mirrors/nodejs
mirrors/nodejs
0
0
Fork 0
mirror of https://github.com/nodejs/node.git synced 2024-12-01 16:10:02 +01:00
Code Issues Releases Activity
14c76f8671
nodejs/lib/internal/readline.js
James M Snell 72547fe28d readline: use icu based string width calculation 
Rather than the pseudo-wcwidth impl used currently, use the ICU
character properties database to calculate string width and
determine if a character is full width or not. This allows the
algorithm to correctly identify emoji's as full width, ensures
the algorithm will continue to fucntion properly as new unicode
codepoints are added, and it's faster.

This was originally part of a proposal to add a new unicode module,
but has been split out.

Refs: https://github.com/nodejs/node/pull/8075
PR-URL: https://github.com/nodejs/node/pull/9040
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: Steven R Loomis <srloomis@us.ibm.com>
2016-10-25 09:00:45 -07:00

411 lines
13 KiB
JavaScript
Raw Blame History

'use strict';
// Regex used for ansi escape code splitting
// eslint-disable-next-line no-control-regex
// Adopted from https://github.com/chalk/ansi-regex/blob/master/index.js
// License: MIT, authors: @sindresorhus, Qix-, and arjunmehta
// Matches all ansi escape code sequences in a string
const ansi =
/[\u001b\u009b][[()#;?]*(?:[0-9]{1,4}(?:;[0-9]{0,4})*)?[0-9A-ORZcf-nqry=><]/g;
module.exports = {
emitKeys,
stripVTControlCharacters
};
if (process.binding('config').hasIntl) {
const icu = process.binding('icu');
module.exports.getStringWidth = function getStringWidth(str, options) {
options = options || {};
if (!Number.isInteger(str))
str = stripVTControlCharacters(String(str));
return icu.getStringWidth(str,
Boolean(options.ambiguousAsFullWidth),
Boolean(options.expandEmojiSequence));
};
module.exports.isFullWidthCodePoint =
function isFullWidthCodePoint(code, options) {
if (typeof code !== 'number')
return false;
return icu.getStringWidth(code, options) === 2;
};
} else {
/**
* Returns the number of columns required to display the given string.
*/
module.exports.getStringWidth = function getStringWidth(str) {
if (Number.isInteger(str))
return module.exports.isFullWidthCodePoint(str) ? 2 : 1;
let width = 0;
str = stripVTControlCharacters(String(str));
for (var i = 0; i < str.length; i++) {
const code = str.codePointAt(i);
if (code >= 0x10000) { // surrogates
i++;
}
if (module.exports.isFullWidthCodePoint(code)) {
width += 2;
} else {
width++;
}
}
return width;
};
/**
* Returns true if the character represented by a given
* Unicode code point is full-width. Otherwise returns false.
*/
module.exports.isFullWidthCodePoint = function isFullWidthCodePoint(code) {
if (!Number.isInteger(code)) {
return false;
}
// Code points are derived from:
// http://www.unicode.org/Public/UNIDATA/EastAsianWidth.txt
if (code >= 0x1100 && (
code <= 0x115f || // Hangul Jamo
0x2329 === code || // LEFT-POINTING ANGLE BRACKET
0x232a === code || // RIGHT-POINTING ANGLE BRACKET
// CJK Radicals Supplement .. Enclosed CJK Letters and Months
(0x2e80 <= code && code <= 0x3247 && code !== 0x303f) ||
// Enclosed CJK Letters and Months .. CJK Unified Ideographs Extension A
0x3250 <= code && code <= 0x4dbf ||
// CJK Unified Ideographs .. Yi Radicals
0x4e00 <= code && code <= 0xa4c6 ||
// Hangul Jamo Extended-A
0xa960 <= code && code <= 0xa97c ||
// Hangul Syllables
0xac00 <= code && code <= 0xd7a3 ||
// CJK Compatibility Ideographs
0xf900 <= code && code <= 0xfaff ||
// Vertical Forms
0xfe10 <= code && code <= 0xfe19 ||
// CJK Compatibility Forms .. Small Form Variants
0xfe30 <= code && code <= 0xfe6b ||
// Halfwidth and Fullwidth Forms
0xff01 <= code && code <= 0xff60 ||
0xffe0 <= code && code <= 0xffe6 ||
// Kana Supplement
0x1b000 <= code && code <= 0x1b001 ||
// Enclosed Ideographic Supplement
0x1f200 <= code && code <= 0x1f251 ||
// CJK Unified Ideographs Extension B .. Tertiary Ideographic Plane
0x20000 <= code && code <= 0x3fffd)) {
return true;
}
return false;
};
}
/**
* Tries to remove all VT control characters. Use to estimate displayed
* string width. May be buggy due to not running a real state machine
*/
function stripVTControlCharacters(str) {
return str.replace(ansi, '');
}
/*
Some patterns seen in terminal key escape codes, derived from combos seen
at http://www.midnight-commander.org/browser/lib/tty/key.c
ESC letter
ESC [ letter
ESC [ modifier letter
ESC [ 1 ; modifier letter
ESC [ num char
ESC [ num ; modifier char
ESC O letter
ESC O modifier letter
ESC O 1 ; modifier letter
ESC N letter
ESC [ [ num ; modifier char
ESC [ [ 1 ; modifier letter
ESC ESC [ num char
ESC ESC O letter
- char is usually ~ but $ and ^ also happen with rxvt
- modifier is 1 +
(shift * 1) +
(left_alt * 2) +
(ctrl * 4) +
(right_alt * 8)
- two leading ESCs apparently mean the same as one leading ESC
*/
function* emitKeys(stream) {
while (true) {
let ch = yield;
let s = ch;
let escaped = false;
const key = {
sequence: null,
name: undefined,
ctrl: false,
meta: false,
shift: false
};
if (ch === '\x1b') {
escaped = true;
s += (ch = yield);
if (ch === '\x1b') {
s += (ch = yield);
}
}
if (escaped && (ch === 'O' || ch === '[')) {
// ansi escape sequence
let code = ch;
let modifier = 0;
if (ch === 'O') {
// ESC O letter
// ESC O modifier letter
s += (ch = yield);
if (ch >= '0' && ch <= '9') {
modifier = (ch >> 0) - 1;
s += (ch = yield);
}
code += ch;
} else if (ch === '[') {
// ESC [ letter
// ESC [ modifier letter
// ESC [ [ modifier letter
// ESC [ [ num char
s += (ch = yield);
if (ch === '[') {
// \x1b[[A
// ^--- escape codes might have a second bracket
code += ch;
s += (ch = yield);
}
/*
* Here and later we try to buffer just enough data to get
* a complete ascii sequence.
*
* We have basically two classes of ascii characters to process:
*
*
* 1. `\x1b[24;5~` should be parsed as { code: '[24~', modifier: 5 }
*
* This particular example is featuring Ctrl+F12 in xterm.
*
* - `;5` part is optional, e.g. it could be `\x1b[24~`
* - first part can contain one or two digits
*
* So the generic regexp is like /^\d\d?(;\d)?[~^$]$/
*
*
* 2. `\x1b[1;5H` should be parsed as { code: '[H', modifier: 5 }
*
* This particular example is featuring Ctrl+Home in xterm.
*
* - `1;5` part is optional, e.g. it could be `\x1b[H`
* - `1;` part is optional, e.g. it could be `\x1b[5H`
*
* So the generic regexp is like /^((\d;)?\d)?[A-Za-z]$/
*
*/
const cmdStart = s.length - 1;
// skip one or two leading digits
if (ch >= '0' && ch <= '9') {
s += (ch = yield);
if (ch >= '0' && ch <= '9') {
s += (ch = yield);
}
}
// skip modifier
if (ch === ';') {
s += (ch = yield);
if (ch >= '0' && ch <= '9') {
s += (ch = yield);
}
}
/*
* We buffered enough data, now trying to extract code
* and modifier from it
*/
const cmd = s.slice(cmdStart);
let match;
if ((match = cmd.match(/^(\d\d?)(;(\d))?([~^$])$/))) {
code += match[1] + match[4];
modifier = (match[3] || 1) - 1;
} else if ((match = cmd.match(/^((\d;)?(\d))?([A-Za-z])$/))) {
code += match[4];
modifier = (match[3] || 1) - 1;
} else {
code += cmd;
}
}
// Parse the key modifier
key.ctrl = !!(modifier & 4);
key.meta = !!(modifier & 10);
key.shift = !!(modifier & 1);
key.code = code;
// Parse the key itself
switch (code) {
/* xterm/gnome ESC O letter */
case 'OP': key.name = 'f1'; break;
case 'OQ': key.name = 'f2'; break;
case 'OR': key.name = 'f3'; break;
case 'OS': key.name = 'f4'; break;
/* xterm/rxvt ESC [ number ~ */
case '[11~': key.name = 'f1'; break;
case '[12~': key.name = 'f2'; break;
case '[13~': key.name = 'f3'; break;
case '[14~': key.name = 'f4'; break;
/* from Cygwin and used in libuv */
case '[[A': key.name = 'f1'; break;
case '[[B': key.name = 'f2'; break;
case '[[C': key.name = 'f3'; break;
case '[[D': key.name = 'f4'; break;
case '[[E': key.name = 'f5'; break;
/* common */
case '[15~': key.name = 'f5'; break;
case '[17~': key.name = 'f6'; break;
case '[18~': key.name = 'f7'; break;
case '[19~': key.name = 'f8'; break;
case '[20~': key.name = 'f9'; break;
case '[21~': key.name = 'f10'; break;
case '[23~': key.name = 'f11'; break;
case '[24~': key.name = 'f12'; break;
/* xterm ESC [ letter */
case '[A': key.name = 'up'; break;
case '[B': key.name = 'down'; break;
case '[C': key.name = 'right'; break;
case '[D': key.name = 'left'; break;
case '[E': key.name = 'clear'; break;
case '[F': key.name = 'end'; break;
case '[H': key.name = 'home'; break;
/* xterm/gnome ESC O letter */
case 'OA': key.name = 'up'; break;
case 'OB': key.name = 'down'; break;
case 'OC': key.name = 'right'; break;
case 'OD': key.name = 'left'; break;
case 'OE': key.name = 'clear'; break;
case 'OF': key.name = 'end'; break;
case 'OH': key.name = 'home'; break;
/* xterm/rxvt ESC [ number ~ */
case '[1~': key.name = 'home'; break;
case '[2~': key.name = 'insert'; break;
case '[3~': key.name = 'delete'; break;
case '[4~': key.name = 'end'; break;
case '[5~': key.name = 'pageup'; break;
case '[6~': key.name = 'pagedown'; break;
/* putty */
case '[[5~': key.name = 'pageup'; break;
case '[[6~': key.name = 'pagedown'; break;
/* rxvt */
case '[7~': key.name = 'home'; break;
case '[8~': key.name = 'end'; break;
/* rxvt keys with modifiers */
case '[a': key.name = 'up'; key.shift = true; break;
case '[b': key.name = 'down'; key.shift = true; break;
case '[c': key.name = 'right'; key.shift = true; break;
case '[d': key.name = 'left'; key.shift = true; break;
case '[e': key.name = 'clear'; key.shift = true; break;
case '[2$': key.name = 'insert'; key.shift = true; break;
case '[3$': key.name = 'delete'; key.shift = true; break;
case '[5$': key.name = 'pageup'; key.shift = true; break;
case '[6$': key.name = 'pagedown'; key.shift = true; break;
case '[7$': key.name = 'home'; key.shift = true; break;
case '[8$': key.name = 'end'; key.shift = true; break;
case 'Oa': key.name = 'up'; key.ctrl = true; break;
case 'Ob': key.name = 'down'; key.ctrl = true; break;
case 'Oc': key.name = 'right'; key.ctrl = true; break;
case 'Od': key.name = 'left'; key.ctrl = true; break;
case 'Oe': key.name = 'clear'; key.ctrl = true; break;
case '[2^': key.name = 'insert'; key.ctrl = true; break;
case '[3^': key.name = 'delete'; key.ctrl = true; break;
case '[5^': key.name = 'pageup'; key.ctrl = true; break;
case '[6^': key.name = 'pagedown'; key.ctrl = true; break;
case '[7^': key.name = 'home'; key.ctrl = true; break;
case '[8^': key.name = 'end'; key.ctrl = true; break;
/* misc. */
case '[Z': key.name = 'tab'; key.shift = true; break;
default: key.name = 'undefined'; break;
}
} else if (ch === '\r') {
// carriage return
key.name = 'return';
} else if (ch === '\n') {
// enter, should have been called linefeed
key.name = 'enter';
} else if (ch === '\t') {
// tab
key.name = 'tab';
} else if (ch === '\b' || ch === '\x7f') {
// backspace or ctrl+h
key.name = 'backspace';
key.meta = escaped;
} else if (ch === '\x1b') {
// escape key
key.name = 'escape';
key.meta = escaped;
} else if (ch === ' ') {
key.name = 'space';
key.meta = escaped;
} else if (!escaped && ch <= '\x1a') {
// ctrl+letter
key.name = String.fromCharCode(ch.charCodeAt(0) + 'a'.charCodeAt(0) - 1);
key.ctrl = true;
} else if (/^[0-9A-Za-z]$/.test(ch)) {
// letter, number, shift+letter
key.name = ch.toLowerCase();
key.shift = /^[A-Z]$/.test(ch);
key.meta = escaped;
} else if (escaped) {
// Escape sequence timeout
key.name = ch.length ? undefined : 'escape';
key.meta = true;
}
key.sequence = s;
if (s.length !== 0 && (key.name !== undefined || escaped)) {
/* Named character or sequence */
stream.emit('keypress', escaped ? undefined : s, key);
} else if (s.length === 1) {
/* Single unnamed character, e.g. "." */
stream.emit('keypress', s, key);
}
/* Unrecognized or broken escape sequence, don't emit anything */
}
}
View Git Blame Copy Permalink