Bitcoin Descriptor Blockchain Retrieval Library

The @bitcoinerlab/discovery library, written in TypeScript, provides a method for retrieving blockchain information essential in Bitcoin wallet development and other applications that require interaction with the Bitcoin blockchain. This library enables querying of data using the Bitcoin descriptors syntax, facilitating retrieval of blockchain data such as balance, UTXOs, and transaction history.

Features

• Descriptor-Based Data Retrieval: Retrieves transaction history for various sources, including: ranged descriptors, accounts (comprising internal & external descriptors), and addresses (a descriptor specialized for a specific index).

• Transaction Status Filter: Offers the ability to filter results by TxStatus: ALL (including transactions in the mempool), CONFIRMED (assuming one confirmation) and IRREVERSIBLE (for transactions with more than a user-defined number of confirmations).

• Pluggable Explorer Interface: Implements a plugin interface that separates data storage and data derivation from blockchain information retrieval, referred to as the Explorer interface. This design enables easy addition of various network explorers. Esplora and Electrum Explorer implementations have already been implemented. The slim Explorer interface makes it easy to add others, such as a Bitcoin Node implementation (planned for a future release).

• Immutability Principles: The library revolves around Immutability, which allows for quick comparisons of objects. For instance, you can retrieve the UTXOs of a certain group of ranged descriptors filtered by a particular TxStatus. The library will return the same reference if the result did not change, reducing the need for deep comparisons of complex data arrays. This feature greatly simplifies developing applications using reactive rendering engines such as React.

• Data Derivation: The library maintains a compact internal structure of information and provides methods to query and derive useful data from it. For example, Balances and UTXOs are computed on-the-fly from raw data, while advanced memoization techniques ensure efficiency. This compact data model allows the library to focus on retrieving and storing only transaction data, eliminating the need to download and keep balances or UTXOs in sync.

Usage

To get started, follow the steps below:

1. Install the Libraries:

   npm install @bitcoinerlab/explorer @bitcoinerlab/discovery bitcoinjs-lib

2. Create the Explorer Client Instance: The explorer client is an interface to communicate with the blockchain. You can create an instance to an Electrum server or an Esplora server.

   • Esplora is a Blockstream-developed open-source explorer for the Bitcoin blockchain. It offers an HTTP REST API that allows interaction with the Bitcoin blockchain.

   • Electrum is a popular Bitcoin wallet that also provides server software (Electrum server) that facilitates communication between clients and the Bitcoin network. The Electrum server uses a different protocol than the standard Bitcoin protocol used by Esplora.

   The @bitcoinerlab/explorer library provides two classes for creating the explorer client instances: EsploraExplorer and ElectrumExplorer.

   import { EsploraExplorer, ElectrumExplorer } from '@bitcoinerlab/explorer';
   import { networks } from 'bitcoinjs-lib';
   const esploraExplorer = new EsploraExplorer({
     url: 'https://blockstream.info/api'
   });
   const electrumExplorer = new ElectrumExplorer({
     host: 'electrum.blockstream.info',
     port: 60002,
     protocol: 'ssl', // 'ssl' and 'tcp' allowed
     network: networks.testnet // Specify the server's network; defaults to networks.bitcoin (mainnet)
   });

   In the code snippet, we create instances for both an Electrum client and an Esplora client using the ElectrumExplorer and EsploraExplorer classes, respectively.

   Please refer to the Explorer documentation for more details.

3. Create the Discovery Class: After creating the explorer client instance, you can create the Discovery class, which you will use to query the Blockchain. The Discovery class is created using the DiscoveryFactory function, passing the previously created explorer instance.

   import { DiscoveryFactory } from '@bitcoinerlab/discovery';
   const { Discovery } = DiscoveryFactory(explorer, network);
   // where 'explorer' corresponds to 'esploraExplorer' or 'electrumExplorer' above
   await explorer.connect();
   const discovery = new Discovery();
   // Perform discovery operations...
   explorer.close();

   The Discovery constructor, new Discovery({ descriptorsCacheSize, outputsPerDescriptorCacheSize }), accepts an optional object with two properties that are crucial for managing the application's memory usage:

   • descriptorsCacheSize: This property represents the cache size limit for descriptor expressions. The cache, implemented using memoizers, serves a dual purpose: it speeds up data derivation by avoiding unnecessary recomputations, and it helps maintain immutability. Reaching the limit of the cache size may lead to a loss of immutability and the returned reference may change. This is not a critical issue, as the returned data is still correct, but it may trigger extra renders in the UI. The default value is 1000, and you can set it to 0 for unbounded caches.
   • outputsPerDescriptorCacheSize: This property represents the cache size limit for indices per expression, related to the number of addresses in ranged descriptor expressions. Similar to the descriptorsCacheSize, reaching the limit of this cache size may lead to the same immutability challenges. The default value is 10000, and you can set it to 0 for unbounded caches.

   It's noteworthy that the default settings for descriptorsCacheSize and outputsPerDescriptorCacheSize are adequate for most use cases. Yet, for projects handling a large volume of descriptor expressions or addresses, increasing these limits may be necessary. On the flip side, if conserving memory is a priority, particularly for projects with minimal resource needs, consider lowering these values.

   Note: The connect method must be run before starting any data queries to the blockchain, and the close method should be run after you have completed all necessary queries and no longer need to query the blockchain.

4. Using the Discovery Methods

   Once you've instantiated the Discovery class, you have access to a variety of methods to fetch and derive blockchain data from Bitcoin Output Descriptors.

   Descriptor expressions are a simple language used to describe collections of Bitcoin output scripts. They enable the Discovery class to fetch detailed blockchain information about specific outputs. For more comprehensive insights into descriptor expressions, refer to the BitcoinerLab descriptors module.

   To initiate (or update) the data retrieval process for addresses associated with a descriptor, whether ranged or fixed, execute fetch:

   await discovery.fetch({ descriptor });

   This method retrieves all associated outputs for a given descriptor. If the descriptor is ranged, you can also specify an index to target a specific output within that range. When dealing with multiple descriptors, use the descriptors parameter with an array of strings. See the fetch API documentation for detailed usage.

   Note: To ensure accurate data computations, fetch descriptor data (using the query above) before employing methods like getUtxos, getBalance, or others described below. An error will alert you when attempting to derive data from descriptors that have not been previously fetched. This ensures you do not compute data based on incomplete information. If you are unsure whether a descriptor has been previously fetched or need to ensure that the data is up-to-date, use whenFetched:

   const fetchStatus = discovery.whenFetched({ descriptor });
   if (fetchStatus === undefined) {
     // The descriptor has not been fetched.
   } else {
     const secondsSinceFetched = (Date.now() - fetchStatus.timeFetched * 1000) / 1000;
     if (secondsSinceFetched > SOME_TIME_THRESHOLD) {
       // The descriptor data is outdated and may need to be fetched again.
     }
   }

   If fetch status is verified or known, proceed directly to the data derivation methods:

   • Deriving UTXOs: Use getUtxos to derive all unspent transaction outputs (UTXOs) from the fetched data:

     const { utxos } = discovery.getUtxos({ descriptor });

   • Calculating Balance: Use getBalance to calculate the total balance from the fetched data:

     const { balance } = discovery.getBalance({ descriptor });

   Other methods to derive or calculate data include:

   • Determining the Next Index: For ranged descriptor expressions, determine the next unused index:

     const index = discovery.getNextIndex({ descriptor });

     See the getNextIndex API documentation for detailed usage.

   • Identifying Descriptors by UTXO: Find the descriptor that corresponds to a specific UTXO using getDescriptor:

     const descriptorData = discovery.getDescriptor({ utxo });
     // Returns: { descriptor, index? }, with 'index' provided for ranged descriptors.

     This is particularly useful for transaction preparation when you need to instantiate a new Output({ descriptor }) using the descriptor associated with the UTXO, as facilitated by the @bitcoinerlab/descriptors library.

   • Accessing Transaction History: Access all transactions associated with a specific descriptor expression (or an array of them):

     const history = discovery.getHistory({ descriptors });

     Refer to the getHistory API for the details.

   • Fetching Standard Accounts: The fetchStandardAccounts method is a helper that automates the common task of retrieving or updating standard accounts (pkh, sh(wpkh), wpkh) associated with a master node. This method saves developers time and eliminates repetitive coding tasks.

     Efficiently retrieve wallet accounts with:

     await discovery.fetchStandardAccounts({
       masterNode,
       gapLimit: 20, // The default gap limit
       onAccountUsed: (account) => {
         // Optional: Trigger app updates when an account with transactions is found.
       },
       onAccountChecking: (account) => {
         // Optional: Implement app-specific logic when the check for an account begins.
       }
     });

     Implement the onAccountUsed and onAccountChecking callbacks as needed for your app's functionality, such as UI updates or logging.

   The methods listed above are only a part of all the Discovery class's functionality. For a complete overview of all available methods and their usage, refer to the API documentation.

API Documentation

To generate the API documentation for this module, you can run the following command:

npm run docs

However, if you'd prefer to skip this step, the API documentation has already been compiled and is available for reference at bitcoinerlab.com/modules/discovery/api.

Authors and Contributors

The project was initially developed and is currently maintained by Jose-Luis Landabaso. Contributions and help from other developers are welcome.

Here are some resources to help you get started with contributing:

Building from source

To download the source code and build the project, follow these steps:

1. Clone the repository:

git clone https://github.com/bitcoinerlab/discovery.git

2. Install the dependencies:

npm install

3. Build the project:

npm run build

This will build the project and generate the necessary files in the dist directory.

Testing

Before finalizing and committing your code, it's essential to make sure all tests are successful. To run these tests:

1. A Bitcoin regtest node must be active.
2. Utilize the Express-based bitcoind manager which should be operational at 127.0.0.1:8080.
3. An Electrum server and an Esplora server are required, both indexing the regtest node.

To streamline this setup, you can use the Docker image, bitcoinerlab/tester, which comes preconfigured with the required services. The Docker image can be found under Dockerfile for bitcoinerlab/tester. When you run the test script using:

npm test

it will automatically download and start the Docker image if it's not already present on your machine. However, ensure you have the docker binary available in your path for this to work seamlessly.

License

This project is licensed under the MIT License.