fix!: TXT decode preserves string-boundary info

[msimerson]

BREAKING

RFC 1035 §3.3.14

A TXT RDATA can contain multiple <character-string> items, each length-prefixed. Packet.Resource.TXT.decode (packet.js:618–633) concatenates all strings into a single this.data value, discarding the boundaries. Applications that rely on multi-string TXT records (SPF, DKIM) will receive incorrect data.

The encoder handles arrays correctly; the asymmetry means decode→encode is lossy.

PS: it was back around node.js 0.11 or so when c-ares had this exact same issue, and had to resolve it this same way. It affected us over in Haraka, a SMTP daemon, where we consume a lot of SPF and DKIM records.

[Copilot]

Pull request overview

This PR fixes lossy decoding of TXT/SPF records by preserving RFC 1035 §3.3.14 <character-string> boundaries on decode (so multi-string TXT records like SPF/DKIM round-trip correctly), and updates the project’s public behavior to return TXT data as an array consistently.

Changes:

• Update Packet.Resource.TXT.decode (and SPF alias) to return data as an array of strings, preserving per-string boundaries.
• Update server/packet tests to assert TXT data as string[] (including a new boundary-preservation test).
• Mark the change as breaking via a major version bump to 3.0.0 and a changelog entry.

Reviewed changes

Copilot reviewed 5 out of 5 changed files in this pull request and generated 3 comments.

Show a summary per file

File	Description
packet.js	Changes TXT/SPF decode to preserve <character-string> boundaries by producing an array of strings.
test/packet.js	Updates TXT-related assertions and adds a test ensuring multi-string TXT boundaries are preserved.
test/server.js	Updates end-to-end expectations so TXT answers return data as string[].
package.json	Bumps package version to 3.0.0 to reflect the breaking API change.
CHANGELOG.md	Documents the breaking TXT data shape change and the decode fix.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.