nv-buf-utf8-ab

• get/decode code-points directly from valid-utf8-array-buffer
• similiar to for(let ch of str) { yield(ch.codePointAt(0))} , just from array-buffer, NOT from string
• this pkg just rewrite v8/src/third_party/utf8-decoder to js.
• just for test, its slow than c version

install

• npm install nv-buf-utf8-ab

splitted

usage

  const { creat_mach, ab2cds }   = require("nv-buf-utf8-ab");

example

    var buf = Buffer.from("aÿ我𝑒")
    /*
     > buf
     <Buffer 61 c3 bf e6 88 91 f0 9d 91 92>
    */

one-by-one (fast)

    var mach = x.creat_mach()
    /*
        > mach.input(0x61)
        97
        > mach.input(0xc3)
        undefined
        > mach.input(0xbf)
        255
        > mach.input(0xe6)
        undefined
        > mach.input(0x88)
        undefined
        > mach.input(0x91)
        25105
        > mach.input(0xf0)
        undefined
        > mach.input(0x9d)
        undefined
        > mach.input(0x91)
        undefined
        > mach.input(0x92)
        119890
        > 
    */

get detail description (slow)

        var arr =[
           97, 195, 191, 230,
          136, 145, 240, 157,
          145, 146
        ]

        var u8a = new Uint8Array(arr)
        var ab = u8a.buffer




        var mach = creat_mach()

        var f1 = (ab)=>x.ab2cds(ab,mach);

        console.dir(f1(ab))
        /*
        {
          codes:   [ 97, 255, 25105, 119890 ],
          offsets: [ 0, 1, 3, 6 ],               // byte index  of each-utf8-char
          sizes:   [ 1, 2, 3, 4 ],               // byte length of each-utf8-char 
        }
         */

METHODS

APIS

LICENSE

• ISC