erabo/spiegel-keyman
1
0
Fork 0
mirror of https://github.com/keymanapp/keyman.git synced 2026-08-05 00:15:32 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 9

feat(developer): remove convet_utils and convert_utils.tests

Browse source
This commit is contained in:
Sabine 2026-08-04 18:01:02 +02:00
parent 0934b20f2b
commit 91aaae0e32
4 changed files with 1 additions and 248 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
common/web/types/src/main.ts
View file

@ -19,8 +19,6 @@ export * as KmpJsonFile from './package/kmp-json-file.js';
export { Uni_IsSurrogate1, Uni_IsSurrogate2 } from './util/util.js';
export * as util from './util/util.js';
export * as convertUtil from './util/convert-utils.js';
export { ObjectWithCompileContext } from './util/types.js';
export * as KeymanFileTypes from './util/file-types.js';

173
common/web/types/src/util/convert-utils.ts
View file

@ -1,173 +0,0 @@
/*
* Keyman is copyright (C) SIL Global. MIT License.
*
* Created by S. Schmitt on 2026-01-19
*
* util for conversion functions for Keyman
*
*/
import { util } from '@keymanapp/common-types';
export class UnicodeCharacterConversion {
// &#x followed by 1.-6. hex digits will later be used for conversion
private static re_hex = /^&#x([0-9a-f]{1,6});$/i;
// &# followed by 1.-6. decimal digits will later be used for conversion
private static re_dec = /^&#([0-9]{1,7});$/;
// & followed by gt, lt, quot, amp, apos will later be used for conversion
private static re_nam = /^&(gt|lt|quot|amp|apos);$/i;
// &# followed by anything will later be refused for conversion
private static re_html_inv = /^(&#)+(.?)+$/i;
// one or more characters except starting & will later be used for conversion
private static re_chr = /^(?!&).+$/i;
// '&', '&#','&#x' with or without ; will later be refused for conversion
private static re_chr_inv = /^((&;?)+|(&#;?)+|(&#x;?)+;?)$|^$/i;
public convert(inputString: string): string | undefined {
const m_hex = UnicodeCharacterConversion.re_hex.exec(inputString);
const m_dec = UnicodeCharacterConversion.re_dec.exec(inputString);
const m_nam = UnicodeCharacterConversion.re_nam.exec(inputString);
const m_html_inv = UnicodeCharacterConversion.re_html_inv.exec(inputString);
const m_chr = UnicodeCharacterConversion.re_chr.exec(inputString);
const m_chr_inv = UnicodeCharacterConversion.re_chr_inv.exec(inputString);
// valid '&#x...'
if (m_hex) {
const codePoint_h = parseInt(m_hex[1], 16);
// Reject surrogates and invalid codepoints
if (!(util.isValidUnicode(codePoint_h))) {
return undefined;
}
return String.fromCodePoint(codePoint_h);
}
// valid '&#...'
else if (m_dec) {
const codePoint_d = parseInt(m_dec[1], 10);
// Reject surrogates and invalid codepoints
if (!(util.isValidUnicode(codePoint_d))) {
return undefined;
}
return String.fromCodePoint(codePoint_d);
}
// valid '&gt', '&lt',..
else if (m_nam) {
switch (m_nam[1]) {
case 'gt': return '>';
case 'lt': return '<';
case 'quot': return '"';
case 'amp': return '&';
case 'apos': return "'";
default: return undefined;
}
}
// invalid '&...'
else if (m_html_inv) {
return undefined;
}
// single '&', ''
else if (m_chr_inv) {
return inputString;
}
// if no matches so far, check for one or more characters ('a','ab', 'ẘ','😎', '😎😎', )
else if (m_chr) {
return inputString;
}
return undefined;
}
};
/**
* @brief function to convert a character, numeric character reference or a unicode value to a character or unicode Codepoint
* if input is a valid single character or Codepoint like 'c','ä', 'ሴ', 'ẘ', '😎', the same character or Codepoint is returned (e.g. 'c' -> 'c', '😎' -> '😎')
* if input is a valid multi character/Codepoint string like 'ab', 'abcde', '😎😆', the same string is returned (e.g. 'abcde' -> 'abcde' , 😎😆 ->😎😆 )
* if input is a valid numeric character reference in hex or decimal, the corresponding character or unicode Codepoint is returned (e.g. &#x1F60E; -> 😎)
* if input is a one of the named character reference &gt; &lt; &amp; &quot; &apos;, the corresponding character is returned ( e.g. '&gt;' -> '>')
* @param inputString the string or stringvalue that will converted
* @return the input character/Codepoint if input is a valid character/Codepoint
* a converted character if input is a valid Unicode value, numeric character reference in hex or decimal, or named character reference
* or undefined if input is null or undefined, half a surrogate pair, or not recognized
*/
export function convertToUnicodeCharacter(inputString: string): string | undefined {
// null, undefined will later be refused for conversion
if (inputString == null || inputString == undefined) {
return undefined;
}
// &#x followed by 1.-6. hex digits will later be used for conversion
const m_hex = /^&#x([0-9a-f]{1,6});$/i.exec(inputString);
// &# followed by 1.-6. decimal digits will later be used for conversion
const m_dec = /^&#([0-9]{1,7});$/.exec(inputString);
// & followed by gt, lt, quot, amp, apos will later be used for conversion
const m_nam = /^&(gt|lt|quot|amp|apos);$/i.exec(inputString);
// &# followed by anything will later be refused for conversion
const m_html_inv = /^(&#)+(.?)+$/i.exec(inputString);
// one or more characters except starting & will later be used for conversion
const m_chr = /^(?!&).+$/i.exec(inputString);
// '&', '&#','&#x' with or without ; will later be refused for conversion
const m_chr_inv = /^((&;?)+|(&#;?)+|(&#x;?)+;?)$|^$/i.exec(inputString);
// valid '&#x...'
if (m_hex) {
const codePoint_h = parseInt(m_hex[1], 16);
if (!(util.isValidUnicode(codePoint_h))) {
return undefined;
}
return String.fromCodePoint(codePoint_h);
}
// valid '&#...'
else if (m_dec) {
const codePoint_d = parseInt(m_dec[1], 10);
if (!(util.isValidUnicode(codePoint_d))) {
return undefined;
}
return String.fromCodePoint(codePoint_d);
}
// valid '&gt', '&lt',..
else if (m_nam) {
switch (m_nam[1]) {
case 'gt': return '>';
case 'lt': return '<';
case 'quot': return '"';
case 'amp': return '&';
case 'apos': return "'";
default: return undefined;
}
}
// invalid '&...'
else if (m_html_inv) {
return undefined;
}
// single '&', ''
else if (m_chr_inv) {
return inputString;
}
// if no matches so far, check for one or more characters ('a','ab', 'ẘ','😎', '😎😎', )
else if (m_chr) {
return inputString;
}
return undefined;
}

71
common/web/types/tests/util/convert-utils.tests.ts
View file

@ -1,71 +0,0 @@
import 'mocha';
import { assert } from 'chai';
import { convertUtil } from '@keymanapp/common-types';
describe('convert-utils', function () {
describe('convertToUnicodeCharacter', function () {
[
["a&gt;b", 'a&gt;b'],
["<", '<'],
["a", 'a'],
["ሴ", 'ሴ'],
["W̊", "W̊"],
['😎', '😎'],
["ab", 'ab'],
["ሴЖ", 'ሴЖ'],
["ẘẈ", "ẘẈ"],
["😎😆", '😎😆'],
["aሴ😆", 'aሴ😆'],
["U+0061", 'U+0061'],
["&#x61;", 'a'],
["&#x1234;", 'ሴ'],
["&#x1E98;", "ẘ"],
["&#x1F60F;", '😏'],
["&#x0002;", '\u0002'],
["&#x1000000;", undefined],
["&#97;", 'a'],
["&#4660;", 'ሴ'],
["&#7835;", "ẛ"],
["&#128518;", '😆'],
["&#x1F606;", '😆'],
["&#0003;", '\u0003'],
["&#1000000;", '󴉀'],
["&commat;", undefined],
["&gt;", '>'],
["&lt;", '<'],
["&quot;", '"'],
["&apos;", "'"],
["&gt", undefined],
["␤", '␤'],
["␕", '␕'],
["", ''],
["", ''],
[undefined, undefined],
[null, undefined],
["U+", "U+"],
['&', '&'],
['&;', '&;'],
['&&', '&&'],
['&&;', '&&;'],
["&#&#", undefined],
["&#x&#x", undefined],
["&#", undefined],
["&#;", undefined],
["&#x", undefined],
["&#x;", undefined],
['&##', undefined],
['&##;', undefined],
["&#x10FFFF;", undefined],
["&#1114111;", undefined],
["&#x110000;", undefined],
["&#x1000000;", undefined],
['&#1234;56', undefined],
].forEach(function (values) {
it(('should convert "' + values[0] + '"').padEnd(30, " ") + 'to "' + values[1] + '"', async function () {
const result = convertUtil.convertToUnicodeCharacter(values[0] as string);
assert.equal(result, values[1]);
});
});
});
});

3
developer/src/kmc-convert/src/keylayout-to-kmn/kmn-file-writer.ts
View file

@ -68,7 +68,7 @@ export class UnicodeCharacterConversion {
public static re_dec = /^&#([0-9]{1,7});$/;
// &#x followed by 1.-6. hex digits or &# followed by 1.-7. decimal digits
private static re_hexdec = /^&#x?([0-9a-f]{1,7};)/ig;
private static re_hexdec = /^&#x?([0-9a-f]{1,7};)/i;
// & followed by gt, lt, quot, amp, apos and ; (&gt;)
private static re_nam = /&(gt|lt|quot|amp|apos);/i;
@ -155,7 +155,6 @@ export class UnicodeCharacterConversion {
inputString.input = this.unescape_string(inputString.input) ?? '';
const re_hexdec = UnicodeCharacterConversion.re_hexdec.exec(inputString.input);
UnicodeCharacterConversion.re_hexdec.lastIndex = 0;
// if the (remaining) input string starts with a hex or dec html entity ( &#x...; or &#...;) we need to convert this part to a character
if (re_hexdec) {