Translate a CSV string table into Italian

What Italian changes in a string table

Moving from an en column to an Italian one raises no script issues: Latin alphabet, left to right. The difficulties lie in short strings with little context.

• Length: Italian generally runs longer than English. Settings becomes Impostazioni, Delete becomes Elimina. Buttons sized for English are the first to overflow.
• Register: the translation uses the formal Lei form (Inserisca la password). Many Italian games and apps use the informal tu (Inserisci) or the infinitive (Inserire), so check the tone on the free sample before paying.
• Accents: perché, è, città, and the capital È at the start of a sentence, often wrongly typed as E' in interfaces.
• Elision: Italian elides before vowels (l'utente, dell'account). A string such as Delete {item}? needs an article that depends on the inserted word; the translation usually rephrases around it, but review strings where a placeholder directly follows an article.

Plurals, gender and placeholders in a flat CSV

A CSV has no plural structure: each form is a separate row, with keys such as msg_count_one and msg_count_other. Italian has three CLDR categories: one, many and other. The many category only covers certain large round numbers (1 milione di utenti, with ‘di’), and most projects do without it. If your library expects it, add the row to the source before uploading: Manuce does not create rows, only values change.

Placeholders are copied as they are: {0}, %s, {player_name}, formatting tags. Each translated value is checked against the source; if a {count} goes missing or is duplicated and the mismatch persists, the source value is kept and the report says so.

Grammatical gender is the issue the file cannot carry: Welcome, {name} could be Benvenuto or Benvenuta. A neutral wording such as Le diamo il benvenuto, {name} avoids it; check greeting strings first.

What Manuce returns, and what it does not do

The expected file is a developer CSV: one key column (key, id, keys in Godot) and language columns (en, fr, de_DE…). Manuce fills the Italian column if it exists, otherwise adds it as the last column. A comment or context column is used as context: with ‘button’, Open becomes Apri; with ‘status’, Aperto.

• Kept identical: row and key order, quoting, comments, line endings, encoding and BOM.
• Not translated: keys, empty values, URLs, email addresses, numbers, colours (#FF8800), date formats (MM/dd/yyyy stays as written; the Italian day-first order is handled in your code), identifiers and anything marked as non-translatable.
• A report lists the strings translated and those left as they were, with the reason.

Manuce does not test the result in your game or app, does not shorten strings to fit a button, and does not add a missing plural form. A CSV without a key column is not treated as a language file: it is translated as plain text. Pricing is based on the words in the values, keys and tags excluded; Italian is charged at the base rate (×1 coefficient).

Preparing the CSV

• Check the key column header: it is what identifies the file as a language file.
• Create the Italian column yourself if your engine expects a specific code (it or it_IT): it will be filled in place rather than relying on a column appended at the end.
• Save as UTF-8. Common Italian letters exist in older Western encodings, but typographic apostrophes and guillemets are easily lost. Since the original encoding and BOM are preserved, the file comes back exactly as you sent it.
• Fill in the context column for one- or two-word strings: Post, Share or Free translate differently as verb, noun or price (Gratis versus Libero).
• Mark game or product names as non-translatable, then check them in the free sample.

Checking the Italian column on return

1. Drop the file straight back into the project. Opening and re-saving a CSV in a spreadsheet can change the delimiter, quoting or encoding; a spreadsheet set to an Italian locale often uses semicolons.
2. If a spreadsheet shows perché instead of perché, the cause is a UTF-8 file without BOM being opened with another encoding, not the content.
3. Read the report: a value left in the source language almost always points to a placeholder or tag worth examining.
4. Run the interface in Italian and look at the narrowest buttons, tabs and menus.
5. Check register consistency: a mix of Lei and tu usually comes from source strings that are themselves inconsistent.

In practice

1. Check the key column and, if needed, create an it or it_IT column.
2. Upload the CSV: the quote is based on the words in the values.
3. Read the free sample to judge register and terminology.
4. Download the translated CSV and its report, then test the Italian interface.