Practical guide

Keep Unicode and structured fields in exports

Preserve Japanese, Arabic, accents, arrays, and grouped attributes across exports.

UTF-8 across all three exports

FormatEncoding behaviorNested values
JSONUTF-8 textNative arrays and objects
CSVUTF-8 with BOMJSON strings in quoted cells
MarkdownUTF-8 textReadable headings and lists

Real non-English evidence

The Saudi Arabia fixture preserves an Arabic title and Arabic category labels. The Germany fixture preserves a German title and category path. The exporter does not translate or transliterate those source values.

شنطة ظهر لطيفة للسفر واللابتوب…
Netter Rucksack für Reise, Laptop…

Avoid common corruption

  • Open CSV through an UTF-8-aware importer.
  • Do not remove the CSV BOM before spreadsheet import.
  • Keep quoted newlines and commas inside their cells.
  • Use a font and interface that support Arabic directionality; the underlying text remains Unicode.

Results may be cached and optional source fields can be absent. See the data methodology for collection and field boundaries. Read the data methodology.

Frequently asked questions

What prevents mojibake in spreadsheet imports?

Import the downloaded CSV explicitly as UTF-8 and keep comma delimiters and quoted cells enabled.