remove fixed schema parameter

ohadmata · May 29, 2024 · 2175ba1 · 2175ba1
1 parent ee7bea3
commit 2175ba1
Show file tree

Hide file tree

Showing 3 changed files with 4 additions and 14 deletions.
diff --git a/README.md b/README.md
@@ -149,10 +149,7 @@ shmessy.infer_schema(df: Dataframe) -> ShmessySchema
 
 ### fix_schema
 ```python
-shmessy.fix_schema(
-    df: Dataframe,
-    fixed_schema: Optional[ShmessySchema] = None,  # Fix the given DF according to this schema
-) -> DataFrame
+shmessy.fix_schema(df: Dataframe) -> DataFrame
 ```
 
 ### get_inferred_schema

diff --git a/examples/read_csv.py b/examples/read_csv.py
@@ -7,7 +7,7 @@
     init_logger()
     shmessy = Shmessy()
     start_time = time.time()
-    df = shmessy.read_csv('../tests/data/data_1.csv')
+    df = shmessy.read_csv('../tests/data/data_30.csv')
     inferred_schema = shmessy.get_inferred_schema()
     pretty_print_df(df=df, inferred_schema=inferred_schema)
     print(f"Duration: {int((time.time() - start_time) * 1000)}")
diff --git a/src/shmessy/__init__.py b/src/shmessy/__init__.py
@@ -74,17 +74,10 @@ def infer_schema(self, df: DataFrame) -> ShmessySchema:
         self.__inferred_schema = inferred_schema
         return inferred_schema
 
-    def fix_schema(
-        self,
-        df: DataFrame,
-        *,
-        fixed_schema: Optional[ShmessySchema] = None,
-    ) -> DataFrame:
+    def fix_schema(self, df: DataFrame) -> DataFrame:
         try:
             _check_number_of_columns(df=df, max_columns_num=self.__max_columns_num)
-
-            if fixed_schema is None:
-                fixed_schema = self.infer_schema(df)
+            fixed_schema = self.infer_schema(df)
 
             for column in fixed_schema.columns:
                 df[column.field_name] = self.__types_handler.fix_field(