Files
jpmschweitzerandClaude Opus 5 a384ec0c7c feat(config): T-1283 — the godot and visual domains
reach godot parse-sweep / cold-parse, reach visual diff / blank-check /
thumbnail. Five scripts retired, and the callers rewired — tests/run-visual
invoked three of them by path at four sites, which is a wider blast radius than
the make targets were.

The godot pair were grep pipelines encoding five hard-won lessons as comments
nobody could test. They are Python filters now, with the reasons attached, and
the engine invocation is a guarded exec. Verified on the real client: 229
scripts, clean.

Their three not-ok states stay distinct, because only one is a verdict about
the code. An engine that crashed or is missing is not a parse failure —
reporting it as one blames the tree for a broken toolchain. A sweep that
emitted no completion marker checked nothing, and zero errors from a check that
never ran reads as clean, which is the false-green the sweep exists to close.
The deliberate asymmetry between the two checks is preserved and documented:
cold-parse filters "Cannot infer the type", the sweep does not, because that
suppression is why cold-parse stayed silent about a helper that genuinely does
not parse.

All three visual scripts carried the same root bug as validate-checklist:
Path(__file__).parent.parent, correct at tooling/ and two levels too deep at
tooling/domains/visual. Fixed during the move rather than after, having learned
that it fails silently — paths resolve to nothing, the work appears to have
nothing to do, and the tool reports success. Three domains now where that would
have shipped a false pass.

Two bugs my own transformation introduced, both found by running rather than
reading. Multi-line print(..., file=sys.stderr) became console.event(...,
file=sys.stderr), and console puts unknown kwargs into the payload — a file
object would have reached json.dumps at the exact moment something was already
being reported as an error. And the replacement script wrote escaped quotes
into three files. Mechanical transformations need mechanical verification.

sys.exit removed from four sites: a service must not end the process.

Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
2026-09-02 13:59:29 +02:00

410 lines
15 KiB
Bash
Executable File

#!/usr/bin/env bash
# tests/run-visual: Visual regression test suite.
# Captures scenarios from tests/visual.json, compares against golden PNGs.
#
# Modes:
# (no args) Run all scenario golden comparisons (xvfb-wrapped)
# --screenshot NAME Ad-hoc single capture to .cache/screenshots/ (no xvfb)
# --movie NAME Flow capture to .cache/screenshots/ (no xvfb)
# --update Regenerate all goldens and stage for commit
# --filter PATTERN Accepted and ignored (compat with run-all)
#
# Exit: 0=pass (or skip), 1=fail, 2=error
# Stdout (golden mode): {"suite":"visual","total":N,"passed":N,"failed":N,"duration_ms":N}
set -euo pipefail
SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
ROOT="$(cd "$SCRIPT_DIR/.." && pwd)"
CONFIG="$ROOT/tests/visual.json"
GODOT="$(command -v godot4 2>/dev/null || command -v godot 2>/dev/null || true)"
CACHE_DIR="$ROOT/.cache/screenshots"
DIFF_DIR="$ROOT/.cache/visual-diff"
MODE="golden" # golden | screenshot | movie | update
TARGET=""
INTERVAL=""
# Server state for live scenarios
SERVER_BIN=""
SERVER_PID=""
SERVER_PORT=""
# Cleanup server on exit
trap '[[ -n "${SERVER_PID:-}" ]] && kill "$SERVER_PID" 2>/dev/null; wait "$SERVER_PID" 2>/dev/null || true' EXIT
# -- Parse args ----------------------------------------------------------------
while [[ $# -gt 0 ]]; do
case "$1" in
--screenshot) MODE="screenshot"; TARGET="${2:-atlas_GJ820Bc_Global}"; shift 2 ;;
--movie) MODE="movie"; TARGET="${2:-flow_dialogue}"; shift 2 ;;
--update) MODE="update"; shift ;;
--filter) shift 2 ;; # Accept and ignore (run-all compat)
--filter=*) shift ;;
--interval) INTERVAL="${2:-3}"; shift 2 ;;
# Override the config resolution for ONE ad-hoc capture (T-1239). The
# Atlas derives canvas extent, and therefore metres-per-gridunit, from
# the viewport — so "does this defect depend on resolution?" is a real
# diagnostic question, and answering it by hand-editing the committed
# tests/visual.json invites leaving it edited. Goldens are shot at the
# config resolution; this flag deliberately does not touch them.
--resolution) RESOLUTION_OVERRIDE="${2:-}"; shift 2 ;;
*) echo "Unknown argument: $1" >&2; exit 2 ;;
esac
done
# -- Preflight -----------------------------------------------------------------
if [[ -z "$GODOT" ]]; then
echo "Warning: Godot not found — skipping visual tests" >&2
printf '{"suite":"visual","total":0,"passed":0,"failed":0,"skipped":1,"duration_ms":0}\n'
exit 0
fi
if ! python3 -c "pass" 2>/dev/null; then
echo "Warning: Python 3 not found — skipping visual tests" >&2
printf '{"suite":"visual","total":0,"passed":0,"failed":0,"skipped":1,"duration_ms":0}\n'
exit 0
fi
if [[ ! -f "$CONFIG" ]]; then
echo "Error: $CONFIG not found" >&2
exit 2
fi
# Read config values
GOLDEN_DIR="$ROOT/$(python3 -c "import json; c=json.load(open('$CONFIG')); print(c.get('golden_dir','client/tests/golden/visual'))")"
TOLERANCE="$(python3 -c "import json; c=json.load(open('$CONFIG')); print(c.get('tolerance', 5))")"
RESOLUTION="$(python3 -c "import json; c=json.load(open('$CONFIG')); r=c.get('resolution',[960,540]); print(f'{r[0]}x{r[1]}')")"
if [[ -n "${RESOLUTION_OVERRIDE:-}" ]]; then
RESOLUTION="$RESOLUTION_OVERRIDE"
echo "Note: resolution overridden to $RESOLUTION (goldens are shot at the config resolution)" >&2
fi
# Read scenario names from config
SCENARIOS=($(python3 -c "
import json
c = json.load(open('$CONFIG'))
for name in c.get('scenarios', {}):
print(name)
"))
# -- Helpers -------------------------------------------------------------------
# Offscreen wrapper for EVERY capture path.
#
# Captures must never steal the desktop: this box is also the developer's own
# session. Until 2026-08-06 only the golden path even tried, via xvfb-run —
# which is not installed here, so it hit the "using visible window" fallback,
# and --screenshot/--movie never wrapped at all. Every capture grabbed focus.
#
# gamescope is preferred, for two independent reasons:
# 1. It renders on the REAL GPU. The goldens are pinned to this box's
# Mesa/AMD output (T-1121 — they do not port across rendering stacks), and
# xvfb-run would fall back to llvmpipe software rendering, shifting every
# pixel. Offscreen must not mean a different renderer.
# 2. It is the only option that lets us SET the output size. cage is also
# installed and also GPU-backed, but it is a kiosk compositor: it forces
# its client to the headless output's default 1280x720 and silently
# overrides --resolution (measured — a 960x540 request produced a 1280x720
# PNG). A wrapper that quietly changes resolution is worse than none,
# because RESOLUTION is load-bearing here: D-255's extent inversion makes
# the shorter viewport axis span exactly one cell of the rung, so window
# size decides how much world a rung shows. cage is deliberately not used.
CAPTURE_PREFIX=()
if command -v gamescope >/dev/null 2>&1; then
CAPTURE_PREFIX=(gamescope --backend headless -W "${RESOLUTION%x*}" -H "${RESOLUTION#*x}" --)
elif command -v xvfb-run >/dev/null 2>&1; then
echo "Note: gamescope not found — falling back to xvfb-run (software GL; goldens may drift)" >&2
CAPTURE_PREFIX=(xvfb-run -a --server-args="-screen 0 ${RESOLUTION}x24")
else
echo "Warning: no offscreen compositor (gamescope/xvfb-run) — capturing in a VISIBLE window" >&2
fi
godot_capture() {
local mode_flag="$1" # --scenario or --flow
local name="$2"
local output="$3"
local extra_args=("${@:4}")
# Isolate `user://` per run (T-1239). Godot resolves user:// under
# XDG_DATA_HOME, which is how the Atlas disk cache (D-255, T-1183) persisted
# across captures — including across the server changes that made its
# contents wrong. A capture then rendered a canvas generated by a build that
# no longer existed: Ferrath Global replayed a pre-T-1237 canvas from weeks
# earlier and showed 375 hop-fragments where the live server produces 73
# whole rivers, and every golden shot in that window silently inherited it.
# A visual test must exercise the code in the tree, so the cache it warms
# must not outlive the run. The dir is recreated fresh each capture; the
# user's real cache at ~/.local/share/godot is never touched.
local user_data="$ROOT/.cache/visual-user-data"
rm -rf "$user_data"
mkdir -p "$user_data"
XDG_DATA_HOME="$user_data" \
"${CAPTURE_PREFIX[@]}" "$GODOT" --rendering-driver opengl3 --fixed-fps 60 \
--resolution "$RESOLUTION" \
--path "$ROOT/client" -s res://tests/visual_capture.gd -- \
$mode_flag "$name" --output "$output" "${extra_args[@]}" 2>&1
}
# Retained as the golden-path name; offscreen handling now lives in
# godot_capture, so both paths get it.
xvfb_capture() {
godot_capture "$@"
}
# -- Server lifecycle (live scenarios) -----------------------------------------
# Check if a scenario has "live": true in config
is_live_scenario() {
python3 -c "
import json, sys
c = json.load(open('${CONFIG}'))
s = c.get('scenarios', {}).get('${1}', {})
sys.exit(0 if s.get('live') else 1)
"
}
# Build server binary (once, cached)
ensure_server_built() {
if [[ -n "$SERVER_BIN" ]]; then return 0; fi
echo " Building server for live visual tests..."
(cd "$ROOT/server" && cargo build --bin settled-reach-server 2>&1) || {
echo "Error: server build failed" >&2
return 1
}
SERVER_BIN="$ROOT/server/target/debug/settled-reach-server"
}
# Start server with --test-mode --port 0, parse LISTENING:{port}
start_server() {
# Keep the server's own output (T-1239). This was `mktemp` + `2>/dev/null`
# + `rm` as soon as LISTENING was parsed, which meant two things: every
# tracing::warn!/error! the simulation emitted was discarded, and everything
# after startup went to an unlinked file. A server quietly degrading mid-
# capture looked identical to a healthy one — the capture only ever showed
# what the CLIENT thought it received. Now stderr is merged in and the log
# lives at a stable path that survives the run for inspection.
local stdout_log="$ROOT/.cache/visual-server.log"
mkdir -p "$ROOT/.cache"
"$SERVER_BIN" --test-mode --port 0 >"$stdout_log" 2>&1 &
SERVER_PID=$!
local attempts=0
while [[ $attempts -lt 150 ]]; do
if ! kill -0 "$SERVER_PID" 2>/dev/null; then
echo " Error: server exited unexpectedly — see $stdout_log" >&2
SERVER_PID=""
return 1
fi
if grep -q "^LISTENING:" "$stdout_log" 2>/dev/null; then
SERVER_PORT=$(sed -n 's/^LISTENING://p' "$stdout_log")
echo " Server started: pid=$SERVER_PID port=$SERVER_PORT log=$stdout_log"
return 0
fi
sleep 0.1
attempts=$((attempts + 1))
done
echo " Error: no LISTENING signal after 15s — see $stdout_log" >&2
kill "$SERVER_PID" 2>/dev/null || true
SERVER_PID=""
return 1
}
# Stop server (called after each live capture; server may have exited on disconnect)
stop_server() {
if [[ -n "$SERVER_PID" ]]; then
kill "$SERVER_PID" 2>/dev/null || true
wait "$SERVER_PID" 2>/dev/null || true
SERVER_PID=""
SERVER_PORT=""
fi
}
# -- Screenshot mode -----------------------------------------------------------
if [[ "$MODE" == "screenshot" ]]; then
mkdir -p "$CACHE_DIR"
echo "Capturing scenario: $TARGET"
if is_live_scenario "$TARGET"; then
ensure_server_built || exit 2
start_server || exit 2
export SR_LIVE=1 SR_PORT="$SERVER_PORT"
fi
godot_capture --scenario "$TARGET" "$CACHE_DIR"
stop_server
unset SR_LIVE SR_PORT 2>/dev/null || true
PNG="$CACHE_DIR/$TARGET.png"
if [[ -f "$PNG" ]]; then
echo "Screenshot: $PNG ($(stat -c%s "$PNG" 2>/dev/null || stat -f%z "$PNG") bytes)"
# Advisory here rather than fatal — an ad-hoc capture of a rung that
# genuinely renders nothing is a legitimate thing to want to look at
# (that is how the empty deep rungs were found). But say so out loud,
# because file size alone reads as success.
reach --no-input visual blank-check "$PNG" --quiet || true
else
echo "Error: capture failed — $PNG not found" >&2
exit 1
fi
exit 0
fi
# -- Movie mode ----------------------------------------------------------------
if [[ "$MODE" == "movie" ]]; then
mkdir -p "$CACHE_DIR"
echo "Capturing flow: $TARGET"
EXTRA=()
[[ -n "$INTERVAL" ]] && EXTRA+=(--interval "$INTERVAL")
godot_capture --flow "$TARGET" "$CACHE_DIR" "${EXTRA[@]}"
FLOW_DIR="$CACHE_DIR/$TARGET"
if [[ -d "$FLOW_DIR" ]]; then
FRAME_COUNT=$(find "$FLOW_DIR" -name "*.png" | wc -l)
echo "Flow: $FRAME_COUNT frames in $FLOW_DIR"
# Generate contact sheet if visual-thumbnail is available
if command -v reach >/dev/null 2>&1; then
reach --no-input visual thumbnail "$FLOW_DIR" --config "$CONFIG"
SHEET="$FLOW_DIR/${TARGET}_sheet.png"
[[ -f "$SHEET" ]] && echo "Contact sheet: $SHEET"
fi
else
echo "Error: flow capture failed — $FLOW_DIR not found" >&2
exit 1
fi
exit 0
fi
# -- Golden mode (default) / Update mode --------------------------------------
START_MS=$(date +%s%3N)
TOTAL=0
PASSED=0
FAILED=0
mkdir -p "$CACHE_DIR" "$DIFF_DIR"
if [[ "$MODE" == "update" ]]; then
mkdir -p "$GOLDEN_DIR"
fi
for scenario in "${SCENARIOS[@]}"; do
TOTAL=$((TOTAL + 1))
echo "--- $scenario ---"
# Start server for live scenarios
IS_LIVE=false
if is_live_scenario "$scenario"; then
IS_LIVE=true
ensure_server_built || { FAILED=$((FAILED + 1)); continue; }
start_server || { FAILED=$((FAILED + 1)); continue; }
export SR_LIVE=1 SR_PORT="$SERVER_PORT"
fi
# Capture
set +e
CAPTURE_OUT=$(xvfb_capture --scenario "$scenario" "$CACHE_DIR" 2>&1)
CAPTURE_RC=$?
set -e
# Stop server after capture (server exits on client disconnect anyway)
if [[ "$IS_LIVE" == "true" ]]; then
unset SR_LIVE SR_PORT 2>/dev/null || true
stop_server
fi
CAPTURED="$CACHE_DIR/$scenario.png"
if [[ $CAPTURE_RC -ne 0 ]] || [[ ! -f "$CAPTURED" ]]; then
echo " FAIL: capture failed (exit $CAPTURE_RC)" >&2
echo "$CAPTURE_OUT" >&2
FAILED=$((FAILED + 1))
continue
fi
# Verify non-empty
if [[ ! -s "$CAPTURED" ]]; then
echo " FAIL: captured PNG is empty" >&2
FAILED=$((FAILED + 1))
continue
fi
# Verify the renderer actually drew something.
#
# "Non-empty file" was the only content check until 2026-08-06, and a blank
# screen is a perfectly valid ~19 KB PNG. Worse, once a blank capture was
# recorded as a golden, every later blank capture matched it at 0.0% and
# the scenario PASSED — atlas_GJ338Bd_Block and atlas_GJ445c-m1_Chunk were
# green against blank goldens. A test that cannot fail is worse than no
# test, because it is counted as coverage.
#
# Checked in BOTH modes, and the update mode matters most: refusing to
# RECORD a blank golden is what stops the trap being re-armed.
set +e
BLANK_OUT=$(reach --no-input visual blank-check "$CAPTURED" 2>&1)
BLANK_RC=$?
set -e
if [[ $BLANK_RC -ne 0 ]]; then
echo " FAIL: $BLANK_OUT" >&2
FAILED=$((FAILED + 1))
continue
fi
if [[ "$MODE" == "update" ]]; then
cp "$CAPTURED" "$GOLDEN_DIR/$scenario.png"
echo " Updated golden: $GOLDEN_DIR/$scenario.png"
PASSED=$((PASSED + 1))
else
GOLDEN="$GOLDEN_DIR/$scenario.png"
if [[ ! -f "$GOLDEN" ]]; then
echo " FAIL: golden not found — run 'make visual-update' first" >&2
FAILED=$((FAILED + 1))
continue
fi
# Compare
set +e
DIFF_OUT=$(reach --no-input visual diff "$GOLDEN" "$CAPTURED" \
--tolerance "$TOLERANCE" \
--diff-output "$DIFF_DIR/$scenario-diff.png" \
--config "$CONFIG" 2>&1)
DIFF_RC=$?
set -e
if [[ $DIFF_RC -eq 0 ]]; then
echo " $DIFF_OUT"
PASSED=$((PASSED + 1))
elif [[ $DIFF_RC -eq 1 ]]; then
echo " $DIFF_OUT"
echo " Diff image: $DIFF_DIR/$scenario-diff.png"
FAILED=$((FAILED + 1))
else
echo " ERROR: visual-diff failed (exit $DIFF_RC)" >&2
echo " $DIFF_OUT" >&2
FAILED=$((FAILED + 1))
fi
fi
done
END_MS=$(date +%s%3N)
DURATION_MS=$((END_MS - START_MS))
if [[ "$MODE" == "update" ]]; then
# Stage golden files
cd "$ROOT"
git add "$GOLDEN_DIR/" 2>/dev/null || true
echo ""
echo "=== Visual goldens updated ($PASSED of $TOTAL) ==="
echo "Review with: git diff --cached -- $GOLDEN_DIR/"
fi
printf '{"suite":"visual","total":%d,"passed":%d,"failed":%d,"duration_ms":%d}\n' \
"$TOTAL" "$PASSED" "$FAILED" "$DURATION_MS"
[[ $FAILED -gt 0 ]] && exit 1
exit 0