Skip to content

fix(peer-deps): ⬆️ update dependency amqplib to ^2.4.0 - #111

Open
renovate[bot] wants to merge 1 commit into
mainfrom
renovate/amqplib-2.x
Open

renovate[bot] wants to merge 1 commit into
mainfrom
renovate/amqplib-2.x

Conversation

@renovate

@renovate renovate Bot commented Nov 5, 2026 •

Copy link
Copy Markdown
Contributor

This PR contains the following updates:

Package Change Age Confidence
amqplib (source) ^2.2.0 → ^2.4.0 age confidence

Release Notes

amqp-node/amqplib (amqplib)

v2.4.0

Compare Source

  • Decode a table that contains a key named !, and is the value of a header, argument or array element, as { '!': 'table', value: <the table> }. Previously it was decoded as the bare table, which the encoder reads as an explicit type such as { '!': 'timestamp', value: 1 }. Publishing headers that had been received with such a table therefore threw Unknown type to encode, or sent a different value without an error when the key held a valid type name. A received table can now also be told apart from a received timestamp or decimal. Upgrade note: a consumer that reads a table containing a ! key must now read it from value. Tables without a ! key, and a ! key at the top level of the headers, are decoded as before
  • Add table as an explicit type. { '!': 'table', value: { ... } } is encoded as a table whose keys are taken literally, so it can contain a key named !

v2.3.3

Compare Source

  • Encode numbers below -2^63, NaN, -Infinity and fractions of 2^50 or more as doubles in headers and arguments tables. Previously publish threw a RangeError for these, including when the table had been received from the broker and was being published again. Numbers that could be encoded before are encoded as before
  • Encode an explicit timestamp of 2^64, or an explicit long of 2^63, in a headers or arguments table as the largest value of that type, as versions before 2.0.1 did. The largest timestamp is decoded as 2^64, because 2^64 - 1 cannot be represented as a number, so since 2.0.1 a received table containing it could not be published again. Larger numbers still throw a RangeError
  • Encode a message timestamp property of 2^64, or any other unsigned 64-bit method field of 2^64, as the largest value in the same way, so a message received with a timestamp of 2^64 - 1 can be published again (also a regression in 2.0.1)

v2.3.2

Compare Source

  • Encode and decode nested field tables and arrays without recursion. Previously a table nested a few thousand levels deep, which RabbitMQ accepts and delivers, overflowed the call stack while a consumer decoded the message headers. The connection closed with Maximum call stack size exceeded, the message was requeued, and each reconnecting consumer failed the same way. There is no depth limit, and everything that encoded or decoded before produces the same result (GHSA-gfjv-mq5g-xjfw, reported by Dickson Wu (@​DicksonWu654))
  • Decode a field table key named __proto__ as an ordinary property. Previously it replaced the decoded table's prototype: a table value was inherited from, a null value removed the prototype, and any other value was dropped. With a byte array value followed by a numeric key, decoding threw a TypeError and closed the connection in the same way. Upgrade note: headers received with a __proto__ key now have it as an own property (GHSA-gfjv-mq5g-xjfw)

v2.3.1

Compare Source

  • Throw a clear RangeError from publish when an encoded headers or arguments table exceeds the 64 KiB encoding buffer. Previously the outcome depended on where the boundary fell: a string or Buffer value crossing the end of the buffer was silently truncated, producing a frame whose table length prefix did not match its contents, which RabbitMQ rejected by closing the connection with a 541 INTERNAL_ERROR after publish had returned; at other positions a bare offset is out of range error escaped from publish instead (fixes #​867)

v2.3.0

Compare Source

  • Reject a connection.tune whose frameMax is below the protocol minimum of 4096 (the spec allows 0 or at least frame-min-size). Previously such a value was accepted, and the first publish whose encoded frames total 2048 bytes or more entered a loop in sendContent that never advanced, exhausting memory. sendContent now also throws if frameMax leaves no room for a body frame (GHSA-37f9-7m79-6866, reported by Rizwan Bedekar (@​r-bedekar); fix based in part on their patch and tests)
  • Add a maxFrameSize connection option, also settable as a URL query parameter, bounding the payload size of frames accepted from the server. The default is 64 MiB, or the negotiated frameMax if that is larger, so raising frameMax needs no second option; 0 disables the limit. Incoming frame sizes were previously unchecked, so a frame header declaring up to 4 GiB made the client buffer, and repeatedly re-copy, that much data before looking at it. An oversized frame is now refused from its header alone and closes the connection with Frame size N exceeds maxFrameSize M. The limit is deliberately separate from the negotiated frameMax, since some brokers send larger frames without splitting them (see #​784). Upgrade note: a broker that delivers an unsplit frame larger than the limit will now have that connection closed; pass maxFrameSize: 0, or a higher value, to restore the previous behaviour (GHSA-37f9-7m79-6866)

v2.2.1

Compare Source

  • Return the server acknowledgement from promise-based channel.cancel, matching the callback API (thanks @​edsonsims).
  • Type cancel replies as Replies.Cancel ({ consumerTag }) in both the promise and callback API declarations, which previously claimed Replies.Empty.
  • Fix callback-based waitForConfirms never calling back when invoked after the confirm channel closed with unconfirmed messages; it now yields a channel closed error, matching the promise API (thanks @​youdie006).
  • waitForConfirms (both APIs) now always fails with channel closed after the channel closes with unconfirmed messages. Previously it reported success if every outstanding message had its own callback, or on any call after the first.
  • The callback API's waitForConfirms callback is now required in the TypeScript declarations, matching the docs. Calling it without one threw at runtime.
  • Replace .npmignore with a files allowlist in package.json so the published package no longer includes tests, tooling config, dotfiles or the defs generator. Install size drops from 577 kB to 358 kB (fixes #​859, thanks @​rubiesonthesky)

Configuration

📅 Schedule: (in timezone Europe/Madrid)

  • Branch creation
    • At any time (no schedule defined)
  • Automerge
    • At any time (no schedule defined)

🚦 Automerge: Disabled by config. Please merge this manually once you are satisfied.

♻ Rebasing: Whenever PR becomes conflicted, or you tick the rebase/retry checkbox.

🔕 Ignore: Close this PR and you won't be reminded about this update again.


  • If you want to rebase/retry this PR, check this box

This PR was generated by Mend Renovate. View the repository job log.

@renovate renovate Bot added the dependencies label Nov 5, 2026
@codecov

codecov Bot commented Nov 5, 2026

Copy link
Copy Markdown

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!

@renovate
renovate Bot force-pushed the renovate/amqplib-2.x branch 2 times, most recently from 8905569 to 67ed3af Compare November 5, 2026 19:29
@renovate renovate Bot changed the title fix(peer-deps): ⬆️ update dependency amqplib to ^2.2.1 fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.0 Nov 8, 2026
@renovate
renovate Bot force-pushed the renovate/amqplib-2.x branch from 67ed3af to dace84c Compare November 8, 2026 18:14
@renovate renovate Bot changed the title fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.0 fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.1 Nov 8, 2026
@renovate
renovate Bot force-pushed the renovate/amqplib-2.x branch 2 times, most recently from cf1cd32 to cd11e46 Compare November 11, 2026 18:30
@renovate renovate Bot changed the title fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.1 fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.3 Nov 16, 2026
@renovate
renovate Bot force-pushed the renovate/amqplib-2.x branch from cd11e46 to 9d3d136 Compare November 16, 2026 04:21
@renovate renovate Bot changed the title fix(peer-deps): ⬆️ update dependency amqplib to ^2.3.3 fix(peer-deps): ⬆️ update dependency amqplib to ^2.4.0 Nov 16, 2026
@renovate
renovate Bot force-pushed the renovate/amqplib-2.x branch from 9d3d136 to c2852c8 Compare November 16, 2026 14:51

This branch has not been deployed

No deployments
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

0 participants