Re-implement standard slugification with IRIs instead of URIs.

src/backends/backend.js

@@ -5,7 +5,8 @@ import GitGatewayBackend from "./git-gateway/implementation";
 import { resolveFormat } from "../formats/formats";
 import { selectListMethod, selectEntrySlug, selectEntryPath, selectAllowNewEntries, selectFolderEntryExtension } from "../reducers/collections";
 import { createEntry } from "../valueObjects/Entry";
-import sanitize from 'sanitize-filename';
+import { sanitizeIRI } from "../lib/urlHelper";
+import sanitizeFilename from 'sanitize-filename';
 
 class LocalStorageAuthStore {
   storageKey = "netlify-cms-user";
@@ -57,7 +58,19 @@ const slugFormatter = (template = "{{slug}}", entryData) => {
     }
   });
 
-  return sanitize(slug, {replacement: "-"}).replace(/[.]/g, '-');
+  // Convert slug to lower-case;
+  slug = slug.toLocaleLowerCase();
+
+  // Replace periods and spaces with dashes.
+  slug = slug.replace(/[.\s]/g, '-');
+  // Sanitize as IRI (i18n URI) and as filename.
+  slug = sanitizeIRI(slug, {replacement: "-"});
+  slug = sanitizeFilename(slug, {replacement: "-"});
+
+  // Remove any doubled or trailing replacement characters (that were added in the sanitizers).
+  slug = slug.replace(/-+/g, '-').replace(/-$/, '');
+
+  return slug;
 };
 
 class Backend {

src/lib/urlHelper.js

@@ -12,6 +12,22 @@ export function getNewEntryUrl(collectionName, direct) {
   return getUrl(`/collections/${ collectionName }/entries/new`, direct);
 }
 
+// Unreserved chars from RFC3987.
+const uriChars = /[\w\-.~]/i;
+const ucsChars = /[\xA0-\u{D7FF}]|[\u{F900}-\u{FDCF}]|[\u{FDF0}-\u{FFEF}]|[\u{10000}-\u{1FFFD}]|[\u{20000}-\u{2FFFD}]|[\u{30000}-\u{3FFFD}]|[\u{40000}-\u{4FFFD}]|[\u{50000}-\u{5FFFD}]|[\u{60000}-\u{6FFFD}]|[\u{70000}-\u{7FFFD}]|[\u{80000}-\u{8FFFD}]|[\u{90000}-\u{9FFFD}]|[\u{A0000}-\u{AFFFD}]|[\u{B0000}-\u{BFFFD}]|[\u{C0000}-\u{CFFFD}]|[\u{D0000}-\u{DFFFD}]|[\u{E1000}-\u{EFFFD}]/u;
+export function sanitizeIRI(str, { replacement }) {
+  let result = "";
+  // We cannot use a `map` function here because `string.split()` splits things like emojis into surrogate pairs.
+  for (const char of str) {
+    if (uriChars.test(char) || ucsChars.test(char)) {
+      result += char;
+    } else {
+      result += replacement;
+    }
+  }
+  return result;
+}
+
 export function urlize(string) {
   const sanitized = makePathSanitized(string);
   const parsedURL = url.parse(sanitized);