Arseniy Klempner 16253026c6
feat: implement lp-v3 error codes with backwards compatibility (#2501)
* feat: implement LightPush v3 protocol support

Add comprehensive LightPush v3 protocol implementation with:

Core Features:
- LightPush v3 protocol codec and multicodec detection
- Status code-based error handling and validation
- Protocol version inference and compatibility layers
- Enhanced error types with detailed failure information

Protocol Support:
- Automatic v3/v2 protocol negotiation and fallback
- Status code mapping to LightPush error types
- Protocol version tracking in SDK results
- Mixed protocol environment support

Testing Infrastructure:
- Comprehensive v3 error code handling tests
- Mock functions for v3/v2 response scenarios
- Protocol version detection and validation tests
- Backward compatibility verification

Implementation Details:
- Clean separation between v2 and v3 response handling
- Type-safe status code validation with isSuccess helper
- Enhanced failure reporting with protocol version context
- Proper error propagation through SDK layers

This implementation maintains full backward compatibility with v2
while providing enhanced functionality for v3 protocol features.

* feat: handle both light push protocols

* fix: unsubscribe test

* feat: consolidate lpv2/v3 types

* feat(tests): bump nwaku to 0.36.0

* fix: remove extraneous exports

* fix: add delay to tests

* fix: remove protocol result types

* feat: consolidate light push codec branching

* fix: revert nwaku image

* fix: remove multicodec

* fix: remove protocolversion

* feat: simplify v2/v3 branching logic to use two stream managers

* fix: remove unused utils

* fix: remove comments

* fix: revert store test

* fix: cleanup lightpush sdk

* fix: remove unused util

* fix: remove unused exports

* fix: rename file from public to protocol_handler

* fix: use proper type for sdk result

* fix: update return types in filter

* fix: rebase against latest master

* fix: use both lightpush codecs when waiting for peer

* fix: handle both lp codecs

* fix: remove unused code

* feat: use array for multicodec fields

* fix: add timestamp if missing in v3 rpc

* fix: resolve on either lp codec when waiting for peer

* fix: remove unused util

* fix: remove unnecessary abstraction

* feat: accept nwaku docker image as arg, test lp backwards compat

* fix: revert filter error

* feat: add legacy flag to enable lightpushv2 only

* Revert "feat: accept nwaku docker image as arg, test lp backwards compat"

This reverts commit 857e12cbc73305e5c51abd057665bd34708b2737.

* fix: remove unused test

* feat: improve lp3 (#2597)

* improve light push core

* move back to singualar multicodec property, enable array prop only for light push

* implement v2/v3 interop e2e test, re-add useLegacy flag, ensure e2e runs for v2 and v3

* fix v2 v3 condition

* generate message package earlier

* add log, fix condition

---------

Co-authored-by: Sasha <118575614+weboko@users.noreply.github.com>
Co-authored-by: Sasha <oleksandr@status.im>
2025-09-05 00:52:37 +02:00
..
2025-07-21 15:54:05 +10:00
2025-04-18 20:44:22 +02:00
2022-12-02 15:54:30 +11:00

Description

This package contains tests for the js-waku library.

Pre-requisites

Some of the tests from this package require a running nwaku node. These nodes are setup to be run in a docker container. Therefore, you need to have docker installed on your machine to run the tests.

Running interop tests

  • The tests by default run against an nwaku node with the image name specified in nwaku.ts and packages/tests/package.json. The tests can be run against a different image by setting the environment variable WAKUNODE_IMAGE to the desired image.

  • Whatever WAKUNODE_IMAGE is set to, the tests will run against that image. If the image is not available locally, the tests will pull the image from the docker hub. You can run the tests by running the following command:

    WAKUNODE_IMAGE=explicit-image-name npm run test:node
    

    Or against the default docker image by running:

    npm run test:node
    
  • You can also run the tests against a local nwaku by setting the environment variable WAKUNODE_IMAGE to the name of the image. The tests will then run against the local image.

    • For example, to run the tests against a local checkout of nwaku, build the image first manually. You can build the image by running the following command:

      docker build path-to-dockerfile -t image-name
      

      Then, you can run the tests by running the following command:

      WAKUNODE_IMAGE=image-name npm run test:node
      
  • Locally, tests are executed serially, allowing the use of .only for focused testing. If you wish to run all tests locally and expedite the process, you can enable parallel execution in the Mocha configuration.

  • Logs from nwaku nodes can be found in packages/tests/logs folder from latest execution.

Running tests in the CI

  • Tests are being run on standard Ubuntu GitHub Actions instances.
  • To speed up execution, we run tests in parallel. After numerous attempts, we determined that using 6 threads strikes the best balance between execution speed and test reliability. Using more than this doesn't significantly decrease execution time and might even slow it down.
  • To address occasional test flakiness, primarily due to Docker containers starting and stopping for each test and the concurrent execution of tests, we utilize the Mocha retry mechanism.