Various helper utilities for working with buffers and binary data in TypeScript.
npm install @jsonjoy.com/buffers
This package provides high-performance utilities for working with binary data, buffers, and UTF-8 text encoding/decoding. It includes optimized implementations for both Node.js and browser environments.
A growable binary data writer with automatic buffer expansion.
import {Writer} from '@jsonjoy.com/buffers/lib/Writer';
const writer = new Writer();
writer.u8(0x42); // Write unsigned 8-bit integer
writer.u16(0x1234); // Write unsigned 16-bit integer
writer.u32(0x12345678); // Write unsigned 32-bit integer
writer.u64(0x123456789abcdefn); // Write unsigned 64-bit integer
writer.f32(3.14); // Write 32-bit float
writer.f64(3.141592653589793); // Write 64-bit float
writer.utf8('Hello π'); // Write UTF-8 string
writer.ascii('Hello'); // Write ASCII string
const data = writer.flush(); // Get written data as Uint8Array
A binary data reader for parsing binary buffers.
import {Reader} from '@jsonjoy.com/buffers/lib/Reader';
const reader = new Reader();
reader.reset(someUint8Array);
const byte = reader.u8(); // Read unsigned 8-bit integer
const word = reader.u16(); // Read unsigned 16-bit integer
const dword = reader.u32(); // Read unsigned 32-bit integer
const qword = reader.u64(); // Read unsigned 64-bit integer
const float = reader.f32(); // Read 32-bit float
const double = reader.f64(); // Read 64-bit float
const text = reader.utf8(5); // Read UTF-8 string of 5 bytes
const ascii = reader.ascii(5); // Read ASCII string of 5 characters
A streaming binary reader that can handle data arriving in chunks.
import {StreamingReader} from '@jsonjoy.com/buffers/lib/StreamingReader';
const reader = new StreamingReader();
reader.push(chunk1);
reader.push(chunk2);
// Read data as it becomes available
const value = reader.u32();
reader.consume(); // Mark consumed data for cleanup
A specialized streaming reader for byte-oriented protocols with optional XOR masking.
import {StreamingOctetReader} from '@jsonjoy.com/buffers/lib/StreamingOctetReader';
const reader = new StreamingOctetReader();
reader.push(dataChunk);
const byte = reader.u8();
const masked = reader.bufXor(length, [0x12, 0x34, 0x56, 0x78], 0);
// Array creation and manipulation
import {b} from '@jsonjoy.com/buffers/lib/b';
import {concat, concatList} from '@jsonjoy.com/buffers/lib/concat';
import {copy} from '@jsonjoy.com/buffers/lib/copy';
const buffer = b(0x48, 0x65, 0x6c, 0x6c, 0x6f); // Create from bytes
const combined = concat(buffer1, buffer2); // Concatenate two buffers
const list = concatList([buf1, buf2, buf3]); // Concatenate array of buffers
const duplicate = copy(originalBuffer); // Copy buffer
import {cmpUint8Array} from '@jsonjoy.com/buffers/lib/cmpUint8Array';
import {cmpUint8Array2} from '@jsonjoy.com/buffers/lib/cmpUint8Array2';
import {cmpUint8Array3} from '@jsonjoy.com/buffers/lib/cmpUint8Array3';
const isEqual = cmpUint8Array(buf1, buf2); // Returns boolean
const comparison = cmpUint8Array2(buf1, buf2); // Returns -1, 0, or 1 (byte-first)
const comparison2 = cmpUint8Array3(buf1, buf2); // Returns -1, 0, or 1 (length-first)
import {isUint8Array} from '@jsonjoy.com/buffers/lib/isUint8Array';
import {isArrayBuffer} from '@jsonjoy.com/buffers/lib/isArrayBuffer';
import {isFloat32} from '@jsonjoy.com/buffers/lib/isFloat32';
if (isUint8Array(data)) { /* data is Uint8Array or Buffer */ }
if (isArrayBuffer(data)) { /* data is ArrayBuffer */ }
if (isFloat32(3.14)) { /* number can fit in float32 */ }
import {toUint8Array} from '@jsonjoy.com/buffers/lib/toUint8Array';
import {bufferToUint8Array} from '@jsonjoy.com/buffers/lib/bufferToUint8Array';
import {toBuf} from '@jsonjoy.com/buffers/lib/toBuf';
const uint8 = toUint8Array(data); // Convert various types to Uint8Array
const converted = bufferToUint8Array(buf); // Convert Buffer to Uint8Array
const encoded = toBuf('Hello π'); // Convert string to UTF-8 bytes
import {ascii, utf8} from '@jsonjoy.com/buffers/lib/strings';
const asciiBytes = ascii`Hello World`; // ASCII string to bytes
const utf8Bytes = utf8`Hello π`; // UTF-8 string to bytes
import {decodeUtf8} from '@jsonjoy.com/buffers/lib/utf8/decodeUtf8';
const text = decodeUtf8(uint8Array, offset, length);
The package includes multiple optimized UTF-8 decoding implementations that automatically choose the best strategy based on:
- Environment (Node.js vs Browser)
- String length
- Available APIs
import {encode} from '@jsonjoy.com/buffers/lib/utf8/encode';
const bytesWritten = encode(targetArray, 'Hello π', offset, maxLength);
import {CachedUtf8Decoder} from '@jsonjoy.com/buffers/lib/utf8/CachedUtf8Decoder';
import {isUtf8} from '@jsonjoy.com/buffers/lib/utf8/isUtf8';
import {decodeAscii} from '@jsonjoy.com/buffers/lib/utf8/decodeAscii';
const decoder = new CachedUtf8Decoder();
const text = decoder.decode(uint8Array, start, length);
const isValidUtf8 = isUtf8(uint8Array);
const asciiText = decodeAscii(uint8Array, start, length);
A lightweight view into a buffer without copying data.
import {Slice} from '@jsonjoy.com/buffers/lib/Slice';
const slice = new Slice(uint8Array, dataView, start, end);
const subarray = slice.subarray(); // Get the actual data
import {decodeF16} from '@jsonjoy.com/buffers/lib/f16';
const float32Value = decodeF16(binaryF16Value);
import {printOctets} from '@jsonjoy.com/buffers/lib/printOctets';
console.log(printOctets(uint8Array, 16)); // Print hex dump of first 16 bytes
This library is designed for high performance with:
- Optimized UTF-8 handling: Multiple implementations that choose the fastest method for each environment
- Minimal allocations: Reusable readers and writers with buffer pooling
- Zero-copy operations: Slices and views avoid unnecessary data copying
- Environment-specific optimizations: Leverages Node.js Buffer APIs when available
Works in all modern browsers and Node.js environments. The library automatically detects available APIs and chooses the most appropriate implementation.
Full TypeScript support with comprehensive type definitions included.
Apache-2.0