{"version":"2.1.246","anchor":"cloud-sync-no-longer-mangles-filenames-that-are-not-valid-ut","canonical_anchor":"cloud-sync-no-longer-mangles-filenames-that-are-not-valid-ut","heading":"Cloud sync no longer mangles filenames that are not valid UTF-8","tier":"notice","area":"Cloud Sync","scope":null,"heads_up":null,"url":"https:\/\/changelogs.core-directive.com\/v\/2.1.246\/e\/cloud-sync-no-longer-mangles-filenames-that-are-not-valid-ut","release_url":"https:\/\/changelogs.core-directive.com\/v\/2.1.246","markdown":"### Cloud sync no longer mangles filenames that are not valid UTF-8\n\nCloud sync holds back files with non-UTF-8 names instead of corrupting them or breaking uploads.\n\n**What**\n\nDirectory sync now carries raw filename bytes through the whole path rather than decoded strings, so files with names that are not valid UTF-8 are detected and held back with the message \"could not hold the paths whose names are not valid UTF-8\" instead of being corrupted or breaking the upload.\n\n**Details**\n\n- The git index payload is now assembled by concatenating raw byte buffers rather than building a text template.\n- Applies to anyone using cloud directory sync; no flag guards it.\n\n**Evidence**\n\n`could not hold the paths whose names are not valid UTF-8`\n\n- Area: Cloud Sync\n- Tier: You'll notice\n- Useful: 3\/5\n- Signal: 2\/5"}