twineconvert

free · in-browser · no upload

CSV to PO
Converter

Drop your CSV file. We'll convert it to .po right here in your browser, your file never leaves your device.

or drop your file

Select your file here to get started

or drop your file here.

Accepts .csv

nothing uploaded no file size cap no signup

How it works

Three steps. No upload, no signup.

  1. 1

    Drop your file

    Click the dropzone above or drag a CSV from your desktop. Files of any size, there's no upload, so there's no upload limit.

  2. 2

    Convert in your browser

    The conversion runs entirely in this tab using JavaScript and WebAssembly. Your file never touches our servers, we don't have any.

  3. 3

    Download

    Get your .po the moment the conversion finishes. Convert another, or close the tab.

Files stay on your device

Your file is never uploaded. The entire conversion runs in your browser using WebAssembly. We can't see what you convert because we have no server to see it.

No file size limit

Server converters cap free users at 1-2 GB and gate larger files behind a paid plan. Since nothing uploads, our limit is whatever your browser can handle.

Free, no signup, no ads on conversions

No account required. No watermark on the output. No queue. Drop a file, get a converted file.

Why convert CSV → PO

What this conversion is actually for

After translators edit your strings in a spreadsheet, you need to push them back into the codebase as a proper PO file that gettext, react-i18next, Django, or Poedit can consume. csv-to-po reads the same column layout po-to-csv emits, rebuilds plural arrays from the JSON-encoded column, and reattaches contexts, comments, references, and flags so the resulting PO drops in cleanly without breaking your build.

A real example

Your translation team sends back `messages.fr.csv` with reviewed `msgstr` cells. You drop the CSV here, get `messages.fr.po`, commit it to `locales/fr/LC_MESSAGES/`, run `msgfmt` (or your tool's equivalent), and the French build picks it up.

Troubleshooting

"Row N is missing a value in the msgid column" error.

Empty rows or rows that lost their msgid (often from a stray Excel sort) crash the build because PO entries are keyed by msgid + msgctxt. Open the CSV in any text editor, find the empty row (or one with only a comma), and either fill in the msgid or delete the row entirely. The header entry (msgid "" with PO file metadata) is allowed and only valid on row 2.

I converted a translator's CSV but the plurals came out as a single string.

Make sure the `msgstr_plurals` column contains a JSON array (`["form1","form2"]`) and not a single line. If your translator delivered each plural form in a separate column (`msgstr_0`, `msgstr_1`), our parser won't pick them up — you'll need to combine them into one JSON-encoded cell first.

Formats involved

About CSV and .po

CSV, Comma-Separated Values

CSV is plain text, one row per line, fields separated by commas. The simplest possible tabular data format, which is exactly why it remains the most portable: every spreadsheet, database, programming language, and analytics tool reads CSV. Tradeoffs: no formulas, no formatting, no multiple sheets, and various edge cases around quoting fields that contain commas or newlines.

How to open

Excel, Google Sheets, Apple Numbers, LibreOffice Calc, any text editor, every database import wizard, every programming language with one line of code.

Full CSV format guide →
.po, Gettext Portable Object

A PO file is the interchange format used by every gettext-based software localization toolchain: GNU gettext, Babel (Python), Poedit, Lokalise, Crowdin, Weblate, Transifex, polib, react-i18next, Django, Symfony, WordPress translations, and many more. Each entry pairs a source string (`msgid`) with its translation (`msgstr`), plus optional disambiguation context (`msgctxt`), plural forms (`msgid_plural` / `msgstr[N]`), translator/developer comments, source-file references, and gettext flags. Spec dates to 1995 with the original GNU gettext release; still the de-facto standard 30 years later.

How to open

Poedit (free, all platforms) is the most common editor. Lokalise/Crowdin/Weblate/Transifex import PO directly via web upload. Any text editor opens them as plain text. Linguists prefer Poedit for translation memory and validation; developers usually edit them in their IDE.

Full .po format guide →

Related tools

Convert other files to .po

Convert your CSV to other formats

FAQ

Common questions

Is this CSV → PO converter really free?

Yes. No signup, no watermark, no daily file count limit. Every conversion runs in your browser, your file never touches our servers because there are no servers.

Where does my file go when I convert it?

Nowhere. The conversion runs entirely in your browser using JavaScript and WebAssembly. Your file is never uploaded to our servers. We don't have any servers handling files, there's nothing for us to log, store, or accidentally leak.

What's the maximum file size?

Whatever your browser can hold in memory. Practically, this means a few hundred MB on most computers, significantly larger than the 1-2 GB caps that server-upload converters charge for. Very large files (multi-GB) may require closing other browser tabs first.

Why convert CSV to .po?

Universal tabular data interchange. Software localization (UI strings, error messages, in-app text in every language). The most common reason to convert is compatibility, .po works in places where CSV doesn't, or vice versa.

How do I open a CSV file in the first place?

Excel, Google Sheets, Apple Numbers, LibreOffice Calc, any text editor, every database import wizard, every programming language with one line of code.

Does this work offline?

Once the page is loaded, the conversion itself runs entirely offline. The first time you use a tool, your browser downloads the conversion library (a one-time cache). If you reload while offline, the page won't load, but you can install the site as a Progressive Web App for full offline use.

Can I convert multiple files at once?

Single file at a time for now. Batch conversion is on the roadmap, for now, drop one file, download the result, then convert the next.