Text Filter

Extract Data from Any Text Instantly

Text Filter extracts specific data from any text — paste and choose a filter to get a clean, deduplicated list instantly. Supports all languages and Unicode scripts. All processing happens locally in your browser — your data never leaves your device. No registration, no ads, no limits, forever free.

Useful for social media managers collecting hashtags, marketers pulling discount codes from campaign copy, developers extracting URLs or identifiers from logs, and anyone scanning a document for numbers or links before publishing. Unlike AI tools, Text Filter processes everything deterministically — no hallucinations, no length limits, no internet required.

How to Use

• Paste your text: Enter or paste any text into the text area above — plain text, Markdown, HTML, rich text copied from websites, AI-generated content (ChatGPT, Claude, Gemini), social media posts, logs, documents, and more. All formats work.
• #Hashtags: Extracts all standalone words prefixed with # — e.g., #design, #SocialMedia. Supports full Unicode: Latin, Arabic, Chinese, Japanese, Korean, Cyrillic, and more. Only standalone hashtags are captured; # symbols inside headings or other formatting are ignored.
• Emails: Extracts all valid email addresses found in the text — e.g., [email protected]. Trailing punctuation attached to an email (such as a period or comma at the end of a sentence) is automatically stripped from results.
• Links: Extracts all URLs starting with http://, https://, or www., as well as short URLs without a scheme — e.g., bit.ly/eRy3lr, tinyurl.com/Av2Eic. Trailing punctuation is automatically stripped from results.
• Numbers: Extracts standalone numeric tokens — integers, decimals, negative numbers, thousands-formatted numbers, percentages, and numeric ranges connected by a dash — e.g., 42, 3.14, 1,000,000, -99, 30%, 123-789, 2025–2030. Any Unicode dash character (hyphen, en dash, em dash, and others) is recognised as a range separator when both sides are digits. Consecutive plain digit groups separated by a single space are joined into one result — e.g., 0800 123 456 and 1234 5678 9012 3456 each appear as a single entry, making it easy to extract phone numbers and bank account numbers formatted with spaces. Digits inside mixed tokens like ORD-2026, version strings like v1.2.3, or amounts with a unit directly attached like 100kg or 50USD are not extracted by Numbers — they may appear in Codes results instead.
• Codes: Extracts standalone tokens that contain both letters and digits — order codes, discount codes, serial numbers, version strings, license plates, and tracking numbers — e.g., ORD-2026-AB12, ORD–2026–AB12, SUMMER30, v1.2.3, 69A-12345, 100kg. Any Unicode dash character (hyphen, en dash, em dash, and others) as well as _, ., @, / are recognised as joining characters within a token — so codes exported from Word, PDF, or ERP systems that use typographic dashes are captured correctly. Tokens must begin and end with a letter or digit. Tokens containing only digits or only letters are excluded. Supports full Unicode, including non-Latin scripts mixed with digits (e.g., 商品A123, 주문2026).
• Emojis: Extracts all emoji characters from the text — smileys, objects, symbols, hand gestures, and more — e.g., 😀, 🎉, ❤️, 🇻🇳, 👨‍💻. Handles complex sequences correctly, including flags (🇯🇵), skin tone variants (👍🏻), and ZWJ combinations (👨‍👩‍👧‍👦, 🧑‍🚀).
• Horizontal / Vertical: Toggle between results displayed on a single line (space-separated) or stacked one per line. Default is vertical. Switch at any time without re-running the filter.
• Copy: Copies all results currently displayed to your clipboard, ready to paste anywhere.
• Delete: Clears both input and output so you can start a fresh session.
• Theme Options:
  • Light: Bright interface for daytime use.
  • Dark: Dark interface that's easy on eyes at night.
  • Auto: Matches your device's theme preference.

How It Works

Text Filter works with any text format — no cleanup needed before pasting. Whether your text is plain, wrapped in Markdown syntax, embedded in HTML, or copied from an AI response, a website, or a document, each filter extracts only what you need and ignores everything else. Duplicates are automatically removed while preserving the original order of appearance.

Each filter targets only standalone entities — tokens that appear independently, not directly adjacent to other letters, digits, or joining characters. This prevents noise: digits inside a code like ORD-2026 won't appear in Numbers results, and a plain word like contact won't appear in Codes results. Processing is entirely client-side — no server calls, no external libraries, no AI.

Install as App

Install Text Filter on your device for instant access from your home screen. Works fully offline after the first load.

• Android (Chrome): Menu → "Add to Home screen" or "Install app".
• iPhone/iPad (Safari): Share → "Add to Home Screen".
• Desktop (Chrome/Edge): Click install icon in address bar.

Privacy & Security

• 100% local processing: All filtering happens in your browser. Your text never reaches any server.
• Works offline: After the first load, the app runs without any internet connection.
• No tracking: We use Cloudflare Web Analytics for anonymous traffic stats only. No cookies, no ads, no personal data collected.
• No registration: Open the page and start using it immediately.

By using Text Filter, you agree to the Privacy & Terms.

About

• Text Filter: textfilter.sonlt.app
• Email: [email protected]
• Built by Lâm Thái Sơn