Bring e2e/ back in sync, put it under CI, and drop the upgrade section - #19
Merged
Conversation
`e2e/` declares its own `[workspace]`, so the root build, clippy and test runs never reach it and CI never compiled it. It had drifted on two counts. Its lockfile still recorded `unicity-token` at 0.1.0 and carried a `num-traits` dependency the SDK dropped some time ago. Building with `--locked` now fails if that happens again. The live test read `e2e/unicity-service`, a `key: value` file that exists nowhere in the tree, is not tracked, is not covered by `e2e/.gitignore`, and is mentioned in no README. Anyone following the root README's run instruction got a panic naming a file they had no way to know the format of, and anyone who guessed and created one would have had an API key sitting untracked but un-ignored. It now reads `e2e/.env` through `dotenvy` like the three examples and the demo crate do, with the same `UNICITY_GATEWAY` / `UNICITY_API_KEY` / `UNICITY_TRUSTBASE` names and the same "process environment wins" precedence, so CI can supply a key without writing a file. The trust base path follows `UNICITY_TRUSTBASE` rather than being hardcoded, matching `examples/mint.rs`, and a missing file now says which path it tried. CI gains an fmt, clippy and `--locked` build of the demo crate.
There is nothing to upgrade from. This crate had never been released: it sat at 0.1.0, unpublished, and 3.0.1 is its first release. A section walking through migration from wire versions no consumer ever received is noise in the README, and the parts of it that were corrections rather than protocol changes read as if they had once been shipped behaviour. The migration story belongs in the v3.0.1 release notes, where it is addressed to someone comparing this crate against the TypeScript and Java SDKs rather than against an earlier Rust release. The interop line near the top stays, because which SDK and aggregator versions a 3.0.1 client talks to is a current fact rather than a historical one.
ristik
added a commit
that referenced
this pull request
Aug 31, 2026
Bring e2e/ back in sync, put it under CI, and drop the upgrade section
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
The
e2e/directory declares its own[workspace], so the rootcargo build,cargo clippyandcargo testnever reach it and CI never compiled it. It had drifted on two counts, both invisible to every check the repo runs.The lockfile was stale
e2e/Cargo.lockrecordedunicity-tokenat 0.1.0 (the crate is 3.0.1) and still carried anum-traitsdependency the SDK dropped some time ago. CI now builds the demo with--locked, so this fails rather than drifting.The live test read a file that does not exist
tests/e2e.rsread connection parameters frome2e/unicity-service, akey: valuefile thate2e/.gitignore(which covers.env,.env.*),So the run instruction in the root README panicked with
e2e/unicity-service, naming a file whose format was documented nowhere. Worse, anyone who guessed the format and created one would have had a testnet API key sitting in the working tree, untracked but not ignored, onegit add -Afrom being committed.Everything else in the repo already shares one convention:
examples/mint.rs,examples/transfer.rs,examples/split.rsand thee2e/demo crate all reade2e/.envthroughdotenvywithUNICITY_GATEWAY/UNICITY_API_KEY/UNICITY_TRUSTBASE, documented ine2e/.env.example,e2e/README.mdandexamples/README.md, and all let the process environment take precedence. The live test now does the same, so CI can supply a key without writing a file.The trust base path follows
UNICITY_TRUSTBASErather than being hardcoded, matchingexamples/mint.rs, and a missing file now reports which path it tried.The README carried an upgrade section for releases that never happened
This crate had never been released: it sat at 0.1.0, unpublished, and 3.0.1 is its first release. A section walking through migration from wire versions no consumer ever received is noise, and the parts of it that were corrections rather than protocol changes read as if they had once been shipped behaviour. It is dropped.
The migration story stays in the v3.0.1 release notes, where it is addressed to someone comparing this crate against the TypeScript and Java SDKs rather than against an earlier Rust release. The interop line near the top stays, because which SDK and aggregator versions a 3.0.1 client talks to is a current fact rather than a historical one.
CI
A step running fmt, clippy and a
--lockedbuild of the demo crate, since nothing else does.Validation
cargo fmt --all --check,cargo clippy --all-targets --all-features -- -D warningscargo test --all-features: 122 unit, 19 transport, 10 cross-SDK, 1 ignored livecargo fmt --manifest-path e2e/Cargo.toml --check,cargo clippy --manifest-path e2e/Cargo.toml --all-targets -- -D warnings,cargo build --manifest-path e2e/Cargo.toml --lockedNo library code changes, so v3.0.1 as published is unaffected.