front: handle accents in STDCM vias autocomplete

Signed-off-by: Leo Valais <leo.valais97@gmail.com>

front/src/applications/stdcmV2/components/StdcmOperationalPoint.tsx

@@ -7,6 +7,7 @@ import nextId from 'react-id-generator';
 import type { SearchResultItemOperationalPoint } from 'common/api/osrdEditoastApi';
 import useSearchOperationalPoint from 'common/Map/Search/useSearchOperationalPoint';
 import type { PathStep } from 'reducers/osrdconf/types';
+import { normalized } from 'utils/strings';
 import { createFixedSelectOptions } from 'utils/uiCoreHelpers';
 
 type StdcmOperationalPointProps = {
@@ -39,7 +40,7 @@ const StdcmOperationalPoint = ({
       sortedSearchResults
         .filter(
           (op) =>
-            op.name.toLowerCase().startsWith(searchTerm.toLowerCase()) ||
+            normalized(op.name).startsWith(normalized(searchTerm)) ||
             op.trigram === searchTerm.toUpperCase()
         )
         .reduce((acc, p) => {

front/src/utils/strings.ts

@@ -71,6 +71,20 @@ export function formatUicToCi(uic: number) {
   return uic.toString().replace(/^87/, '');
 }
 
+/**
+ * Normalizes a string by removing accents and converting to lowercase.
+ *
+ * Accent removal is done using `String.normalize` which converts accented characters to their
+ * decomposed form (e.g. é -> e + ´). We then remove all characters in the range of U+0300 to U+036f
+ * which contains all unicode combining diacritical marks (https://en.wikipedia.org/wiki/Combining_Diacritical_Marks).
+ */
+export function normalized(s: string): string {
+  return s
+    .toLowerCase()
+    .normalize('NFD')
+    .replace(/[\u0300-\u036f]/g, '');
+}
+
 export const TEXT_AREA_MAX_LENGTH = 4096;
 export const SMALL_TEXT_AREA_MAX_LENGTH = 1024;
 export const TEXT_INPUT_MAX_LENGTH = 255;