We are thrilled to announce the General Availability (GA) of the C++ SDKfor Couchbase! This release adds support for native C++ language to our existing comprehensive set of SDK libraries in 11 programming languages and marks a significant milestone in our commitment to providing robust, high-performance tools for developers to build modern, scalable applications. In fact, this C++ SDK is the core library behind our existing Python, NodeJS, Ruby, and PHP SDKs and strives to provide a consistent, performant and efficient experience across all these programming languages.
C++ remains a vital language for many developers due to its performance, efficiency, and control over system resources. By introducing the C++ SDK, Couchbase empowers developers to leverage these advantages while taking full benefit of Couchbase’s advanced NoSQL capabilities. Here are some key reasons why this release is a game-changer:
C++ is renowned for its speed and low-level memory management, making it ideal for applications where performance is critical. The C++ SDK for Couchbase ensures that you can build high-performance applications without sacrificing speed or efficiency.
The C++ SDK provides seamless integration with Couchbase Server, enabling developers to easily perform operations such as KV operations, SQL++ queries, and transactions. This tight integration ensures that your applications can scale effectively while maintaining high performance.
The C++ SDK features a modern, idiomatic API that aligns with the latest C++ standards. This makes it not only powerful but also intuitive for C++ developers, reducing the learning curve and allowing you to start building applications quickly.
Here are some of the standout features of the C++ SDK for Couchbase:
The C++ SDK simplifies connection management, allowing you to establish and manage connections to your Couchbase cluster with minimal effort. This includes handling connection pooling, load balancing, and failover, ensuring your application remains resilient and performant. This means that the SDK can automatically read the topology of your cluster and provide seamless connections during topology changes such as cluster resizes or cluster upgrades. Below is an example of C++ code to connect to your couchbase database.
// Attempt to connect to the Couchbase cluster
auto [connect_err, cluster] = couchbase::cluster::connect(config.connection_string, options).get();
if (connect_err) {
// Output the error message if connection fails
std::cout << "Unable to connect to the cluster. Error code: " << connect_err.message() << "\n";
} else {
auto collection = cluster.bucket(config.bucket_name)
.scope(config.scope_name)
.collection(config.collection_name);
// d something interesting
cluster.close().get();
}
The C++ SDK natively supports performing key value operations. Key-value operations are unique to couchbase and provide very fast CRUD operations for documents stored in couchbase. Below is an example of C++ code to run a simple KV get & upsert of a document.
auto collection = cluster.bucket(config.bucket_name)
.scope(config.scope_name)
.collection(config.collection_name);
// KV- get
auto record = collection.get(document_id);
std::cout << record.content_as() << "\n";
// create a new document
const std::string document_id{ "minimal_example" };
const tao::json::value basic_doc{
{ "a", 1.0 },
{ "b", 2.0 },
};
// KV
auto [err, resp] = collection.upsert(document_id, basic_doc, {}).get();
if (err.ec()) {
std::cout << "ec: " << err.message() << ", ";
} else {
std::cout << "id: " << document_id << ", CAS: " << resp.cas().value() << "\n";
}
cluster.close().get();
The C++ SDK supports SQL++ queries, full-text search (FTS), and vector search, providing you with powerful tools to perform complex data operations. Whether you need to run sophisticated queries or perform full-text searches, the SDK has you covered.
Below is a code snippet to execute a simple SQL++ query to fetch records from airlines collection:
auto scope = cluster.bucket(config.bucket_name).scope(config.scope_name);
auto [err, resp] = scope.query("SELECT * FROM airline LIMIT 10").get();
std::cout << "--- Iterating as JSON objects:\n";
for (auto row : resp.rows_as_json()) {
std::cout << "Airline(id: " << row["airline"]["id"] << ", name: \"" << row["airline"]["name"] << "\")\n";
}
Below is a code snippet to execute a FTS query to do a full text search for “nice restaurants” the index created on the landmark collection:
auto scope = cluster.bucket(config.bucket_name).scope(config.scope_name);
auto [err, resp] = scope
.search("travel-inventory-landmarks",
couchbase::search_request(couchbase::query_string_query("nice restaurants")),
couchbase::search_options{}.fields({ "content" }))
.get();
for (const auto& row : resp.rows()) {
auto fields = row.fields_as<couchbase::codec::tao_json_serializer>();
std::cout << "score: " << row.score() << ", id: \"" << row.id() << "\", content: \"" << fields["content"].get_string() << "\"\n";
}
Below is a code snippet to execute a vector search query:
auto scope = cluster.bucket(config.bucket_name).scope(config.scope_name);
// weights could be retrieved from your llm model or openAI
std::vector<double> weights{ 0.1, 0.2, 0.3, 0.4 };
auto [err, resp] = scope
.search("travel-inventory-landmarks",
couchbase::search_request(couchbase::vector_search(
couchbase::vector_query(field_name, weights))))
.get();
for (const auto& row : resp.rows()) {
auto fields = row.fields_as<couchbase::codec::tao_json_serializer>();
std::cout << "score: " << row.score() << ", id: \"" << row.id() << "\", content: \"" << fields["content"].get_string() << "\"\n";
}
The SDK supports asynchronous programming models, enabling you to build responsive, non-blocking applications. This is particularly useful for high-throughput applications where maintaining responsiveness is essential. Here’s an example of running a transaction in a asynchronous context:
std::shared_ptr<async_attempt_context> ctx) -> couchbase::error {
ctx->get(
collection,
some_id,
// do something in this transaction
}
This new C++ SDK includes support for transactions and replaces our existing C++ transaction support by adding more performance improvements and features. Below is a sample code snippet for create a transaction context:
std::shared_ptr<couchbase::transactions::attempt_context> ctx) -> couchbase::error {
ctx.insert(collection, "doc-a", nlohmann::json({}));
couchbase::transactions::transaction_get_result doc_a = ctx->get( collection, “doc-a”)
}
Error handling is crucial in any application, and the C++ SDK, just like our other SDKs, provides comprehensive error handling capabilities including retry for connection drops, connection pooling and informative error messages. This ensures that you can gracefully handle and recover from errors, enhancing the stability and reliability of your applications.
To help you get started with the C++ SDK for Couchbase, we’ve prepared a detailed getting started guide on our documentation website. Here’s a quick overview of how to begin:
We believe in the power of community and open-source development. The C++ SDK for Couchbase is open-source, and we encourage you to contribute, provide feedback, and join the conversation. For support, if you are our enterprise licensed customer, you can reach out via support, otherwise, you can access our comprehensive documentation, join the Couchbase Forums or Couchbase Discord, or reach out via our support portal.
To learn more, check out our documentation website. It goes into more detail on the API, especially around transactions and asynchronous operations and provides other reference material and sample bindings links for you to dig deeper:
Supported operating systems are listed on our documentation website.
Happy coding!
The Couchbase Team
The post Announcing General Availability of the C++ SDK for Couchbase appeared first on The Couchbase Blog.
The above is the detailed content of Announcing General Availability of the C++ SDK for Couchbase. For more information, please follow other related articles on the PHP Chinese website!
![[Web front-end] Node.js quick start](https://img.php.cn/upload/course/000/000/067/662b5d34ba7c0227.png)
