diff --git a/preprocessing/process_sql.py b/preprocessing/process_sql.py
index 686ee1f..63b48ae 100644
--- a/preprocessing/process_sql.py
+++ b/preprocessing/process_sql.py
@@ -29,9 +29,10 @@
 import json
 import sqlite3
 from nltk import word_tokenize
+import pdb
 
 CLAUSE_KEYWORDS = ('select', 'from', 'where', 'group', 'order', 'limit', 'intersect', 'union', 'except')
-JOIN_KEYWORDS = ('join', 'on', 'as')
+JOIN_KEYWORDS = ('left', 'join', 'on', 'as')
 
 WHERE_OPS = ('not', 'between', '=', '>', '<', '>=', '<=', '!=', 'in', 'like', 'is', 'exists')
 UNIT_OPS = ('none', '-', '+', "*", '/')
@@ -397,10 +398,15 @@ def parse_from(toks, start_idx, tables_with_alias, schema):
             idx, sql = parse_sql(toks, idx, tables_with_alias, schema)
             table_units.append((TABLE_TYPE['sql'], sql))
         else:
-            if idx < len_ and toks[idx] == 'join':
+            join_type = None
+            if idx < len_ - 1 and toks[idx] == 'left' and toks[idx+1] == 'join':
+                join_type = 'left'
+                idx += 2  # skip left join
+            elif idx < len_ and toks[idx] == 'join':
+                join_type = 'inner'
                 idx += 1  # skip join
             idx, table_unit, table_name = parse_table_unit(toks, idx, tables_with_alias, schema)
-            table_units.append((TABLE_TYPE['table_unit'],table_unit))
+            table_units.append((TABLE_TYPE['table_unit'],table_unit,join_type))
             default_tables.append(table_name)
         if idx < len_ and toks[idx] == "on":
             idx += 1  # skip on