' % ', '.join(extract_tables(sql))