Update version info to 0.6.0.dev3 (#2511)

    ### What problem does this PR solve?

    Update version info.

    ### Type of change

    - [x] Refactoring
    - [x] Python SDK impacted, Need to update PyPI

Signed-off-by: Jin Hai <[email protected]>

Author: JinHai-CN

README.md

@@ -60,7 +60,7 @@ Supports a wide range of data types including strings, numerics, vectors, and mo
 Infinity supports two working modes, embedded mode and client-server mode. Infinity's embedded mode enables you to quickly embed Infinity into your Python applications, without the need to connect to a separate backend server. The following shows how to operate in embedded mode:
 
    ```bash
-   pip install infinity-embedded-sdk==0.6.0.dev2
+   pip install infinity-embedded-sdk==0.6.0.dev3
    ```
    Use Infinity to conduct a dense vector search:
    ```python

benchmark/remote_infinity/remote_query_benchmark.cpp

@@ -51,7 +51,7 @@ struct InfinityClient {
         transport->open();
         CommonResponse response;
         ConnectRequest request;
-        request.__set_client_version(28); // 0.6.0.dev1 and 0.6.0.dev2
+        request.__set_client_version(29); // 0.6.0.dev3
         client->Connect(response, request);
         session_id = response.session_id;
     }

client/cpp/infinity_client.cpp

@@ -25,7 +25,7 @@ Client Client::Connect(const std::string &ip_address, uint16_t port) {
     transport->open();
     CommonResponse response;
     ConnectRequest request;
-    request.__set_client_version(28); // 0.6.0.dev1 and 0.6.0.dev2
+    request.__set_client_version(29); // 0.6.0.dev3
     client->Connect(response, request);
     return {socket, transport, protocol, std::move(client), response.session_id};
 }

docs/getstarted/build_from_source.mdx

@@ -264,7 +264,7 @@ cmake --build . -t test_main
 
 2. Install Python sdk of infinity:
    ```bash
-   pip install infinity-sdk==0.6.0.dev2
+   pip install infinity-sdk==0.6.0.dev3
    ```
 
 3. Run the functional tests:
@@ -286,7 +286,7 @@ cmake --build . -t test_main
 
 2. Install Python sdk of infinity:
    ```bash
-   pip install infinity-sdk==0.6.0.dev2
+   pip install infinity-sdk==0.6.0.dev3
    ```
 
 3. Run the functional tests: 
@@ -309,7 +309,7 @@ cmake --build . -t test_main
 
 2. Install Python sdk of infinity:
    ```bash
-   pip install infinity-sdk==0.6.0.dev2
+   pip install infinity-sdk==0.6.0.dev3
    ```
 
 3. Run the functional tests: 

docs/getstarted/deploy_infinity_server.mdx

@@ -34,7 +34,7 @@ This approach allows you to call Infinity as a Python module. To deploy Infinity
 ### Install Infinity as a module
 
 ```
-pip install infinity-embedded-sdk==0.6.0.dev2
+pip install infinity-embedded-sdk==0.6.0.dev3
 ```
 
 ### Create an Infinity object
@@ -101,7 +101,7 @@ If you are on Windows 10+, you must enable WSL or WSL2 to deploy Infinity using
 ### Install Infinity client
 
 ```
-pip install infinity-sdk==0.6.0.dev2
+pip install infinity-sdk==0.6.0.dev3
 ```
 
 ### Run a vector search
@@ -151,7 +151,7 @@ This section provides instructions on deploying Infinity using binary package on
 
 Fedora/RHEL/CentOS/OpenSUSE
 ```bash
-sudo rpm -i infinity-0.6.0.dev2-x86_64.rpm
+sudo rpm -i infinity-0.6.0.dev3-x86_64.rpm
 ```
 
 ```bash
@@ -162,7 +162,7 @@ sudo systemctl start infinity
   <TabItem value="ubuntu">
 
 ```bash
-sudo dpkg -i infinity-0.6.0.dev2-x86_64.deb
+sudo dpkg -i infinity-0.6.0.dev3-x86_64.deb
 ```
 
 ```bash
@@ -175,7 +175,7 @@ sudo systemctl start infinity
 ### Install Infinity client
 
 ```
-pip install infinity-sdk==0.6.0.dev2
+pip install infinity-sdk==0.6.0.dev3
 ```
 
 ### Run a vector search

docs/getstarted/quickstart.md

@@ -21,7 +21,7 @@ If you wish to embed Infinity into your Python application without the need for
 
 1. Install the Infinity-embedded SDK:
    ```bash
-   pip install infinity-embedded-sdk==0.6.0.dev2
+   pip install infinity-embedded-sdk==0.6.0.dev3
    ```
 2. Use Infinity to conduct a dense vector search:
    ```python

docs/guides/search_guide.md

@@ -91,7 +91,7 @@ Infinity supports the following syntax or full-text search expressions:
 - AND multiple terms
 - OR multiple terms
 - Phrase search
-- CARAT opertor
+- CARAT operator
 - Sloppy phrase search
 - Field-specific search
 - Escape character
@@ -164,7 +164,7 @@ Infinity supports real-time sparse vector search, which is based on a sparse vec
 
 Tensors work in scenarios where embeddings of all tokens of each document need to be preserved. Tensor is another form of multi-vector, but *not* the multi-vector mentioned in dense vector search. They share similarities but are different in several aspects:
 
-Multi-vectors are mainly used for dense vector search, while tensors are mainly used for tensor reranking. Tensor search is *not* recommended due to the potential for significant storage overhead when converting raw documents to tensors; tensors are more appropriate for reranking in a [hyprid search](#hybrid-search).
+Multi-vectors are mainly used for dense vector search, while tensors are mainly used for tensor reranking. Tensor search is *not* recommended due to the potential for significant storage overhead when converting raw documents to tensors; tensors are more appropriate for reranking in a [hybrid search](#hybrid-search).
 
 A multi-vector search remains a KNN search, where search results are aggregated using vector identifiers. In contrast, a tensor search uses 'Colxxx' models, such as ColBERT, ColPali, and ColQwen2. In this context, 'Col' denotes 'contextualized late interaction' and 'xxx' refers to specific models that can be added using a 'Col adapter'. Additionally, a tensor search uses the MaxSim method for similarity operations, which accumulates similarities of all embedding pairs between a query tensor and a document tensor.
 
@@ -209,4 +209,4 @@ Filters based on secondary index can have arbitrary logical combinations. Suppor
 
 ### Full-text index filters
 
-Infinity's full-text index supports conditional filtering through the `filter_fulltext` parameter. Full-text index filters are keyword-based and do not support the expressions available for secondary index filters. They use the `minimum_should_match` parameter to specify the minimum number of keywords that must be satisfied during fitlering.
+Infinity's full-text index supports conditional filtering through the `filter_fulltext` parameter. Full-text index filters are keyword-based and do not support the expressions available for secondary index filters. They use the `minimum_should_match` parameter to specify the minimum number of keywords that must be satisfied during filtering.

example/README.md

@@ -5,13 +5,13 @@
 The Python scripts in the current directory provides the examples of the infinity SDK API. Before running these scripts, please ensure that the correct version of infinity SDK is installed. If you are using the server version of Infinity, you can use the following command to install it
 
 ```shell
-pip install infinity-sdk==0.6.0.dev2
+pip install infinity-sdk==0.6.0.dev3
 ```
 
 To use Infinity locally as a Python module, install the Infinity Embedded SDK:
 
 ```shell
-pip install infinity-embedded-sdk==0.6.0.dev2
+pip install infinity-embedded-sdk==0.6.0.dev3
 ```
 
 Once the installation completes, run some simple examples:

pyproject.toml

@@ -11,7 +11,7 @@ build-backend = "scikit_build_core.build"
 
 [project]
 name = "infinity_embedded_sdk"
-version = "0.6.0.dev2"
+version = "0.6.0.dev3"
 requires-python = ">=3.10"
 dependencies = [
     "sqlglot~=11.7.0",

python/infinity_sdk/README.md

@@ -63,7 +63,7 @@ Infinity supports two working modes, embedded mode and client-server mode. The f
 
 2. Install the `infinity-sdk` package:
    ```bash
-   pip install infinity-sdk==0.6.0.dev2
+   pip install infinity-sdk==0.6.0.dev3
    ```
 
 3. Use Infinity to conduct a dense vector search:

python/infinity_sdk/infinity/remote_thrift/client.py

@@ -99,7 +99,8 @@ def _reconnect(self):
         # version: 0.5.0.dev4 and 0.5.0.dev5, client_version: 26
         # version: 0.5.0.dev6 and 0.5.0.dev7 and 0.5.0 and 0.5.1 and 0.5.2, client_version: 27
         # version: 0.6.0.dev1 and 0.6.0.dev2, client_version: 28
-        res = self.client.Connect(ConnectRequest(client_version=28)) # 0.6.0.dev1 and 0.6.0.dev2
+        # version: 0.6.0.dev3, client_version: 29
+        res = self.client.Connect(ConnectRequest(client_version=29)) # 0.6.0.dev3
         if res.error_code != 0:
             raise InfinityException(res.error_code, res.error_msg)
         self.session_id = res.session_id

python/infinity_sdk/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "infinity-sdk"
-version = "0.6.0.dev2"
+version = "0.6.0.dev3"
 requires-python = ">=3.10"
 dependencies = [
     "sqlglot~=11.7.0",

src/network/infinity_thrift_service.cpp

@@ -96,6 +96,7 @@ ClientVersions::ClientVersions() {
     client_version_map_[26] = String("0.5.0.dev5");
     client_version_map_[27] = String("0.5.2");
     client_version_map_[28] = String("0.6.0.dev2");
+    client_version_map_[29] = String("0.6.0.dev3");
 }
 
 Pair<const char *, Status> ClientVersions::GetVersionByIndex(i64 version_index) {

src/network/infinity_thrift_service.cppm

@@ -60,7 +60,7 @@ struct ClientVersions {
 export class InfinityThriftService final : public infinity_thrift_rpc::InfinityServiceIf {
 private:
     static constexpr std::string_view ErrorMsgHeader = "[THRIFT ERROR]";
-    static constexpr i64 current_version_index_{28}; // 0.6.0.dev1 and 0.6.0.dev2
+    static constexpr i64 current_version_index_{29}; // 0.6.0.dev3
 
     static std::mutex infinity_session_map_mutex_;
     static HashMap<u64, SharedPtr<Infinity>> infinity_session_map_;

src/planner/binder/project_binder.cpp

@@ -14,6 +14,8 @@
 
 module;
 
+#include <string>
+
 module project_binder;
 
 import stl;