wehub-resource-sync
f3d80b4628
Auto Release / auto-release (push) Failing after 1s
CI / lint (push) Failing after 0s
CI / screenshot-quality (push) Failing after 3s
CI / pr-policy (push) Has been skipped
CI / test (3.11) (push) Failing after 0s
CI / test (3.12) (push) Failing after 0s
CI / test (3.13) (push) Failing after 3s
CI / coverage (push) Failing after 1s
Legibility / legibility (push) Failing after 3s
117 lines
3.1 KiB
Bash
Executable File
117 lines
3.1 KiB
Bash
Executable File
#!/usr/bin/env bash
|
|
set -euo pipefail
|
|
|
|
# Run a real Claude E2E flow through claude-tap forward proxy mode without tmux.
|
|
# Usage:
|
|
# scripts/run_real_e2e.sh
|
|
# Env:
|
|
# PROMPT_ONE / PROMPT_TWO override default prompts.
|
|
# CLAUDE_ARGS extra claude CLI args (e.g. "--model sonnet").
|
|
|
|
REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
|
|
cd "$REPO_ROOT"
|
|
|
|
if ! command -v uv >/dev/null 2>&1; then
|
|
echo "error: uv not found." >&2
|
|
exit 1
|
|
fi
|
|
|
|
if ! command -v claude >/dev/null 2>&1; then
|
|
echo "error: claude CLI not found." >&2
|
|
exit 1
|
|
fi
|
|
|
|
TS="$(date +%s)"
|
|
TRACE_DIR="/tmp/ctap-real-e2e-$TS"
|
|
RUN_LOG="/tmp/claude-tap-recordings/real-e2e-$TS.log"
|
|
PROMPT_ONE="${PROMPT_ONE:-Reply with exactly: REAL_E2E_TURN_ONE_OK}"
|
|
PROMPT_TWO="${PROMPT_TWO:-What was the exact text I asked you to reply with in the previous turn?}"
|
|
CLAUDE_ARGS="${CLAUDE_ARGS:-}"
|
|
|
|
mkdir -p "$TRACE_DIR" /tmp/claude-tap-recordings
|
|
|
|
run_turn() {
|
|
local prompt="$1"
|
|
shift
|
|
local extra_args=("$@")
|
|
|
|
local claude_args=()
|
|
if [[ -n "$CLAUDE_ARGS" ]]; then
|
|
# shellcheck disable=SC2206
|
|
claude_args=($CLAUDE_ARGS)
|
|
fi
|
|
local cmd=(
|
|
uv run python -m claude_tap
|
|
--tap-output-dir "$TRACE_DIR"
|
|
--tap-proxy-mode forward
|
|
--
|
|
)
|
|
if [[ ${#claude_args[@]} -gt 0 ]]; then
|
|
cmd+=("${claude_args[@]}")
|
|
fi
|
|
cmd+=(-p "$prompt")
|
|
if [[ ${#extra_args[@]} -gt 0 ]]; then
|
|
cmd+=("${extra_args[@]}")
|
|
fi
|
|
|
|
echo "==> ${cmd[*]}" | tee -a "$RUN_LOG"
|
|
"${cmd[@]}" 2>&1 | tee -a "$RUN_LOG"
|
|
}
|
|
|
|
run_turn "$PROMPT_ONE"
|
|
run_turn "$PROMPT_TWO" -c
|
|
|
|
uv run python - "$TRACE_DIR" "$PROMPT_ONE" "$PROMPT_TWO" <<'PY'
|
|
import json
|
|
import sys
|
|
from pathlib import Path
|
|
|
|
from claude_tap import _generate_html_viewer
|
|
|
|
trace_dir = Path(sys.argv[1])
|
|
p1 = sys.argv[2]
|
|
p2 = sys.argv[3]
|
|
|
|
jsonl_files = sorted(trace_dir.glob("trace_*.jsonl"))
|
|
if not jsonl_files:
|
|
raise SystemExit(f"assert failed: no trace_*.jsonl found in {trace_dir}")
|
|
|
|
all_text_parts: list[str] = []
|
|
message_count = 0
|
|
|
|
for jf in jsonl_files:
|
|
text = jf.read_text(encoding="utf-8")
|
|
if text:
|
|
all_text_parts.append(text)
|
|
for line in text.splitlines():
|
|
if not line.strip():
|
|
continue
|
|
try:
|
|
rec = json.loads(line)
|
|
except json.JSONDecodeError:
|
|
continue
|
|
path = rec.get("request", {}).get("path", "")
|
|
if isinstance(path, str) and path.startswith("/v1/messages"):
|
|
message_count += 1
|
|
|
|
html = jf.with_suffix(".html")
|
|
if not html.exists():
|
|
_generate_html_viewer(jf, html)
|
|
|
|
all_text = "\n".join(all_text_parts)
|
|
if p1 not in all_text:
|
|
raise SystemExit("assert failed: prompt one missing in trace jsonl files")
|
|
if p2 not in all_text:
|
|
raise SystemExit("assert failed: prompt two missing in trace jsonl files")
|
|
if message_count < 2:
|
|
raise SystemExit(f"assert failed: expected >=2 /v1/messages requests, got {message_count}")
|
|
|
|
print(f"assert ok: traces={len(jsonl_files)} message_requests={message_count}")
|
|
for jf in jsonl_files:
|
|
print(f"JSONL={jf}")
|
|
print(f"HTML={jf.with_suffix('.html')}")
|
|
PY
|
|
|
|
echo "TRACE_DIR=$TRACE_DIR"
|
|
echo "RUN_LOG=$RUN_LOG"
|