minor changes

Signed-off-by: Praneeth Bedapudi <[email protected]>

liteindex/__init__.py

@@ -1,4 +1,4 @@
-from .defined_index import DefinedIndex
+from .defined_index import DefinedIndex, get_defined_index_names_in_db
 from .defined_serializers import DefinedTypes
 from .kv_index import KVIndex
 from .function_cache import function_cache

liteindex/defined_index.py

@@ -51,6 +51,27 @@
 import threading
 
 
+def get_defined_index_names_in_db(db_path):
+    with sqlite3.connect(db_path, uri=True) as conn:
+        all_table_names = {
+            _[0]
+            for _ in conn.execute(
+                "SELECT name FROM sqlite_master WHERE type='table';"
+            ).fetchall()
+        }
+
+        res = []
+        for table_name in all_table_names:
+            if (
+                not table_name.startswith("__")
+                and not table_name.startswith("sqlite_")
+                and f"__{table_name}_meta" in all_table_names
+            ):
+                res.append(table_name)
+
+        return res
+
+
 class DefinedIndex:
     def __init__(
         self, name, schema=None, db_path=None, ram_cache_mb=64, compression_level=-1
@@ -604,12 +625,14 @@ def distinct(self, key, query={}):
             _[0] for _ in self.__connection.execute(sql_query, sql_params).fetchall()
         }
 
-    def distinct_count(self, key, query={}):
+    def distinct_count(self, key, query={}, min_count=0, top_n=None):
         sql_query, sql_params = distinct_count_query(
             table_name=self.name,
             column=key,
             query={k: v for k, v in query.items()},
             schema=self.schema,
+            min_count=min_count,
+            top_n=top_n,
         )
 
         return {
@@ -729,12 +752,12 @@ def optimize_for_query(self, keys, is_unique=False):
 
         if key_hashes:
             self.__connection.execute(
-                f"CREATE {'UNIQUE' if is_unique else ''} INDEX IF NOT EXISTS idx_{self.name}_{'_'.join(key_hashes)} ON {self.name} ({','.join(key_hashes)})"
+                f"""CREATE {'UNIQUE' if is_unique else ''} INDEX IF NOT EXISTS "idx_{self.name}_{'_'.join(key_hashes)}" ON {self.name} ({','.join(['"' + _ + '"' for _ in key_hashes])})"""
             )
 
             for size_hash in size_hashes:
                 self.__connection.execute(
-                    f"CREATE INDEX IF NOT EXISTS idx_{self.name}_{size_hash} ON {self.name} ({size_hash})"
+                    f"""CREATE INDEX IF NOT EXISTS "idx_{self.name}_{size_hash}" ON {self.name} ({size_hash})"""
                 )
 
             self.__connection.commit()

liteindex/query_parser.py

@@ -200,17 +200,26 @@ def distinct_query(table_name, column, query, schema):
     return query_str, params
 
 
-def distinct_count_query(table_name, column, query, schema):
-    # Prepare the query
+def distinct_count_query(table_name, column, query, schema, min_count=0, top_n=None):
     where_conditions, params = parse_query(query, schema)
 
     if schema[column] == "json":
-        query_str = f"SELECT JSON_EXTRACT({column}, '$[*]'), COUNT(*) FROM {table_name}"
+        query_str = f"SELECT JSON_EXTRACT({column}, '$[*]') AS extracted_column, COUNT(*) AS count FROM {table_name}"
     else:
-        query_str = f"SELECT {column}, COUNT(*) FROM {table_name}"
+        query_str = f"SELECT {column}, COUNT(*) AS count FROM {table_name}"
+
     if where_conditions:
         query_str += f" WHERE {' AND '.join(where_conditions)}"
-    query_str += f" GROUP BY {column}"
+
+    if schema[column] == "json":
+        query_str += " GROUP BY extracted_column"
+    else:
+        query_str += f" GROUP BY {column}"
+
+    query_str += f" HAVING COUNT(*) >= {min_count}"
+
+    if top_n is not None:
+        query_str += f" ORDER BY COUNT(*) DESC LIMIT {top_n}"
 
     return query_str, params
 

setup.py

@@ -18,7 +18,7 @@
 EMAIL = "[email protected]"
 AUTHOR = "BEDAPUDI PRANEETH"
 REQUIRES_PYTHON = ">=3.6.0"
-VERSION = "0.0.2.dev56"
+VERSION = "0.0.2.dev57"
 
 # What packages are required for this module to be executed?
 REQUIRED = []