From 17b8d635fc386e55ad1c19e9b10611f2f9046cb5 Mon Sep 17 00:00:00 2001
From: Stanley Chan <149976039+StanChan03@users.noreply.github.com>
Date: Tue, 14 Jan 2025 01:22:43 -0800
Subject: [PATCH] add missing operator_cache decorator (#80)

mssing decorator and doc updates
---
 docs/configurations.rst           | 10 ++++++++--
 lotus/sem_ops/sem_dedup.py        |  2 ++
 lotus/sem_ops/sem_index.py        |  2 ++
 lotus/sem_ops/sem_partition_by.py |  3 +++
 4 files changed, 15 insertions(+), 2 deletions(-)

diff --git a/docs/configurations.rst b/docs/configurations.rst
index 16cec150..d0fd9bc4 100644
--- a/docs/configurations.rst
+++ b/docs/configurations.rst
@@ -21,8 +21,8 @@ Using the Settings module
 Configurable Parameters
 --------------------------
 
-1. enable_message_cache: 
-    * Description: Enables or Disables cahcing mechanisms
+1. enable_cache: 
+    * Description: Enables or Disables caching mechanisms
     * Default: False
     * Parameters: 
         - cache_type: Type of caching (SQLITE or In_MEMORY)
@@ -31,6 +31,12 @@ Configurable Parameters
     * Note: It is recommended to enable caching
 .. code-block:: python
 
+    import pandas as pd
+
+    import lotus
+    from lotus.models import LM
+    from lotus.cache import CacheFactory, CacheConfig, CacheType
+    
     cache_config = CacheConfig(cache_type=CacheType.SQLITE, max_size=1000)
     cache = CacheFactory.create_cache(cache_config)
 
diff --git a/lotus/sem_ops/sem_dedup.py b/lotus/sem_ops/sem_dedup.py
index 7c900073..3360714d 100644
--- a/lotus/sem_ops/sem_dedup.py
+++ b/lotus/sem_ops/sem_dedup.py
@@ -4,6 +4,7 @@
 import pandas as pd
 
 import lotus
+from lotus.cache import operator_cache
 
 
 @pd.api.extensions.register_dataframe_accessor("sem_dedup")
@@ -19,6 +20,7 @@ def _validate(obj: Any) -> None:
         if not isinstance(obj, pd.DataFrame):
             raise AttributeError("Must be a DataFrame")
 
+    @operator_cache
     def __call__(
         self,
         col_name: str,
diff --git a/lotus/sem_ops/sem_index.py b/lotus/sem_ops/sem_index.py
index cb7d8cd4..ae8d7753 100644
--- a/lotus/sem_ops/sem_index.py
+++ b/lotus/sem_ops/sem_index.py
@@ -3,6 +3,7 @@
 import pandas as pd
 
 import lotus
+from lotus.cache import operator_cache
 
 
 @pd.api.extensions.register_dataframe_accessor("sem_index")
@@ -19,6 +20,7 @@ def _validate(obj: Any) -> None:
         if not isinstance(obj, pd.DataFrame):
             raise AttributeError("Must be a DataFrame")
 
+    @operator_cache
     def __call__(self, col_name: str, index_dir: str) -> pd.DataFrame:
         """
         Index a column in the DataFrame.
diff --git a/lotus/sem_ops/sem_partition_by.py b/lotus/sem_ops/sem_partition_by.py
index 298b5e5c..7886594c 100644
--- a/lotus/sem_ops/sem_partition_by.py
+++ b/lotus/sem_ops/sem_partition_by.py
@@ -2,6 +2,8 @@
 
 import pandas as pd
 
+from lotus.cache import operator_cache
+
 
 @pd.api.extensions.register_dataframe_accessor("sem_partition_by")
 class SemPartitionByDataframe:
@@ -16,6 +18,7 @@ def _validate(obj: Any) -> None:
         if not isinstance(obj, pd.DataFrame):
             raise AttributeError("Must be a DataFrame")
 
+    @operator_cache
     def __call__(
         self,
         partition_fn: Callable[[pd.DataFrame], list[int]],