Update py sdk version (#336)

* Change insert more values * Add hello_infinity.py demo * One thread benchmark * Modify benchmark multithreading * Update py sdk version
infiniflow · Dec 20, 2023 · 067191c · 067191c
1 parent 631d8fc
commit 067191c
Show file tree

Hide file tree

Showing 26 changed files with 307 additions and 949 deletions.
diff --git a/.gitignore b/.gitignore
@@ -99,8 +99,8 @@ benchmark/csv/csv_config.h
 #python sdk
 python/build
 python/dist
-python/infinity.egg-info
 python/.idea
+python/infinity_sdk.egg-info
 
 
 sift_1m

diff --git a/python/README.md b/python/README.md
@@ -17,106 +17,35 @@ sudo apt-get install thrift-compiler
 # build
 
 ```shell
-python setup.py bdist_wheel
+python setup.py sdist bdist_wheel
 ```
 
 # install
 ```shell
-pip install dist/infinity-0.0.1-py3-none-any.whl
+pip install dist/*.whl
 ```
+
+# upload
+twine upload dist/*  
 # using
 
 ```python
 import infinity
-from infinity import NetworkAddress
-
-infinity_obj = infinity.connect(NetworkAddress('127.0.0.1', 23817))
-
-# infinity
-res = infinity_obj.create_database("my_db")
-
-res = infinity_obj.list_databases()
-
-res = infinity_obj.drop_database("my_db")
-
-db_obj = infinity_obj.get_database("default")
-res = db_obj.create_table("my_table1", {"c1": "int, primary key"}, None)
-
-res = db_obj.list_tables()
-
-res = db_obj.drop_table("my_table1")
-
-# index
-res = db_obj.create_table("my_table2", {"c1": "vector,1024,float"}, None)
-
-table_obj = db_obj.get_table("my_table2")
-
-res = table_obj.create_index("my_index",
-                             [index.IndexInfo("c1",
-                                              index.IndexType.IVFFlat,
-                                              [index.InitParameter("centroids_count", "128"),
-                                               index.InitParameter("metric", "l2")])], None)
-
-res = table_obj.drop_index("my_index")
-
-res = db_obj.drop_table("my_table2")
-
-# insert
-res = db_obj.create_table("my_table3", {"c1": "int, primary key", "c2": "float"}, None)
-
-table_obj = db_obj.get_table("my_table3")
-
-res = table_obj.insert([{"c1": 1, "c2": 1.1}, {"c1": 2, "c2": 2.2}])
-
-res = db_obj.create_table("test_insert_varchar", {"c1": "varchar"}, None)
-
-table_obj = db_obj.get_table("test_insert_varchar")
-
-res = table_obj.insert([{"c1": "test_insert_varchar"}])
-
-res = db_obj.create_table("test_insert_embedding", {"c1": "vector,3,int"}, None)
-
-table_obj = db_obj.get_table("test_insert_embedding")
-
-res = table_obj.insert([{"c1": [4, 5, 6]}])
-
-res = table_obj.insert([{"c1": [1, 2, 3]}, {"c1": [4, 5, 6]}, {"c1": [7, 8, 9]}, {"c1": [-7, -8, -9]}])
-
-# search
-
-res = table_obj.search().output(["c1 + 0.1"]).to_df()
-
-res = table_obj.search().output(["*"]).filter("c1 > 1").to_df()
-
-# import
-res = db_obj.create_table("my_table4", {"c1": "int", "c2": "vector,3,int"}, None)
-table_obj = db_obj.get_table("my_table4")
-parent_dir = os.path.dirname(os.path.dirname(os.getcwd()))
-test_csv_dir = parent_dir + "/test/data/csv/embedding_int_dim3.csv"
-
-res = table_obj.import_data(test_csv_dir, None)
-
-# search
-res = table_obj.search().output(["c1"]).filter("c1 > 1").to_df()
-
-res = db_obj.drop_table("my_table4")
-
-
-res = db_obj.create_table("table_4", {"c1": "int, primary key, not null", "c2": "int", "c3": "int"}, None)
-
-table_obj = db_obj.get_table("table_4")
-
-res = table_obj.insert(
-    [{"c1": 1, "c2": 10, "c3": 100}, {"c1": 2, "c2": 20, "c3": 200}, {"c1": 3, "c2": 30, "c3": 300},
-     {"c1": 4, "c2": 40, "c3": 400}])
-
-res = table_obj.update("c1 = 1", [{"c2": 90, "c3": 900}])
-
-res = table_obj.delete("c1 = 1")
-
-res = table_obj.delete()
-
-# disconnect
-res = infinity_obj.disconnect()
+from infinity.common import REMOTE_HOST
+
+infinity_obj = infinity.connect(REMOTE_HOST)
+db = infinity_obj.get_database("default")
+db.drop_table("my_table", if_exists=True)
+table = db.create_table(
+    "my_table", {"num": "integer", "body": "varchar", "vec": "vector,5,float"}, None)
+table.insert(
+    [{"num": 1, "body": "undesirable, unnecessary, and harmful", "vec": [1.0] * 5}])
+table.insert(
+    [{"num": 2, "body": "publisher=US National Office for Harmful Algal Blooms", "vec": [4.0] * 5}])
+table.insert(
+    [{"num": 3, "body": "in the case of plants, growth and chemical", "vec": [7.0] * 5}])
+
+res = table.output(["*"]).knn("vec", [3.0] * 5, "float", "ip", 2).to_pl()
+print(res)
 
 ```