npm

Need private packages and team management tools?Check out npm Orgs. »

@thi.ng/transducers-binary

0.4.1 • Public • Published

@thi.ng/transducers-binary

npm (scoped) npm downloads Twitter Follow

This project is part of the @thi.ng/umbrella monorepo.

About

Binary data related transducers & reducers. Like the transducers and reducers defined in @thi.ng/transducers, all functions defined in this package too accept an optional input iterable for direct use.

Installation

yarn add @thi.ng/transducers-binary

Dependencies

Usage examples

import * as tx from "@thi.ng/transducers";
import * as txb from "@thi.ng/transducers-binary";

Random bits

// 10 samples with 50% probability of drawing a 1
[...txb.randomBits(0.5, 10)]
// [ 1, 0, 1, 1, 0, 1, 0, 1, 1, 0 ]
 
// infinite iterator without 2nd arg, so limit with `take()`
[...tx.take(10, txb.randomBits(0.1))]
// [ 0, 0, 0, 0, 0, 0, 1, 0, 0, 0 ]
 
import { Smush32 } from "@thi.ng/random";
 
// with seeded PRNG
[...txb.randomBits(0.5, 10, new Smush32(12345678))]
// [ 0, 0, 1, 1, 0, 0, 0, 0, 1, 0 ]

Streaming hexdump

This is a higher-order transducer, purely composed from other transducers. See code here.

src = [65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 33, 48, 49, 50, 51, 126, 122, 121, 120]
 
[...txb.hexDump({ cols: 8, address: 0x100 }, src)]
// [ '00000100 | 41 42 43 44 45 46 47 48 | ABCDEFGH',
//   '00000108 | 49 4a 21 30 31 32 33 7e | IJ!0123~',
//   '00000110 | 7a 79 78 00 00 00 00 00 | zyx.....' ]

Structured byte buffer construction

The bytes() reducer transforms a stream of declarative data definitions (optionally with Little-Endian encoding) into an Uint8Array.

const bytes = txb.bytes(
    // initial buffer capacity (grows on demand)
    32,
    // structured data
    [
        // default order is Big-Endian
        txb.u32(0xdecafbad),
        // force Little-endian (also works for floats)
        txb.u32(0x44332211, true),
        // all strings will be utf-8 encoded
        txb.str("vec4"),
        // use little-endian for each of these array vals
        txb.f32array([1, 2, 3, 4], true),
        txb.u8(0x2a)
    ]
);
 
console.log(tx.str("\n", txb.hexDump({}, bytes)));
 
// 00000000 | de ca fb ad 11 22 33 44 76 65 63 34 00 00 80 3f | ....."3Dvec4...?
// 00000010 | 00 00 00 40 00 00 40 40 00 00 80 40 2a 00 00 00 | ...@..@@...@*...

Bitstream

Decompose / transform a stream of fixed size words into their bits:

[...txb.bits(8, [0xf0, 0xaa])];
// [ 1, 1, 1, 1, 0, 0, 0, 0, 1, 0, 1, 0, 1, 0, 1, 0 ]
 
console.log(
    tx.transduce(
        tx.comp(
            txb.bits(8),
            tx.map((x) => (x ? "#" : ".")),
            tx.partition(8),
            tx.map((x) => x.join(""))
        ),
        tx.str("\n"),
        [0x00, 0x18, 0x3c, 0x66, 0x66, 0x7e, 0x66, 0x00]
    )
);
// ........
// ...##...
// ..####..
// .##..##.
// .##..##.
// .######.
// .##..##.
// ........

Extended to transform longer strings (taken from the bitmap-font example, live demo):

// font lookup table
const chars = {
    a: [0x00, 0x18, 0x3c, 0x66, 0x66, 0x7e, 0x66, 0x00],
    b: [0x00, 0x7c, 0x66, 0x7c, 0x66, 0x66, 0x7c, 0x00]
};
 
// re-usable transducer
const xfJoin = tx.map((x) => x.join(""));
 
// higher order transducer to transform single char from string
const xfChar = (i) =>
    tx.comp(
        tx.pluck(i),
        txb.bits(8),
        tx.map((x) => (x ? "#" : ".")),
        tx.partition(8),
        xfJoin
    );
 
// transform entire string
const banner = (src) =>
    tx.transduce(
        tx.comp(
            // dynamically create `xfChar` transducers for each char
            // and run them in parallel via `multiplex()`
            tx.multiplex(...tx.map((i) => xfChar(i), tx.range(src.length))),
            // then join the results for each line
            xfJoin
        ),
        // use `str()` reducer to build string result
        tx.str("\n"),
        // convert input string into stream of row-major bitmap font tuples
        tx.zip(...tx.map((x) => chars[x], src))
    );
 
console.log(banner("abba"));
// ................................
// ...##....#####...#####.....##...
// ..####...##..##..##..##...####..
// .##..##..#####...#####...##..##.
// .##..##..##..##..##..##..##..##.
// .######..##..##..##..##..######.
// .##..##..#####...#####...##..##.
// ................................

Base64 & UTF-8 en/decoding

Unlike JS default btoa() / atob() functions which operate on strings, these transducers stepwise convert byte values to base64 and back.

// here we first add an offset (0x80) to allow negative values to be encoded
// (URL safe results can be produced via opt arg to `base64Encode`)
enc = tx.transduce(
    tx.comp(tx.map((x) => x + 0x80), txb.base64Encode()),
    tx.str(),
    tx.range(-8, 8)
);
// "eHl6e3x9fn+AgYKDhIWGhw=="
 
// remove offset again during decoding, but (for example) only decode while val < 0
[
    ...tx.iterator(
        tx.comp(
            txb.base64Decode(),
            tx.map((x) => x - 0x80),
            tx.takeWhile((x) => x < 0)
        ),
        enc
    )
];
// [ -8, -7, -6, -5, -4, -3, -2, -1 ]
 
buf = tx.transduce(
    tx.comp(txb.utf8Encode(), txb.base64Encode()),
    tx.str(),
    "beer (🍺) or hot beverage (☕️)"
);
// "YmVlciAo8J+Nuikgb3IgaG90IGJldmVyYWdlICjimJXvuI4p"
 
tx.transduce(tx.comp(txb.base64Decode(), txb.utf8Decode()), tx.str(), buf);
// "beer (🍺) or hot beverage (☕️)"

API

Transducers

Reducers

Authors

  • Karsten Schmidt

License

© 2018 Karsten Schmidt // Apache Software License 2.0

install

npm i @thi.ng/transducers-binary

Downloadsweekly downloads

47

version

0.4.1

license

Apache-2.0

homepage

github.com

repository

Gitgithub

last publish

collaborators

  • avatar
Report a vulnerability