أفضل ذكاء اصطناعي لـ Generate a regex pattern

For most validation use cases, this is good enough — it accepts what real email systems will deliver to and rejects the obvious garbage: const emailRegex = /^[^\s@]+@[^\s@]+\.[^\s@]+$/; if (emailRegex.test(input)) { // accept — but still verify with a confirmation email; regex can't catch typos like "gmal.com" } What each piece does: - `^` and `$`: anchors. Without them, the regex matches "looks like there's an email somewhere in this string" — and your validation accepts pasted essays. - `[^\s@]+`: one or more characters that aren't whitespace and aren't @. Used for the local part and each domain segment. - `@`: the literal @. - `\.`: a literal dot. Escaped because `.` matches anything otherwise. What this rejects that the original accepted: - "a@b@c" — would have passed the original since `.+` matches @ symbols - " @ . " — would have passed; this rejects whitespace - "test@example" — passes the original (no dot required); this requires a dot in the domain Gotcha: this still accepts "x@y.z" where y is a single character. RFC-strict regexes get monstrous. For most apps, "x@y.z" is fine to accept and let the email system bounce if it's actually invalid — your regex's job is to catch obvious junk, not validate that an inbox exists. If your case demands stricter (e.g., a list of allowed TLDs, or rejecting plus-addressing): add a negative lookahead or a TLD allow-list rather than trying to write the One True Email Regex.