Files
bot-bottle/tests/unit/test_dashboard.py
T
didericis f3a1b4d667 feat(dashboard): wire cred-proxy-block approval to real apply (PRD 0014)
Phase 3 of PRD 0014. dashboard.approve() now does the real
remediation for cred-proxy-block proposals:

- Calls apply_routes_change(slug, file_to_apply) which fetches the
  current routes.json from the running sidecar, validates the new
  JSON, docker cp's it in, and SIGHUPs the sidecar.
- Audit entry's diff is now the real before→after from the apply
  return — not the empty-string placeholder 0013 wrote.
- On apply failure (CredProxyApplyError): no response file, no
  audit entry. Proposal stays pending so the operator can fix the
  input and retry. The TUI's key handlers catch the exception and
  surface the message in the status line.
- pipelock-block + capability-block remain no-op approvals; their
  remediation lands in PRDs 0015 + 0016 and the audit diff stays
  empty until then.
- reject path unchanged: no apply, audit entry with empty diff.

Tests stub apply_routes_change at the dashboard module level so the
unit suite doesn't need a running sidecar; integration test in
Phase 5 covers the real docker exec/cp/SIGHUP plumbing.

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>
2026-05-25 04:44:33 -04:00

330 lines
13 KiB
Python

"""Unit: dashboard headless paths (PRD 0013 phase 4, PRD 0014).
The curses TUI itself isn't exercised here — these tests cover the
discovery + approve/reject + audit-write paths that the TUI's key
handlers call into.
apply_routes_change is stubbed at the dashboard module level so the
tests don't need a running cred-proxy sidecar; the real docker
exec/cp/SIGHUP plumbing is covered by the integration test.
"""
import os
import tempfile
import unittest
from datetime import datetime, timezone
from pathlib import Path
from claude_bottle import supervise
from claude_bottle.backend.docker.cred_proxy_apply import CredProxyApplyError
from claude_bottle.cli import dashboard
from claude_bottle.supervise import (
Proposal,
STATUS_APPROVED,
STATUS_MODIFIED,
STATUS_REJECTED,
TOOL_CAPABILITY_BLOCK,
TOOL_CRED_PROXY_BLOCK,
TOOL_PIPELOCK_BLOCK,
read_audit_entries,
read_response,
sha256_hex,
)
FIXED = datetime(2026, 5, 25, 12, 0, 0, tzinfo=timezone.utc)
def _proposal(slug: str = "dev", tool: str = TOOL_CRED_PROXY_BLOCK) -> Proposal:
return Proposal.new(
bottle_slug=slug, tool=tool,
proposed_file='{"routes": []}\n',
justification=f"needed for {slug}",
current_file_hash=sha256_hex("{}"),
now=FIXED,
)
class _FakeHomeMixin:
"""Patch supervise.claude_bottle_root to a temp dir for the test."""
def _setup_fake_home(self):
self._tmp = tempfile.TemporaryDirectory(prefix="dashboard-test.")
original = supervise.claude_bottle_root
def fake_root() -> Path:
return Path(self._tmp.name) / ".claude-bottle"
supervise.claude_bottle_root = fake_root # type: ignore[assignment]
self._restore_home = lambda: setattr(supervise, "claude_bottle_root", original)
def _teardown_fake_home(self):
self._restore_home()
self._tmp.cleanup()
class TestDiscoverPending(_FakeHomeMixin, unittest.TestCase):
def setUp(self):
self._setup_fake_home()
def tearDown(self):
self._teardown_fake_home()
def test_empty_when_no_queues(self):
self.assertEqual([], dashboard.discover_pending())
def test_walks_all_slug_subdirs(self):
for slug in ("dev", "api"):
qdir = supervise.queue_dir_for_slug(slug)
qdir.mkdir(parents=True)
supervise.write_proposal(qdir, _proposal(slug=slug))
pending = dashboard.discover_pending()
self.assertEqual({"dev", "api"}, {qp.proposal.bottle_slug for qp in pending})
def test_sorted_by_arrival_across_bottles(self):
early = Proposal.new(
bottle_slug="api", tool=TOOL_CRED_PROXY_BLOCK,
proposed_file="{}", justification="early",
current_file_hash="h",
now=datetime(2026, 5, 25, 10, 0, 0, tzinfo=timezone.utc),
)
late = Proposal.new(
bottle_slug="dev", tool=TOOL_CRED_PROXY_BLOCK,
proposed_file="{}", justification="late",
current_file_hash="h",
now=datetime(2026, 5, 25, 14, 0, 0, tzinfo=timezone.utc),
)
for p in (late, early):
qdir = supervise.queue_dir_for_slug(p.bottle_slug)
qdir.mkdir(parents=True, exist_ok=True)
supervise.write_proposal(qdir, p)
pending = dashboard.discover_pending()
self.assertEqual([early.id, late.id], [qp.proposal.id for qp in pending])
def test_excludes_already_responded(self):
p = _proposal()
qdir = supervise.queue_dir_for_slug("dev")
qdir.mkdir(parents=True)
supervise.write_proposal(qdir, p)
supervise.write_response(qdir, supervise.Response(
proposal_id=p.id, status=STATUS_APPROVED, notes="",
))
self.assertEqual([], dashboard.discover_pending())
class TestApproveReject(_FakeHomeMixin, unittest.TestCase):
def setUp(self):
self._setup_fake_home()
self._original_apply = dashboard.apply_routes_change
# Default stub: succeed with deterministic before/after so the
# audit log shows a non-empty diff.
dashboard.apply_routes_change = lambda slug, content: (
'{"routes": []}\n', content,
)
def tearDown(self):
dashboard.apply_routes_change = self._original_apply
self._teardown_fake_home()
def _enqueue(self, tool: str = TOOL_CRED_PROXY_BLOCK):
p = _proposal(tool=tool)
qdir = supervise.queue_dir_for_slug("dev")
qdir.mkdir(parents=True, exist_ok=True)
supervise.write_proposal(qdir, p)
return dashboard.QueuedProposal(proposal=p, queue_dir=qdir)
def test_approve_writes_response_and_audit(self):
qp = self._enqueue()
dashboard.approve(qp)
resp = read_response(qp.queue_dir, qp.proposal.id)
self.assertEqual(STATUS_APPROVED, resp.status)
self.assertIsNone(resp.final_file)
entries = read_audit_entries("cred-proxy", "dev")
self.assertEqual(1, len(entries))
self.assertEqual("approved", entries[0].operator_action)
def test_approve_with_final_file_marks_modified(self):
qp = self._enqueue()
dashboard.approve(qp, final_file='{"routes": [{"path": "/x/"}]}\n', notes="tweaked")
resp = read_response(qp.queue_dir, qp.proposal.id)
self.assertEqual(STATUS_MODIFIED, resp.status)
self.assertEqual('{"routes": [{"path": "/x/"}]}\n', resp.final_file)
self.assertEqual("tweaked", resp.notes)
entries = read_audit_entries("cred-proxy", "dev")
self.assertEqual("modified", entries[0].operator_action)
def test_reject_writes_rejection(self):
qp = self._enqueue()
dashboard.reject(qp, reason="nope")
resp = read_response(qp.queue_dir, qp.proposal.id)
self.assertEqual(STATUS_REJECTED, resp.status)
self.assertEqual("nope", resp.notes)
entries = read_audit_entries("cred-proxy", "dev")
self.assertEqual("rejected", entries[0].operator_action)
self.assertEqual("nope", entries[0].operator_notes)
def test_capability_block_skips_audit_log(self):
qp = self._enqueue(tool=TOOL_CAPABILITY_BLOCK)
dashboard.approve(qp)
# No audit log for capability-block (per PRD 0013 / 0016).
# cred-proxy and pipelock logs both empty.
self.assertEqual([], read_audit_entries("cred-proxy", "dev"))
self.assertEqual([], read_audit_entries("pipelock", "dev"))
def test_pipelock_audit_distinct_from_cred_proxy(self):
qp = self._enqueue(tool=TOOL_PIPELOCK_BLOCK)
dashboard.approve(qp)
self.assertEqual(1, len(read_audit_entries("pipelock", "dev")))
self.assertEqual(0, len(read_audit_entries("cred-proxy", "dev")))
class TestCredProxyApplyWiring(_FakeHomeMixin, unittest.TestCase):
"""PRD 0014 Phase 3: approve() on a cred-proxy-block proposal
must call apply_routes_change with the right args and surface
its failures."""
def setUp(self):
self._setup_fake_home()
self._original_apply = dashboard.apply_routes_change
def tearDown(self):
dashboard.apply_routes_change = self._original_apply
self._teardown_fake_home()
def _enqueue_cred_proxy(self, proposed: str = '{"routes": []}\n'):
p = Proposal.new(
bottle_slug="dev", tool=TOOL_CRED_PROXY_BLOCK,
proposed_file=proposed,
justification="need a route",
current_file_hash=sha256_hex(proposed),
now=FIXED,
)
qdir = supervise.queue_dir_for_slug("dev")
qdir.mkdir(parents=True, exist_ok=True)
supervise.write_proposal(qdir, p)
return dashboard.QueuedProposal(proposal=p, queue_dir=qdir)
def test_cred_proxy_block_calls_apply_with_proposed_file(self):
calls = []
dashboard.apply_routes_change = lambda slug, content: (
calls.append((slug, content)) or ("before", content)
)
qp = self._enqueue_cred_proxy(proposed='{"routes": [{"path": "/new/"}]}\n')
dashboard.approve(qp)
self.assertEqual(1, len(calls))
slug, content = calls[0]
self.assertEqual("dev", slug)
self.assertEqual('{"routes": [{"path": "/new/"}]}\n', content)
def test_modify_passes_final_file_to_apply(self):
calls = []
dashboard.apply_routes_change = lambda slug, content: (
calls.append(content) or ("before", content)
)
qp = self._enqueue_cred_proxy()
dashboard.approve(qp, final_file='{"routes": [{"path": "/edited/"}]}\n', notes="tweaked")
self.assertEqual(['{"routes": [{"path": "/edited/"}]}\n'], calls)
def test_apply_failure_blocks_response_and_audit(self):
dashboard.apply_routes_change = lambda slug, content: (_ for _ in ()).throw(
CredProxyApplyError("docker exec failed")
)
qp = self._enqueue_cred_proxy()
with self.assertRaises(CredProxyApplyError):
dashboard.approve(qp)
# No response file (proposal stays pending).
self.assertEqual(
[qp.proposal.id],
[p.id for p in supervise.list_pending_proposals(qp.queue_dir)],
)
# No audit entry.
self.assertEqual([], read_audit_entries("cred-proxy", "dev"))
def test_real_diff_lands_in_audit(self):
dashboard.apply_routes_change = lambda slug, content: (
'{"routes": []}\n', # before
'{"routes": [{"path": "/new/"}]}\n', # after
)
qp = self._enqueue_cred_proxy(proposed='{"routes": [{"path": "/new/"}]}\n')
dashboard.approve(qp)
entries = read_audit_entries("cred-proxy", "dev")
self.assertEqual(1, len(entries))
self.assertIn('+{"routes": [{"path": "/new/"}]}', entries[0].diff)
self.assertIn('-{"routes": []}', entries[0].diff)
def test_reject_does_not_call_apply(self):
called = []
dashboard.apply_routes_change = lambda slug, content: (
called.append(True) or ("", content)
)
qp = self._enqueue_cred_proxy()
dashboard.reject(qp, reason="no thanks")
self.assertEqual([], called)
# Reject still writes a response + audit entry with empty diff.
resp = read_response(qp.queue_dir, qp.proposal.id)
self.assertEqual(STATUS_REJECTED, resp.status)
entries = read_audit_entries("cred-proxy", "dev")
self.assertEqual(1, len(entries))
self.assertEqual("", entries[0].diff)
class TestEditInEditor(unittest.TestCase):
def test_runs_editor_returns_edited_content(self):
# Fake "editor" is /bin/sh -c 'cat <<EOF > $1 ... EOF'
original_editor = os.environ.get("EDITOR")
try:
# Use a fake editor that overwrites the file with a known
# marker. EDITOR is split with shlex equivalence by
# subprocess.run when invoked as a list — keep it as a
# single program path that takes the file as argv[1].
os.environ["EDITOR"] = (
"/bin/sh -c 'printf %s \"edited\" > \"$0\"'"
)
# subprocess.run with the str as the first list element
# would try to find a binary literally named "/bin/sh -c ..."
# — that won't work. Use shell mode trick: wrap in a script.
# Easier: build a tiny helper script.
with tempfile.NamedTemporaryFile(
mode="w", suffix=".sh", delete=False, prefix="fake-editor.",
) as script:
script.write('#!/bin/sh\nprintf "%s" "edited" > "$1"\n')
editor_script = script.name
os.chmod(editor_script, 0o755)
os.environ["EDITOR"] = editor_script
try:
result = dashboard.edit_in_editor("original")
self.assertEqual("edited", result)
finally:
os.unlink(editor_script)
finally:
if original_editor is None:
os.environ.pop("EDITOR", None)
else:
os.environ["EDITOR"] = original_editor
def test_returns_none_when_unchanged(self):
original_editor = os.environ.get("EDITOR")
try:
# No-op editor: touch the file (leaves it unchanged).
with tempfile.NamedTemporaryFile(
mode="w", suffix=".sh", delete=False, prefix="noop-editor.",
) as script:
script.write('#!/bin/sh\n: $1\n')
editor_script = script.name
os.chmod(editor_script, 0o755)
os.environ["EDITOR"] = editor_script
try:
result = dashboard.edit_in_editor("original")
self.assertIsNone(result)
finally:
os.unlink(editor_script)
finally:
if original_editor is None:
os.environ.pop("EDITOR", None)
else:
os.environ["EDITOR"] = original_editor
if __name__ == "__main__":
unittest.main()