Extract Emails from Text

The Extract Emails tool pulls every email address out of any pasted text — documents, log files, web pages, email threads, HTML source, code files, whatever you drop in. This is one of those operations everybody needs eventually: a CSV arrived with emails buried in a notes column, a log file contains user addresses you need to audit, an exported contact list needs dedup, a support ticket thread has email addresses in the message bodies. Manually scanning and copying is tedious and error-prone. The tool does it in milliseconds with an RFC-compliant regex pattern.

Output is clean, deduplicated, and optionally validated. You get a newline-separated list of unique email addresses, sorted if you want, with duplicates removed automatically. Optional validation applies stricter regex to catch typos and malformed addresses. Export formats include plain list, CSV, JSON array, or a SQL INSERT statement ready for database loading. All processing is local to your browser — addresses never touch a server, which matters when the source text is internal or regulated. Remember that extracting emails to send unsolicited messages is illegal in most jurisdictions (GDPR, CAN-SPAM, CASL). Use this tool for legitimate purposes: deduping lists you own, auditing logs, or extracting contacts from your own exports.

Matching email addresses with a regex is notoriously tricky because RFC 5322 allows extremely permissive syntax (quoted local parts with almost any character, internationalized domain names, extensive comments). A fully compliant matcher runs hundreds of lines; a practical matcher covers 99%+ of real addresses with a short pattern.

This tool uses a practical pattern: [A-Za-z0-9._%+-]+@[A-Za-z0-9.-]+\\.[A-Za-z]{2,24}

Local part: alphanumeric plus dot, underscore, percent, plus, hyphen. Does not allow quoted local parts (rare) or unusual characters.

Separator: @ symbol.

Domain: alphanumeric with dots and hyphens. Does not allow IP-literal domains or internationalized domain names (IDN punycode required).

TLD: 2 to 24 letters. Modern TLDs like .museum and .photography fit; weird ones longer than 24 characters would fail (rare in practice).

This pattern catches virtually every email you will encounter in real text without false positives on common patterns. For internationalized email (Unicode local parts, IDN domains), enable the strict-Unicode mode which uses \\p{L} Unicode property escapes.

Validation beyond regex: the tool optionally checks that the TLD is in the IANA published list (no fake TLDs), the domain has at least one dot, and the local part has reasonable length (under 64 characters per RFC). It cannot check whether the address is deliverable without sending mail — that requires SMTP verification, which the tool does not do.

Deduplication: extracted addresses are normalized (lowercased) and deduped via Set. Case is preserved in output by default (first occurrence wins); enable \"force lowercase output\" if your downstream system requires it.

Surrounding context removal: emails often appear with angle brackets (<john@example.com>), labels (\"John Doe\" <john@example.com>), or punctuation (comma-separated lists). The tool extracts just the address part, removing decoration.

Performance: regex extraction on multi-megabyte text runs in milliseconds. Works on logs, exports, and documents of any typical size.

Compared to writing a custom regex in code, this tool gives an instant interactive workflow with validation and deduplication built in. For automation in a data pipeline, write code; for interactive extraction from pasted text, this tool is faster.

Compared to dedicated contact management tools, this tool is the extraction step before import. Run it to clean and dedupe a list, then import into Mailchimp, HubSpot, or whatever you use.

Compared to Unix grep with an email regex, this tool has a visual UI, automatic deduplication, and multiple output formats. grep is better for scripting and large-file piping; this tool is better for interactive work.