What is the difference between Auto, Structural, and Exact mode?

Auto mode first checks your examples against 10 common patterns (email, UUID, URL, IPv4, ISO date, hex color, slug, phone, digits, identifier) and uses the matching one if found; otherwise it falls back to Structural. Structural mode extracts shared prefix/suffix and infers a character class for the variable middle — good for custom IDs, codes, and reference numbers. Exact mode escapes every example literally and joins them with | alternation, guaranteeing no false positives at the cost of a longer pattern that won't generalise to new inputs.

What are counter-examples and why should I use them?

Counter-examples are strings that look similar to your examples but must NOT be matched by the generated pattern. For instance, if generating a pattern for "PRD-001" style product IDs, counter-examples like "PRD-" (incomplete) or "001" (missing prefix) quickly reveal if the pattern is too broad. The validation table immediately flags any counter-examples that the pattern accidentally matches, so you know to switch modes or refine the pattern manually.

Are the generated patterns production-ready?

Known-pattern results (email, UUID, IPv4, ISO date, etc.) are well-tested and suitable for production use. Structural and Exact patterns are solid starting points that you should review and tighten against real-world data before deploying, especially for security-sensitive inputs such as authentication tokens, API keys, or form validation.

Does the generated regex work in all programming languages?

The core pattern uses a broadly compatible regex subset. The code snippet panel handles language-specific differences: Python uses re.compile() with the IGNORECASE flag object, Go and Rust prepend (?i) as an inline flag (they don't support trailing /i syntax), Java uses Pattern.compile() with Pattern.CASE_INSENSITIVE, and PHP wraps the pattern in / delimiters for preg_match().

What does the Anchored toggle do?

Anchoring wraps the pattern with ^ (start of string) and $ (end of string), so the entire input string must match from start to finish rather than just containing the pattern somewhere. For form field validation (email, ID, date) you almost always want anchoring. For searching within a larger block of text (log files, documents) leave it unchecked.

Yes. All pattern generation, structural analysis, validation, and code snippet production runs entirely in your browser using JavaScript. None of your example strings, counter-examples, or generated patterns are ever sent to any server — not even CodeLint.Dev servers. You can safely use this tool with sensitive data such as internal reference IDs, log lines, customer identifiers, or API key formats.

Regex Generator

Generate regex patterns from example strings. Auto-detects emails, URLs, UUIDs, and more.

Private Try Regex Tester →

Presets:

Example strings (one per line — must match)

Counter-examples (optional — must NOT match)

Mode

Anchored ^…$ Case insensitive i

Auto: Detects well-known formats (email, UUID, URL…) first; falls back to structural analysis.

Enter example strings on the left, or click a preset to get started.

About Regex Generator

The Regex Generator automatically builds regular expression patterns from your example strings. Instead of writing regex from scratch, you describe what you want to match by example, and the tool derives the pattern for you.

Auto mode first checks your examples against 10 well-known pattern signatures — email addresses, URLs, UUIDs, IPv4 addresses, ISO dates, hex colors, URL slugs, phone numbers, digit-only strings, and code identifiers. If all examples match a known pattern, a precise, production-ready regex is returned with a full explanation.

When examples don't fit a known pattern, Structural mode extracts the longest common prefix and suffix, infers the character class of the variable middle segment (\d, [a-z], [a-zA-Z0-9], etc.), and generates a concise pattern with the correct {min,max} quantifier.

Counter-examples are strings that must not match your pattern. Adding them lets you immediately see if your generated regex is too permissive — a feature rarely found in other regex generators.

Once a pattern is generated, the code snippet panel shows you ready-to-paste code in 8 languages: JavaScript, TypeScript, Python, Go, Java, PHP, Rust, and C#. Language-specific escaping and flag conventions (e.g. re.IGNORECASE for Python, (?i) inline flag for Go and Rust) are handled automatically.

Everything runs entirely in your browser. No example strings, no counter-examples, and no generated patterns are ever transmitted to any server. It is safe to use with log lines, PII, API keys, or any sensitive content.

Frequently Asked Questions

What is the difference between Auto, Structural, and Exact mode?: Auto mode first checks your examples against 10 common patterns (email, UUID, URL, etc.) and uses the matching one if found; otherwise it falls back to Structural. Structural mode extracts shared prefix/suffix and infers a character class for the variable middle — good for custom IDs and codes. Exact mode escapes every example literally and joins them with | alternation, guaranteeing no false positives at the cost of a longer pattern.
Why should I add counter-examples?: Counter-examples are strings that look similar to your examples but should NOT match — for instance, if you are generating a pattern for product IDs like "PRD-001", a counter-example might be "PRD-" (incomplete) or "001" (missing prefix). The validation table immediately shows if the generated pattern is too broad and accidentally matches your counter-examples, so you can switch modes or refine manually.
Is this tool suitable for production regex patterns?: Known-pattern results (email, UUID, IPv4, etc.) are well-tested and production-ready. Structural and Exact patterns are good starting points that you should review and tighten manually — especially for security-sensitive inputs like form validation. Always test generated patterns against a broader set of real-world data before deploying.
Does the generated regex work in all programming languages?: The core pattern is written in a broadly compatible subset of regex syntax. The code snippet panel handles language-specific details: Python uses re.compile() with the IGNORECASE flag, Go and Rust use the inline (?i) flag (since they do not support the /i trailing flag syntax), Java uses Pattern.compile() with Pattern.CASE_INSENSITIVE, and PHP wraps the pattern in / delimiters.
What does "Anchored" mean?: Anchoring wraps the pattern with ^ (start of string) and $ (end of string), which means the entire string must match rather than just containing a match. For validation (checking if a whole input field is an email, for example), anchoring is usually what you want. For searching within a longer text, uncheck it.
Is my data private and secure?: Yes. All pattern generation, validation, and code snippet production runs entirely in your browser using JavaScript. None of your example strings, counter-examples, or generated patterns are sent to any server — not even CodeLint.Dev servers. You can safely use this tool with sensitive data such as log lines, customer IDs, or internal formats.

About

The Regex Generator automatically builds regular expression patterns from your example strings — no regex knowledge required. Instead of writing regex syntax by hand, you provide examples of what you want to match and the tool derives the pattern for you. Auto mode checks your examples against 10 well-known pattern signatures: email addresses, HTTP/HTTPS URLs, UUIDs/GUIDs, IPv4 addresses, ISO 8601 dates, CSS hex colors, URL slugs, phone numbers, digit-only strings, and code identifiers. When a known pattern matches all examples, a precise production-ready regex with a full token-by-token explanation is returned. For custom patterns, Structural mode extracts the longest common prefix and suffix from your examples, infers the character class of the variable middle segment (\d, [a-z], [a-zA-Z0-9], \S, etc.), and generates a quantified pattern such as PREFIX\d{4,6}SUFFIX. Exact mode escapes every example literally and joins them with | alternation — zero false positives, longer pattern. Counter-examples let you verify your pattern doesn't match strings it shouldn't. The code snippet panel outputs ready-to-paste code in JavaScript, TypeScript, Python, Go, Java, PHP, Rust, and C#, handling language-specific escaping and flag conventions automatically. All processing is done in your browser — no example strings or generated patterns ever leave your device.

How to use

1 Type or paste your example strings, one per line, into the Example strings textarea — these are strings the pattern must match.
2 Optionally add counter-examples (strings the pattern must NOT match) in the second textarea to validate that the generated pattern is precise enough.
3 Select a generation mode: Auto (recommended — detects known formats first), Structural (analyzes shared structure), or Exact (literal alternation of all examples).
4 Toggle Anchored to wrap the pattern in ^ and $ anchors (for whole-string matching in form validation) or leave it unchecked for substring search.
5 Click a Preset button (Email, UUID, URL, etc.) to load example strings for common formats and see the generator in action.
6 Copy the generated pattern using the copy button, or switch to the code snippet panel to copy ready-to-paste code in your programming language.

What is the difference between Auto, Structural, and Exact mode?: Auto mode first checks your examples against 10 common patterns (email, UUID, URL, IPv4, ISO date, hex color, slug, phone, digits, identifier) and uses the matching one if found; otherwise it falls back to Structural. Structural mode extracts shared prefix/suffix and infers a character class for the variable middle — good for custom IDs, codes, and reference numbers. Exact mode escapes every example literally and joins them with | alternation, guaranteeing no false positives at the cost of a longer pattern that won't generalise to new inputs.
What are counter-examples and why should I use them?: Counter-examples are strings that look similar to your examples but must NOT be matched by the generated pattern. For instance, if generating a pattern for "PRD-001" style product IDs, counter-examples like "PRD-" (incomplete) or "001" (missing prefix) quickly reveal if the pattern is too broad. The validation table immediately flags any counter-examples that the pattern accidentally matches, so you know to switch modes or refine the pattern manually.
Are the generated patterns production-ready?: Known-pattern results (email, UUID, IPv4, ISO date, etc.) are well-tested and suitable for production use. Structural and Exact patterns are solid starting points that you should review and tighten against real-world data before deploying, especially for security-sensitive inputs such as authentication tokens, API keys, or form validation.
Does the generated regex work in all programming languages?: The core pattern uses a broadly compatible regex subset. The code snippet panel handles language-specific differences: Python uses re.compile() with the IGNORECASE flag object, Go and Rust prepend (?i) as an inline flag (they don't support trailing /i syntax), Java uses Pattern.compile() with Pattern.CASE_INSENSITIVE, and PHP wraps the pattern in / delimiters for preg_match().
What does the Anchored toggle do?: Anchoring wraps the pattern with ^ (start of string) and $ (end of string), so the entire input string must match from start to finish rather than just containing the pattern somewhere. For form field validation (email, ID, date) you almost always want anchoring. For searching within a larger block of text (log files, documents) leave it unchecked.
Is my data private?: Yes. All pattern generation, structural analysis, validation, and code snippet production runs entirely in your browser using JavaScript. None of your example strings, counter-examples, or generated patterns are ever sent to any server — not even CodeLint.Dev servers. You can safely use this tool with sensitive data such as internal reference IDs, log lines, customer identifiers, or API key formats.