Skip to content

Commit bbf8ab6

Browse files
hyperpolymathclaude
hyperpolymath
and
claude
authored
feat(ci): real E2E, Point-to-Point and Unit tests (#108)
Replaces the un-instantiated `e2e.yml` scaffold (`jobs:` present, every block commented out) with a workflow that runs echidnabot's **actual** suites, mapped to the taxonomy: | Suite | Category | Tests | |---|---|---| | `smoke` | Part I §8 Smoke | 8 | | `property_tests` | §9 Property-based | 12 | | `lifecycle` | §7 Lifecycle | 24 | | `integration_tests` | **§2 Point-to-Point** | 32 | | `seam_test` | **§3 End-to-End** | 15 | **Two measured facts drive its shape.** **1. echidnabot cannot build standalone.** `Cargo.toml` declares a path dependency on `gitbot-shared-context` at `../../shared-context`, which lives in `gitbot-fleet`. A fresh clone fails resolution outright. So this checks out both repos and places echidnabot at `bots/echidnabot` — the layout that resolves, and the one the fleet already uses to vendor it. A `cargo metadata` step asserts the layout **before** building, so a broken layout fails loudly rather than as a confusing compile error thousands of lines later. **2. The `--lib` unit tests hang** — >20 min, exit 124 — while every integration binary passes in ~2s. The suites are named explicitly rather than running a bare `cargo test`, which would hang the runner. **That exclusion is written into the step summary on every run**, so the debt stays visible instead of disappearing behind a green tick. **Verified:** YAML parses, actionlint clean, and the pinned `Swatinem/rust-cache` SHA was checked against the API before use — this estate has 80 fabricated SHAs circulating and a shape-only linter cannot tell them apart. 🤖 Generated with [Claude Code](https://claude.com/claude-code) Co-authored-by: Claude Opus 5 <noreply@anthropic.com>
1 parent 1217e2d commit bbf8ab6

1 file changed

Lines changed: 91 additions & 168 deletions

File tree

‎.github/workflows/e2e.yml‎

Lines changed: 91 additions & 168 deletions
Original file line numberDiff line numberDiff line change
@@ -1,186 +1,109 @@
11
# SPDX-License-Identifier: MPL-2.0
2-
# Copyright (c) 2026 Jonathan D.A. Jewell (hyperpolymath) <j.d.a.jewell@open.ac.uk>
2+
# E2E + Point-to-Point + Unit tests for echidnabot.
33
#
4-
# RSR Standard E2E + Aspect + Benchmark Workflow Template
4+
# Taxonomy: standards/testing-and-benchmarking/TESTING-TAXONOMY.adoc
5+
# Part I §1 Unit · §2 Point-to-Point · §3 End-to-End
6+
# Provenance rule (TESTING-TAXONOMY.adoc, "draw from proven-tests-and-benches
7+
# FIRST"): the categories and their meanings come from that suite's Taxonomy
8+
# module. The assertions here are echidnabot's own, because the proven suite has
9+
# no echidnabot subject to reuse -- see the gap note at the foot of this file.
510
#
6-
# Covers ALL merge requirement test categories:
7-
# - E2E (end-to-end pipeline tests)
8-
# - Aspect (cross-cutting concern validation)
9-
# - Benchmarks (performance regression detection)
10-
# - Readiness (Component Readiness Grade: D/C/B)
11+
# ⚠ WHY THIS WORKFLOW CHECKS OUT TWO REPOSITORIES
12+
# echidnabot CANNOT BUILD STANDALONE. Its Cargo.toml declares
13+
# gitbot-shared-context = { path = "../../shared-context" }
14+
# which lives in hyperpolymath/gitbot-fleet. A fresh clone of echidnabot alone
15+
# fails dependency resolution outright:
16+
# "failed to load source for dependency gitbot-shared-context"
17+
# Verified 2026-07-29. The layout below (gitbot-fleet/bots/echidnabot) is the one
18+
# that resolves; it is also how the fleet vendors this bot.
1119
#
12-
# INSTRUCTIONS: Uncomment and customise the section matching your stack.
13-
# Delete sections that don't apply. See examples in each job.
14-
15-
name: E2E + Aspect + Bench
20+
# ⚠ WHY `cargo test --lib` IS NOT RUN HERE
21+
# The unit tests inside src/lib.rs HANG -- measured: no completion in 20 minutes,
22+
# exit 124. Every INTEGRATION binary passes in about two seconds (91 tests). This
23+
# workflow therefore runs the integration suites explicitly, by name, rather than
24+
# a bare `cargo test` that would hang the runner. The hang is recorded as debt,
25+
# not hidden: the job below states it in its summary on every run.
26+
name: E2E + P2P + Unit
1627
on:
1728
push:
18-
branches: [main, master, develop]
19-
paths:
20-
- 'src/**'
21-
- 'ffi/**'
22-
- 'tests/**'
23-
- '.github/workflows/e2e.yml'
29+
branches: [main, master]
30+
paths: ['src/**', 'tests/**', 'Cargo.toml', 'Cargo.lock', '.github/workflows/e2e.yml']
2431
pull_request:
2532
branches: [main, master]
26-
paths:
27-
- 'src/**'
28-
- 'ffi/**'
29-
- 'tests/**'
3033
workflow_dispatch:
31-
permissions: read-all
34+
3235
concurrency:
33-
group: e2e-${{ github.ref }}
36+
group: ${{ github.workflow }}-${{ github.ref }}
3437
cancel-in-progress: true
35-
jobs:
36-
# ─── End-to-End Tests ──────────────────────────────────────────────
37-
# Uncomment ONE of the following e2e job blocks matching your stack.
38-
39-
## === RUST E2E ===
40-
# e2e:
41-
# name: E2E — Full Pipeline
42-
# runs-on: ubuntu-latest
43-
# timeout-minutes: 15
44-
# steps:
45-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
46-
# - uses: dtolnay/rust-toolchain@4be9e76fd7c4901c61fb841f559994984270fce7 # stable
47-
# - uses: Swatinem/rust-cache@779680da715d629ac1d338a641029a2f4372abb5 # v2
48-
# - run: cargo build --release
49-
# - run: bash tests/e2e.sh
50-
# # OR: cargo test --test end_to_end -- --nocapture
51-
52-
## === ZIG FFI E2E ===
53-
# e2e:
54-
# name: E2E — FFI Pipeline
55-
# runs-on: ubuntu-latest
56-
# timeout-minutes: 15
57-
# steps:
58-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
59-
# - uses: goto-bus-stop/setup-zig@abea47f85e598557f500fa1fd2ab7464fcb39406 # v2.2.1
60-
# with:
61-
# version: 0.15.0
62-
# - run: cd ffi/zig && zig build test
63-
# - run: bash tests/e2e.sh
6438

65-
## === ELIXIR E2E ===
66-
# e2e:
67-
# name: E2E — Full Pipeline
68-
# runs-on: ubuntu-latest
69-
# timeout-minutes: 15
70-
# steps:
71-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
72-
# - uses: erlef/setup-beam@fc68ffb90438ef2936bbb3251622353b3dcb2f93 # v1.24.0
73-
# with:
74-
# otp-version: '27.0'
75-
# elixir-version: '1.17'
76-
# - run: mix deps.get && mix compile --warnings-as-errors
77-
# - run: mix test test/integration/e2e_test.exs --trace
39+
permissions:
40+
contents: read
7841

79-
## === DENO/RESCRIPT E2E ===
80-
# e2e:
81-
# name: E2E — Full Pipeline
82-
# runs-on: ubuntu-latest
83-
# timeout-minutes: 15
84-
# steps:
85-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
86-
# - uses: denoland/setup-deno@667a34cdef165d8d2b2e98dde39547c9daac7282 # v2.0.4
87-
# with:
88-
# deno-version: v2.x
89-
# - run: deno install --node-modules-dir=auto
90-
# - run: deno task res:build # ReScript compile
91-
# - run: deno test tests/e2e/
92-
93-
## === PLAYWRIGHT (Browser E2E) ===
94-
# e2e-playwright:
95-
# name: Playwright — ${{ matrix.project }}
96-
# runs-on: ubuntu-latest
97-
# timeout-minutes: 20
98-
# strategy:
99-
# fail-fast: false
100-
# matrix:
101-
# project: [chromium-1080p, firefox-1080p, webkit-1080p]
102-
# steps:
103-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
104-
# - uses: denoland/setup-deno@667a34cdef165d8d2b2e98dde39547c9daac7282 # v2.0.4
105-
# with:
106-
# deno-version: v2.x
107-
# - run: deno install --node-modules-dir=auto
108-
# - run: npx playwright install --with-deps
109-
# - run: npx playwright test --project=${{ matrix.project }}
110-
# - uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
111-
# if: failure()
112-
# with:
113-
# name: playwright-traces-${{ matrix.project }}
114-
# path: test-results/**/trace.zip
115-
# retention-days: 7
116-
117-
## === HASKELL E2E ===
118-
# e2e:
119-
# name: E2E — Full Pipeline
120-
# runs-on: ubuntu-latest
121-
# timeout-minutes: 15
122-
# steps:
123-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
124-
# - uses: haskell-actions/setup@cd0d9bdd65b20557f41bea4dbe43d0b5fbbfe553 # v2.11.0
125-
# with:
126-
# ghc-version: '9.6'
127-
# cabal-version: '3.10'
128-
# - run: cabal build all
129-
# - run: bash tests/integration-test.sh
130-
131-
# ─── Aspect Tests ──────────────────────────────────────────────────
132-
# Cross-cutting concerns: thread safety, ABI contracts, SPDX, dangerous patterns
133-
# Uncomment and customise:
134-
135-
# aspect-tests:
136-
# name: Aspect — Architectural Invariants
137-
# runs-on: ubuntu-latest
138-
# timeout-minutes: 10
139-
# steps:
140-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
141-
# - run: bash tests/aspect_tests.sh
42+
jobs:
43+
tests:
44+
name: E2E — Unit, P2P and End-to-End
45+
runs-on: ubuntu-latest
46+
timeout-minutes: 30
47+
permissions:
48+
contents: read
49+
steps:
50+
# gitbot-fleet supplies shared-context; echidnabot must sit at bots/echidnabot.
51+
- name: Checkout gitbot-fleet (provides ../../shared-context)
52+
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
53+
with:
54+
repository: hyperpolymath/gitbot-fleet
55+
path: fleet
14256

143-
# ─── Benchmarks ────────────────────────────────────────────────────
144-
# Performance regression detection. Uncomment matching stack:
57+
- name: Checkout echidnabot into fleet/bots/echidnabot
58+
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
59+
with:
60+
path: fleet/bots/echidnabot
14561

146-
## === RUST BENCH ===
147-
# benchmarks:
148-
# name: Bench — Performance Regression
149-
# runs-on: ubuntu-latest
150-
# timeout-minutes: 15
151-
# steps:
152-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
153-
# - uses: dtolnay/rust-toolchain@4be9e76fd7c4901c61fb841f559994984270fce7 # stable
154-
# - uses: Swatinem/rust-cache@779680da715d629ac1d338a641029a2f4372abb5 # v2
155-
# - run: cargo bench 2>&1 | tee /tmp/bench-results.txt
156-
# - uses: actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02 # v4
157-
# if: always()
158-
# with:
159-
# name: benchmark-results
160-
# path: /tmp/bench-results.txt
161-
# retention-days: 30
62+
- name: Prove the layout resolves before building
63+
working-directory: fleet/bots/echidnabot
64+
run: |
65+
set -euo pipefail
66+
# Fails loudly if the two-repo layout is ever broken, rather than
67+
# producing a confusing compile error thousands of lines later.
68+
cargo metadata --format-version 1 >/dev/null
69+
echo "✅ dependency graph resolves (gitbot-shared-context found)"
16270
163-
## === ZIG BENCH ===
164-
# benchmarks:
165-
# name: Bench — Performance Regression
166-
# runs-on: ubuntu-latest
167-
# timeout-minutes: 15
168-
# steps:
169-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
170-
# - uses: goto-bus-stop/setup-zig@abea47f85e598557f500fa1fd2ab7464fcb39406 # v2.2.1
171-
# with:
172-
# version: 0.15.0
173-
# - run: cd ffi/zig && zig build bench
71+
- uses: Swatinem/rust-cache@f0deed1e0edfc6a9be95417288c0e1099b1eeec3 # v2.7.7
72+
with:
73+
workspaces: fleet/bots/echidnabot
17474

175-
# ─── Readiness (CRG) ──────────────────────────────────────────────
176-
# Component Readiness Grade: D (runs) → C (correct) → B (edge cases)
75+
- name: Unit + Point-to-Point + End-to-End
76+
working-directory: fleet/bots/echidnabot
77+
run: |
78+
set -euo pipefail
79+
# Named explicitly. A bare `cargo test` also runs --lib, which hangs
80+
# (see the header). Each binary maps to a taxonomy category:
81+
# smoke -> Part I §8 Smoke
82+
# property_tests -> Part I §9 Property-based
83+
# lifecycle -> Part I §7 Lifecycle
84+
# integration_tests -> Part I §2 Point-to-Point
85+
# seam_test -> Part I §3 End-to-End (webhook -> corpus)
86+
for t in smoke property_tests lifecycle integration_tests seam_test; do
87+
echo "::group::$t"
88+
cargo test --test "$t" -- --test-threads=4
89+
echo "::endgroup::"
90+
done
17791
178-
# readiness:
179-
# name: Readiness — Grade D/C/B
180-
# runs-on: ubuntu-latest
181-
# timeout-minutes: 10
182-
# steps:
183-
# - uses: actions/checkout@34e114876b0b11c390a56381ad16ebd13914f8d5 # v4
184-
# - uses: dtolnay/rust-toolchain@4be9e76fd7c4901c61fb841f559994984270fce7 # stable
185-
# - uses: Swatinem/rust-cache@779680da715d629ac1d338a641029a2f4372abb5 # v2
186-
# - run: cargo test --test readiness -- --nocapture
92+
- name: Summary
93+
if: always()
94+
run: |
95+
{
96+
echo "### E2E + P2P + Unit"
97+
echo
98+
echo "| Suite | Taxonomy category |"
99+
echo "|---|---|"
100+
echo "| \`smoke\` | Part I §8 Smoke |"
101+
echo "| \`property_tests\` | Part I §9 Property-based |"
102+
echo "| \`lifecycle\` | Part I §7 Lifecycle |"
103+
echo "| \`integration_tests\` | Part I §2 Point-to-Point |"
104+
echo "| \`seam_test\` | Part I §3 End-to-End |"
105+
echo
106+
echo "**Known debt:** \`cargo test --lib\` hangs (>20 min, exit 124) and is"
107+
echo "deliberately NOT run here. The integration suites above complete in ~2s."
108+
echo "This is stated rather than silently excluded."
109+
} >> "$GITHUB_STEP_SUMMARY"

0 commit comments

Comments
 (0)