Create new v2_to_v2 integration test by benalleng · Pull Request #85 · LtbLightning/payjoin-ffi

benalleng · 2025-04-29T13:46:19Z

Creates new v2_to_v2 integration test

python/test/payjoin_integration_test.py

We need more control over psbts in the integration tests so I am using my branch of bitcoin_ffi until we have that PR merged into main.

Callback methods meant to be overridden by the foreign languange should be exposed with [`with_foreign`](https://mozilla.github.io/uniffi-rs/0.27/foreign_traits.html#1-define-a-rust-trait)

This adds updated integration tests to work with 0.23 and beyond as it now utilizes the updated payjoin_test_utils.

Using filenames that start with test will allow us to run a more seamless workflow.

DanGould · 2025-05-12T15:32:49Z

it does not have access to the same environments by default as ubunut-latest

In case this build fails (Success pls???)

Is there a good reason not to use ubuntu-latest?

benalleng · 2025-05-12T15:36:28Z

it does not have access to the same environments by default as ubunut-latest

In case this build fails (Success pls???)

Is there a good reason not to use ubuntu-latest?

Not really, I was going to try that soon if this current attempt doesn't work, I was just hoping to keep the minimal manylinux image if possible, it might be but perhaps its not worth it

DanGould

cACK 1364c24

Does CI work on commits before the one I cACKed? Or would there be errors if not for the CI changes in the final commit? That's my only real concern assuming integration testing is operating as I imagine it is. The other comments are just notes for the future.

DanGould · 2025-05-13T01:04:02Z

src/test_utils.rs

+    pub fn fetch_ohttp_keys(&self) -> Result<crate::OhttpKeys, crate::io::IoError> {
+        let runtime = RUNTIME.lock().expect("Lock should not be poisoned");
+        runtime.block_on(async {
+            self.0.lock().await.fetch_ohttp_keys().await.map_err(Into::into).map(Into::into)
+        })


The repeated let runtime =, runtime.block_on makes me think this could be dried up by a delarative macro. But there's no rush on this, it can get in after this is merged, just a prettier way to write.

DanGould · 2025-05-13T01:09:58Z

python/requirements-dev.txt

 toml==0.10.2
 yapf==0.43.0
-
+httpx==0.28.1


very pleased we're able to use httpx here instead of exposing rust's reqwest. I'm curious, though not particularly eager, to see if the native http clients for some foreign language can replace fetch_ohttp_keys so we don't have to ship a whole rust TLS stack with our binaries, too.

benalleng · 2025-05-13T16:34:58Z

021dab3 causes a CI falure as the integration test has been added to the workflow there but the followup changes to add support redis are not in until 1364c24

DanGould · 2025-05-13T16:52:21Z

021dab3 causes a CI falure as the integration test has been added to the workflow there but the followup changes to add support redis are not in until 1364c24

shall those commits then be squashed or tests ignored so that each commit in turn passes CI and git bisect works?

benalleng · 2025-05-13T16:57:31Z

I think we should just squash them such that the integration test doesn't get added to the Ci until the redis compatibility is also included as 021dab3 really doesn't do anything other than cause a CI failure with the new tests added

Now that the test is working we should add it to the workflow. I have left this as a seperate commit in case it adds too much time to the workflows. By adding ubuntu latest for linux and a special workflow command for macos we are able to enable redis to create a compatible workflow for running the new integration tests.

DanGould · 2025-05-13T17:36:04Z

How come the mac version is unable to use redis provided by payjoin-test-utils?

DanGould

tACK 83359b5

I ran the python test on my local machine.

I'm still confused why redis setup is re-scripted in the CI yml though. Why not use the payjoin-test-utils docker version so that it's the same as rust? Perhaps this needs to be done in a follow up.

DanGould · 2025-05-13T17:42:52Z

Hmmm but it also seems like there might be a silent test failure in the python tests. Look at the 'Run tests' section

 OHTTP relay binding to port [::]:33681
2025-05-13T17:02:36.074649Z ERROR ohttp_relay::bootstrap::connect: server io error: Transport endpoint is not connected (os error 107)
test_integration_v2_to_v2 (test.test_payjoin_integration_test.TestPayjoin.test_integration_v2_to_v2) ... Executing <Task pending name='Task-2' coro=<TestPayjoin.test_integration_v2_to_v2() running at /home/runner/work/payjoin-ffi/payjoin-ffi/python/test/test_payjoin_integration_test.py:92> cb=[_run_until_complete_cb() at /opt/hostedtoolcache/Python/3.13.3/x64/lib/python3.13/asyncio/base_events.py:181] created at /opt/hostedtoolcache/Python/3.13.3/x64/lib/python3.13/asyncio/runners.py:100> took 3.964 seconds
2025-05-13T17:02:38.449256Z ERROR payjoin_directory: Error serving connection: hyper::Error(Shutdown, Os { code: 107, kind: NotConnected, message: "Transport endpoint is not connected" })
Executing <Task finished name='Task-2' coro=<TestPayjoin.test_integration_v2_to_v2() done, defined at /home/runner/work/payjoin-ffi/payjoin-ffi/python/test/test_payjoin_integration_test.py:71> result=None created at /opt/hostedtoolcache/Python/3.13.3/x64/lib/python3.13/asyncio/runners.py:100> took 0.127 seconds
ok

is this a silent failure or some loop that's waiting and errors the first go around? I think it might just be waiting but ERROR prints are frightening.

benalleng · 2025-05-13T17:50:44Z

some loop that's waiting and errors the first go around?

I'm pretty sure its the latter as the test still resolves to the end to an ok response with all the asserts we have in place, and at least one of these are the same errors on the rust-payjoin.

Database running on 127.0.0.1:32865
Directory server binding to port [::]:45329
OHTTP relay binding to port [::]:37679
session: Receiver {
    context: SessionContext {
        ...
    },
}
2025-05-13T17:49:46.254400Z ERROR ohttp_relay::bootstrap::connect: server io error: Transport endpoint is not
 connected (os error 107)
test integration::v2::v2_to_v2 ... ok

I agree though that it would be nice to not get these errors though

benalleng force-pushed the create-new-v2-tests branch from 8514ad5 to af5c1ff Compare April 29, 2025 13:48

benalleng changed the title ~~Create new v2 tests~~ Create new v2_to_v2 integration test Apr 29, 2025

benalleng force-pushed the create-new-v2-tests branch 6 times, most recently from 9bafb19 to ab25849 Compare April 30, 2025 18:00

DanGould mentioned this pull request Apr 30, 2025

Bind payjoin-test-utils #55

Closed

benalleng force-pushed the create-new-v2-tests branch 5 times, most recently from 549c956 to 0262683 Compare May 1, 2025 12:51

arminsabouri reviewed May 1, 2025

View reviewed changes

python/test/payjoin_integration_test.py Outdated Show resolved Hide resolved

benalleng force-pushed the create-new-v2-tests branch 13 times, most recently from 85559df to 8c745bc Compare May 7, 2025 15:34

benalleng force-pushed the create-new-v2-tests branch 2 times, most recently from 5b754af to aa6c733 Compare May 8, 2025 16:31

benalleng and others added 4 commits May 9, 2025 12:51

Introduce bitcoin_ffi Psbt struct and test_utils runtime

d1c9ff7

We need more control over psbts in the integration tests so I am using my branch of bitcoin_ffi until we have that PR merged into main.

Expose uniffi persistence traits as foreign

fab9db6

Callback methods meant to be overridden by the foreign languange should be exposed with [`with_foreign`](https://mozilla.github.io/uniffi-rs/0.27/foreign_traits.html#1-define-a-rust-trait)

Introduce python integration tests v2_to_v2

2f562f5

This adds updated integration tests to work with 0.23 and beyond as it now utilizes the updated payjoin_test_utils.

rename test files

265ad68

Using filenames that start with test will allow us to run a more seamless workflow.

benalleng force-pushed the create-new-v2-tests branch from bda8b68 to 4ef75b6 Compare May 9, 2025 16:51

This comment was marked as resolved.

Sign in to view

benalleng force-pushed the create-new-v2-tests branch 3 times, most recently from 82d4418 to 4c6fd51 Compare May 12, 2025 15:31

benalleng force-pushed the create-new-v2-tests branch 6 times, most recently from a626dd8 to 1364c24 Compare May 12, 2025 18:10

DanGould self-requested a review May 13, 2025 01:06

DanGould reviewed May 13, 2025

View reviewed changes

benalleng force-pushed the create-new-v2-tests branch from 1364c24 to 83359b5 Compare May 13, 2025 17:00

DanGould approved these changes May 13, 2025

View reviewed changes

DanGould merged commit d3a2b6f into LtbLightning:main May 13, 2025
10 checks passed

spacebear21 mentioned this pull request Jun 3, 2025

Send and Receive test cases for Python #6

Closed

Comments

Conversation

benalleng commented Apr 29, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

This comment was marked as resolved.

DanGould commented May 12, 2025

Uh oh!

benalleng commented May 12, 2025

Uh oh!

DanGould left a comment

Choose a reason for hiding this comment

Uh oh!

DanGould May 13, 2025

Choose a reason for hiding this comment

Uh oh!

DanGould May 13, 2025

Choose a reason for hiding this comment

Uh oh!

benalleng commented May 13, 2025

Uh oh!

DanGould commented May 13, 2025

Uh oh!

benalleng commented May 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

DanGould commented May 13, 2025

Uh oh!

DanGould left a comment

Choose a reason for hiding this comment

Uh oh!

DanGould commented May 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

benalleng commented May 13, 2025 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Uh oh!

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

benalleng commented Apr 29, 2025 •

edited

Loading

benalleng commented May 13, 2025 •

edited

Loading

DanGould commented May 13, 2025 •

edited

Loading

benalleng commented May 13, 2025 •

edited

Loading