Why does my Arabic text appear backwards?

Software lays out text using the Unicode bidirectional algorithm. Inside a left-to-right context, an Arabic run mixed with digits, punctuation, or Latin words can be reordered incorrectly because the surrounding weak and neutral characters resolve to the wrong direction. Wrapping the run in an explicit RTL control forces the right order.

What is the difference between an isolate and an embedding?

An isolate (RLI…PDI or FSI…PDI) sets the direction for its contents while hiding them from the surrounding bidi calculation, so it cannot leak into neighbouring text. The older embedding (RLE…PDF) sets direction but still participates in the outer context. Isolates are the modern recommendation.

When should I use plain RLM marks?

The right-to-left mark (RLM) is an invisible zero-width character that nudges adjacent neutral characters to resolve as RTL. Bracketing a snippet with RLM on both sides is the lightest-weight fix and is handy when you only need to correct trailing punctuation, but full isolates are more robust for mixed content.

Will these characters show up visibly?

No. All of the controls — RLM, RLE, PDF, RLI, FSI, PDI — are zero-width and invisible. They affect ordering only. The visible characters of your text are unchanged; the snippet just renders in the correct direction.

Does the tool send my text anywhere?

No. It simply prepends and appends Unicode control characters in your browser. Nothing is uploaded, and the copy button puts the result, controls included, onto your clipboard.

What is the Arabic RTL Direction Fixer?

Fix scrambled Arabic that renders backwards in left-to-right apps by wrapping it in the correct Unicode bidi controls — RLI isolates, FSI, RLE embedding, or RLM marks. Copy the corrected snippet straight from your browser. It runs free in your browser on Gera Tools, with nothing uploaded.

Arabic RTL Direction Fixer

Name: Arabic RTL Direction Fixer
Creator: Gera Tools
License: https://creativecommons.org/licenses/by/4.0/

Get one useful tool a week

Like this tool? Enter your email and we'll send you one genuinely useful Gera tool a week — plus a link to come back to this one. No spam, one-click unsubscribe any time.

Make Arabic read the right way in LTR apps

You paste a clean Arabic phrase into a left-to-right field — a button label, a JSON value, a chat message — and the punctuation jumps to the wrong side or the whole run reverses. The text is fine; the layout engine just resolved its direction wrong. This tool wraps the snippet in the correct invisible Unicode bidi controls so it renders properly wherever you paste it.

Why bidi goes wrong

The Unicode bidirectional algorithm (UBA, Unicode Standard Annex #9) is responsible for determining the visual order of characters on screen. It works well for text that is clearly one direction, but mixed-direction content — Arabic words alongside Latin letters, digits, or punctuation — requires the algorithm to make decisions about neutral characters such as spaces, parentheses, and colons.

In a left-to-right context, neutral characters resolve to left-to-right direction by default. An Arabic run embedded in an LTR field can have its surrounding punctuation flip to the wrong side because the algorithm sees the LTR context as the base and resolves neutral characters accordingly.

Common symptoms:

A trailing parenthesis or quotation mark appears on the left instead of the right
The entire Arabic phrase appears in reverse visual order
An Arabic phrase embedded in an English label has its first and last words swapped
Phone numbers or codes embedded in Arabic text appear in the wrong order

How it works

The Unicode bidirectional algorithm decides direction from the characters present and their context. By bracketing your text with explicit control characters, you override that guess for just that run. The tool offers four methods, strongest isolation first:

RLI…PDI — right-to-left isolate. Sets RTL and hides the run from the outer context (recommended).
FSI…PDI — first-strong isolate. Direction is taken from the first strong character inside.
RLE…PDF — legacy RTL embedding. Sets RTL but still interacts with surrounding text.
RLM marks — lightweight zero-width bracketing for neutral characters.

A live preview renders the wrapped output inside a deliberately left-to-right box so you can confirm the fix before copying.

Choosing the right method

Situation	Recommended method
Arabic phrase in an English UI label	RLI…PDI
Bidirectional paragraph (mixed Arabic + English)	FSI…PDI
Legacy system that does not support isolates	RLE…PDF
Just a trailing punctuation mark misbehaving	RLM marks

The isolate methods (RLI and FSI) are the modern recommendation because they create a bidi context that cannot leak into the surrounding paragraph. The legacy embedding (RLE…PDF) was the only option before Unicode 6.3 introduced isolates, and some older environments still only support it.

Tips for developers

If you are building a web application that displays Arabic content, the cleanest fix is to set dir="rtl" on the HTML element or to use the CSS property direction: rtl on the containing element, rather than inserting bidi control characters into the string. Reserve the control characters for cases where you cannot control the rendering environment — API responses that go into third-party apps, CSV files pasted into spreadsheets, or strings embedded in JSON that will be displayed by an engine you do not control.