PDK Async Dev Updates - Dec 15 2025

[arminsabouri]

Contributors please post what you have been busy doing, what you doing this week and what are your constraints. Please post them as comments on this discussion by EOD.

[chavic]

Last week:

• I've been working on addressing issues with Move the Default Payjoin FFI API to primitives #1191 and am still on it
• Waiting for a review of uniffi-dart async support

This week:

Once these are fixed, I'll continue with

• Define and add a dedicated primitive-conversion error category in the send/receive public error enums; wire all malformed base64/hex/txid/vout/script cases to it at the FFI boundary.
• Regenerate bindings (JS/Dart/Python) and update tests in each language to assert the new error on bad primitives.
• Update relevant documentation

[zealsham]

Last week:

• Got up to speed with Nix , following the lessons on nix.dev
• fixed minor issues in my fallback handling v1 tests
• started writting v2 tests for fallback handling
• synced up with an LDK btrust grantee who wants to work on integrating payjoin on LDK

This week

• Finish up v2 fallback handling tests
• address issue 1227
• draft LDK payjoin integration PR.

[mehmetefeumit]

Last Week:

• Addressed the outpoint check in Receiver by making it skip the monitoring if the sender has non-SegWit inputs.
• Added integration tests for different address types and sender actions (fallback, Payjoin, non-Segwit, etc.)
• Finalized the BDK CLI change by fixing the issues raised in the team's testing. The PR is approved, and will be merged approx. in 1 week.

This Week:

• Working on refining the repository README.md and mirroring it in docsrs.
• Going to check with Shehu on BDK CLI persistance and see if we can add that to the CLI together.
• Open for other issue recommendations. Feel free to reach out if there is something you'd need worked on in parallel.

[spacebear21]

Last week:

• Merge Fold ohttp-relay into rust-payjoin monorepo #1215
• Update Directory Service Tracking Issue #941 sub-items
• Hack on unified Payjoin service. Draft PR posted today: Unified payjoin service #1232

This week:

• Continue payjoin-service work and get it merged
• Review/merge Move the Default Payjoin FFI API to primitives #1191
• Explore making rust-payjoin async-compatible per discussion in should callbacks accepting functions be async? #816 (comment)

[DanGould]

Curious if there was discussion about bringing the ohttp-relay/flake.nix into this repository as a module, or if that was axed because you knew the follow up with unified service was coming?

[spacebear21]

We chose to go with one big flake to fast-track the unified service work (all that was really needed was the nginx dependency for the ohttp-relay integration tests, which we plan to remove in favor of ACME anyways). There has been discussion about modularizing our flake following the nix-bitcoin model in follow ups.

[benalleng]

Speaking of the ohttp-relay being migrated into the monorepo, should we archive and move the open issues/ PRs over to the rust-payjoin repo?

[nothingmuch]

Curious if there was discussion about bringing the ohttp-relay/flake.nix into this repository as a module, or if that was axed because you knew the follow up with unified service was coming?

do you mean carl's nixos service module?

we should definitely have just one flake.nix at the top of the monorepo, with a unified devshell. multiple flakes in a single repo is possible, but complicates things needlessly

but think of flake.nix as lib.rs, it's just an entry point, the code can be split up into separate files and these files can be organized into subdirectories where they make sense to avoid polluting the project root with a bunch of files

[DanGould]

I meant carl's service module but as module.nix referred to in the top level flake.nix. If it's all one binary this may not make sense.

[arminsabouri]

• Been trying to stay on top off reviews in PDK. Worked with efe to fix the monitoring typestate Skip check_payment in Receiver<Monitor> if the sender is using non-segwit address #1218 and fixing our p2tr input weight calculation Derive P2TR input weight from witness data #1200
• Met with wizardsardine team to explore our treasury options. notes posted and shared on proton.
• Working with nothingmuch on reserach overview docs. Spending time today and last week reading on async BFT protocols. Exploring ideas that are applicable to multiparty payjoins in the byzantine fault tolerant setting. See personal notes Starting working on writing some notes for the semi honest setting.
• Finished OHTTP client side PR and tests Add Async OHTTP client bitcoindevkit/rust-esplora-client#145 . Concept Ack recieved. Will need to do a refactor after bitreq is introduced.
• Looking to get the nostr OHTTP work rebased and in a good shape before Q1 -- as planned.
  I will be intermittenly online during christmas week.

[benalleng]

Last Week

• Reviewing and helping out with some of the recent nix related PRs including merging Introduce treefmt-nix as a universal formatter infra #1221
• Review help where I was needed to try and take up some slack

This Week

• This week I hope to have a PR that has the full receiver set on liana to give some better perspective for their team without being the entire POC in a single commit
• continue to help out with the nix PRs as we see some inconsistency discussed here flake: build dev profile instead of release #1228 caused by some of the new checks slowing down some people's commit workflow

[0xZaddyy]

Last week:

• Conducted an audit of types with unintentional trait implementations.

• Spent a lot of time researching and reviewing relevant research papers (thanks to Yuval and Armin for the pointers).

This week:

• Wrapping up work on Audit serialization trait implementations #1223

• Continuing deeper research on privacy.

• Beginning code review work.

[nothingmuch]

• mainly focused on the multiparty protocol spec work
• helped a bit with nix related PRs