jess
/
Graphite
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Fix logic bug in key internationalization for A-Z keys in alternate layouts

This commit is contained in:
Keavon Chambers 2022-08-14 13:57:01 -07:00
parent 410b3d672c
commit 78a3644c45
1 changed files with 9 additions and 15 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
frontend/src/utility-functions/keyboard-entry.ts
View File

@ -71,8 +71,7 @@ export async function getLocalizedScanCode(e: KeyboardEvent): Promise<string> {
// Use good-enough-for-now heuristics on the writing system keys, which are commonly subject to change by locale
// Number scan codes
const scanCodeDigit = /^Digit[0-9]$/.test(scanCode);
if (scanCodeDigit) {
if (/^Digit[0-9]$/.test(scanCode)) {
// For now it's good enough to treat every digit key, regardless of locale, as just its digit from the standard US layout.
// Even on a keyboard like the French AZERTY layout, where numbers are shifted, users still refer to those keys by their numbers.
// This unfortunately means that any special symbols under these keys are overridden by their number, making it impossible to access some shortcuts that rely on those special symbols.
@ -81,20 +80,15 @@ export async function getLocalizedScanCode(e: KeyboardEvent): Promise<string> {
}
// Letter scan codes
const scanCodeLetter = scanCode.match(/^Key([A-Z])$/);
if (scanCodeLetter) {
const scanCodeLetterValue = String(scanCodeLetter[1]);
if (/^Key([A-Z])$/.test(scanCode)) {
// Get the uppercase letter, with any accents or discritics removed if possible
const rawLetter = keyText
.normalize("NFD")
.replace(/\p{Diacritic}/gu, "")
.toUpperCase();
// If the scan code matches the key letter (ignoring diacritics and case), use that letter directly
const letterOfScanCodeMatchesKey =
scanCodeLetterValue ===
keyText
.normalize("NFD")
.replace(/\p{Diacritic}/gu, "")
.toUpperCase();
if (letterOfScanCodeMatchesKey) {
return scanCode;
}
// If the key letter is in the A-Z range, use the key code for that letter
if (/^[A-Z]$/.test(rawLetter)) return `Key${rawLetter}`;
// If the key text isn't one of the named attribute values, that means it must be the literal unicode value which we use directly
// It is likely a weird symbol that isn't in the A-Z range even with accents removed.