move utf32ToString to TextDecoder.ts

This commit is contained in:
Jörg Breitbart
2019-01-25 17:35:12 +01:00
parent 21165a4286
commit 4855b60ffe
7 changed files with 86 additions and 93 deletions
+1 -1
View File
@@ -6,7 +6,7 @@
import { ParserState, ParserAction, IParsingState, IDcsHandler, IEscapeSequenceParser } from './Types';
import { IDisposable } from 'xterm';
import { Disposable } from './common/Lifecycle';
import { utf32ToString } from './common/TypedArrayUtils';
import { utf32ToString } from './core/input/TextDecoder';
interface IHandlerCollection<T> {
[key: string]: T[];
+2 -2
View File
@@ -14,8 +14,8 @@ import { EscapeSequenceParser } from './EscapeSequenceParser';
import { ICharset } from './core/Types';
import { IDisposable } from 'xterm';
import { Disposable } from './common/Lifecycle';
import { concat, utf32ToString } from './common/TypedArrayUtils';
import { StringToUtf32, stringFromCodePoint } from './core/input/TextDecoder';
import { concat } from './common/TypedArrayUtils';
import { StringToUtf32, stringFromCodePoint, utf32ToString } from './core/input/TextDecoder';
/**
* Map collect to glevel. Used in `selectCharset`.
+1 -9
View File
@@ -3,7 +3,7 @@
* @license MIT
*/
import { assert } from 'chai';
import { fillFallback, concat, utf32ToString } from './TypedArrayUtils';
import { fillFallback, concat } from './TypedArrayUtils';
type TypedArray = Uint8Array | Uint16Array | Uint32Array | Uint8ClampedArray
| Int8Array | Int16Array | Int32Array
@@ -94,12 +94,4 @@ describe('typed array convenience functions', () => {
const merged = concat(a, b);
deepEquals(merged, new Uint8Array([1, 2, 3, 4, 5, 6, 7, 8, 9, 0]));
});
it('utf16ToString', () => {
const s = 'abcdefg';
const data = new Uint16Array(s.length);
for (let i = 0; i < s.length; ++i) {
data[i] = s.charCodeAt(i);
}
assert.equal(utf32ToString(data), s);
});
});
+1 -21
View File
@@ -2,10 +2,7 @@
* Copyright (c) 2018 The xterm.js authors. All rights reserved.
* @license MIT
*/
type TypedArray = Uint8Array | Uint16Array | Uint32Array | Uint8ClampedArray
| Int8Array | Int16Array | Int32Array
| Float32Array | Float64Array;
import { TypedArray } from './Types';
/**
* polyfill for TypedArray.fill
@@ -49,20 +46,3 @@ export function concat<T extends TypedArray>(a: T, b: T): T {
result.set(b, a.length);
return result;
}
/**
* Convert UTF32 char codes into JS string.
*/
export function utf32ToString<T extends TypedArray>(data: T, start: number = 0, end: number = data.length): string {
let result = '';
let cp;
for (let i = start; i < end; ++i) {
if ((cp = data[i]) > 0xFFFF) {
cp -= 0x10000;
result += String.fromCharCode((cp >> 10) + 0xD800) + String.fromCharCode((cp % 0x400) + 0xDC00);
} else {
result += String.fromCharCode(cp);
}
}
return result;
}
+4
View File
@@ -35,3 +35,7 @@ export interface ICircularList<T> extends IEventEmitter {
trimStart(count: number): void;
shiftElements(start: number, count: number, offset: number): void;
}
export type TypedArray = Uint8Array | Uint16Array | Uint32Array | Uint8ClampedArray
| Int8Array | Int16Array | Int32Array
| Float32Array | Float64Array;
+56 -56
View File
@@ -4,64 +4,64 @@
*/
import { assert } from 'chai';
import { StringToUtf32, stringFromCodePoint } from './TextDecoder';
import { StringToUtf32, stringFromCodePoint, utf32ToString } from './TextDecoder';
// convert UTF32 codepoints to string
function toString(data: Uint32Array, length: number): string {
if ((String as any).fromCodePoint) {
return (String as any).fromCodePoint.apply(null, data.subarray(0, length));
}
let result = '';
for (let i = 0; i < length; ++i) {
result += stringFromCodePoint(data[i]);
}
return result;
}
describe('StringToUtf32 Decoder', () => {
describe('full codepoint test', () => {
it('0..65535', () => {
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
for (let i = 0; i < 65536; ++i) {
// skip surrogate pairs
if (i >= 0xD800 && i <= 0xDFFF) {
continue;
}
const length = decoder.decode(String.fromCharCode(i), target);
assert.equal(length, 1);
assert.equal(target[0], i);
assert.equal(toString(target, length), String.fromCharCode(i));
decoder.clear();
}
});
it('65536..0x10FFFF (surrogates)', function(): void {
this.timeout(20000);
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
for (let i = 65536; i < 0x10FFFF; ++i) {
const codePoint = i - 0x10000;
const s = String.fromCharCode((codePoint >> 10) + 0xD800) + String.fromCharCode((codePoint % 0x400) + 0xDC00);
const length = decoder.decode(s, target);
assert.equal(length, 1);
assert.equal(target[0], i);
assert.equal(toString(target, length), s);
decoder.clear();
}
});
describe('text encodings', () => {
it('stringFromCodePoint/utf32ToString', () => {
const s = 'abcdefg';
const data = new Uint32Array(s.length);
for (let i = 0; i < s.length; ++i) {
data[i] = s.charCodeAt(i);
assert.equal(stringFromCodePoint(data[i]), s[i]);
}
assert.equal(utf32ToString(data), s);
});
describe('stream handling', () => {
it('surrogates mixed advance by 1', () => {
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
const input = 'Ä€𝄞Ö𝄞€Ü𝄞€';
let decoded = '';
for (let i = 0; i < input.length; ++i) {
const written = decoder.decode(input[i], target);
decoded += toString(target, written);
}
assert(decoded, 'Ä€𝄞Ö𝄞€Ü𝄞€');
describe('StringToUtf32 Decoder', () => {
describe('full codepoint test', () => {
it('0..65535', () => {
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
for (let i = 0; i < 65536; ++i) {
// skip surrogate pairs
if (i >= 0xD800 && i <= 0xDFFF) {
continue;
}
const length = decoder.decode(String.fromCharCode(i), target);
assert.equal(length, 1);
assert.equal(target[0], i);
assert.equal(utf32ToString(target, 0, length), String.fromCharCode(i));
decoder.clear();
}
});
it('65536..0x10FFFF (surrogates)', function(): void {
this.timeout(20000);
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
for (let i = 65536; i < 0x10FFFF; ++i) {
const codePoint = i - 0x10000;
const s = String.fromCharCode((codePoint >> 10) + 0xD800) + String.fromCharCode((codePoint % 0x400) + 0xDC00);
const length = decoder.decode(s, target);
assert.equal(length, 1);
assert.equal(target[0], i);
assert.equal(utf32ToString(target, 0, length), s);
decoder.clear();
}
});
});
describe('stream handling', () => {
it('surrogates mixed advance by 1', () => {
const decoder = new StringToUtf32();
const target = new Uint32Array(5);
const input = 'Ä€𝄞Ö𝄞€Ü𝄞€';
let decoded = '';
for (let i = 0; i < input.length; ++i) {
const written = decoder.decode(input[i], target);
decoded += utf32ToString(target, written);
}
assert(decoded, 'Ä€𝄞Ö𝄞€Ü𝄞€');
});
});
});
});
+21 -4
View File
@@ -2,6 +2,7 @@
* Copyright (c) 2019 The xterm.js authors. All rights reserved.
* @license MIT
*/
import { TypedArray } from '../../common/Types';
/**
* StringToUtf32 - decodes UTF16 sequences into UTF32 codepoints.
@@ -73,15 +74,31 @@ export class StringToUtf32 {
}
/**
* Polyfill - Convert UTF32 codepoint into JS string.
* Convert UTF32 codepoint into JS string.
*/
export function stringFromCodePoint(codePoint: number): string {
if ((String as any).fromCodePoint) {
return (String as any).fromCodePoint(codePoint);
}
if (codePoint > 0xFFFF) {
codePoint -= 0x10000;
return String.fromCharCode((codePoint >> 10) + 0xD800) + String.fromCharCode((codePoint % 0x400) + 0xDC00);
}
return String.fromCharCode(codePoint);
}
/**
* Convert UTF32 char codes into JS string.
* Basically the same as `stringFromCodePoint` but for multiple codepoints
* in a loop (which is a lot faster).
*/
export function utf32ToString<T extends TypedArray>(data: T, start: number = 0, end: number = data.length): string {
let result = '';
let cp;
for (let i = start; i < end; ++i) {
if ((cp = data[i]) > 0xFFFF) {
cp -= 0x10000;
result += String.fromCharCode((cp >> 10) + 0xD800) + String.fromCharCode((cp % 0x400) + 0xDC00);
} else {
result += String.fromCharCode(cp);
}
}
return result;
}