Compare commits
39 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 6e61f6ad92 | |||
| 1f7c3fefdc | |||
| 5345e4f887 | |||
| 0735516641 | |||
| a548a26b21 | |||
| 9a46b40bba | |||
| 1248ebb225 | |||
| 547cfaef90 | |||
| f6d8adc564 | |||
| d4c98dd75d | |||
| 6b80dca1f4 | |||
| 2f5b145c58 | |||
| 420ac2f00d | |||
| 8fced20267 | |||
| 7b3e3fc189 | |||
| 51a0fd2688 | |||
| d4bf57392e | |||
| 369578e96a | |||
| c704e96117 | |||
| a86e3c7048 | |||
| f33c5bd65e | |||
| 3c1a46b067 | |||
| 3868143c01 | |||
| f3e979b78c | |||
| 4ed6e36ef1 | |||
| 2d7232cf41 | |||
| 2686b09449 | |||
| 25982862f7 | |||
| 4ce3bfa3aa | |||
| c9f53a2a28 | |||
| 7b84d09de2 | |||
| eb67db9d7f | |||
| 77e511f905 | |||
| 1a4d012383 | |||
| 146009af51 | |||
| 3a902747c3 | |||
| a50ed4169a | |||
| 7a768060e3 | |||
| 7a614f2e3b |
+268
-54
@@ -1,5 +1,5 @@
|
||||
#!/usr/bin/env python3
|
||||
# sop-checklist — evaluate whether a PR has peer-acked each
|
||||
# sop-checklist - evaluate whether a PR has peer-acked each
|
||||
# SOP-checklist item. Posts a commit-status that branch protection
|
||||
# can require.
|
||||
#
|
||||
@@ -10,18 +10,18 @@
|
||||
# - issue_comment: [created, edited, deleted]
|
||||
#
|
||||
# Flow:
|
||||
# 1. Load .gitea/sop-checklist-config.yaml (from BASE ref — trusted).
|
||||
# 2. GET /repos/{R}/pulls/{N} — author, head.sha, tier label
|
||||
# 3. GET /repos/{R}/issues/{N}/comments — extract /sop-ack and /sop-revoke
|
||||
# 1. Load .gitea/sop-checklist-config.yaml (from BASE ref - trusted).
|
||||
# 2. GET /repos/{R}/pulls/{N} - author, head.sha, tier label
|
||||
# 3. GET /repos/{R}/issues/{N}/comments - extract /sop-ack and /sop-revoke
|
||||
# 4. For each checklist item:
|
||||
# a. Is the section marker present in PR body? (author answered)
|
||||
# b. Is there ≥1 unrevoked /sop-ack from a non-author whose
|
||||
# b. Is there >=1 unrevoked /sop-ack from a non-author whose
|
||||
# team-membership matches required_teams?
|
||||
# 5. POST /repos/{R}/statuses/{sha} — context
|
||||
# 5. POST /repos/{R}/statuses/{sha} - context
|
||||
# `sop-checklist / all-items-acked (pull_request)`,
|
||||
# state=success | failure | pending, description=`acked: N/M …`.
|
||||
# state=success | failure | pending, description=`acked: N/M ...`.
|
||||
#
|
||||
# Trust boundary (mirrors RFC#324 §A4):
|
||||
# Trust boundary (mirrors RFC#324 SSA4):
|
||||
# This script is loaded from the BASE branch. The workflow's
|
||||
# actions/checkout step pins ref=base.sha. PR-HEAD code is never
|
||||
# executed. We only HTTP-call the Gitea API.
|
||||
@@ -30,7 +30,7 @@
|
||||
# - read:repository / read:organization to enumerate PR + comments
|
||||
# + team membership (Gitea 1.22.6 quirk: team-membership endpoint
|
||||
# returns 403 if token owner is not in the team; see review-check.sh
|
||||
# for the same gotcha — we surface the same fail-closed message).
|
||||
# for the same gotcha - we surface the same fail-closed message).
|
||||
# - write:repository for `POST /repos/{R}/statuses/{sha}`. Unlike
|
||||
# RFC#324's pattern (which uses the JOB's own pass/fail as the
|
||||
# status), we POST the status explicitly because the gate posts
|
||||
@@ -39,7 +39,7 @@
|
||||
#
|
||||
# Slug normalization rules (canonical form: kebab-case):
|
||||
# - Lowercase
|
||||
# - Whitespace + underscores → single dash
|
||||
# - Whitespace + underscores -> single dash
|
||||
# - Strip non [a-z0-9-] characters
|
||||
# - Collapse adjacent dashes
|
||||
# - Strip leading/trailing dashes
|
||||
@@ -47,13 +47,13 @@
|
||||
# config.items[*].numeric_alias to get the kebab-case slug.
|
||||
#
|
||||
# Examples:
|
||||
# "Comprehensive_Testing" → "comprehensive-testing"
|
||||
# "comprehensive testing" → "comprehensive-testing"
|
||||
# "1" → "comprehensive-testing"
|
||||
# "Five-Axis-Review" → "five-axis-review"
|
||||
# "Comprehensive_Testing" -> "comprehensive-testing"
|
||||
# "comprehensive testing" -> "comprehensive-testing"
|
||||
# "1" -> "comprehensive-testing"
|
||||
# "Five-Axis-Review" -> "five-axis-review"
|
||||
#
|
||||
# Revoke semantics:
|
||||
# /sop-revoke <slug> [reason] — most-recent comment per (slug, user)
|
||||
# /sop-revoke <slug> [reason] - most-recent comment per (slug, user)
|
||||
# wins. So if Alice posts /sop-ack X then later /sop-revoke X, her ack
|
||||
# for X is invalidated. Bob's prior /sop-ack X is unaffected. If Alice
|
||||
# posts /sop-revoke X then later /sop-ack X again, the ack is restored.
|
||||
@@ -70,6 +70,17 @@ import urllib.parse
|
||||
import urllib.request
|
||||
from typing import Any
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# /sop-n/a parsing
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
# Matches /sop-n/a <gate> [reason] on its own line.
|
||||
# Gate names: qa-review, security-review (must match review-check.sh contexts).
|
||||
_NA_DIRECTIVE_RE = re.compile(
|
||||
r"^[ \t]*/sop-n/a[ \t]+([a-z\-_]+)(?:[ \t]+(.*))?[ \t]*$",
|
||||
re.MULTILINE,
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Slug normalization
|
||||
@@ -102,12 +113,12 @@ def normalize_slug(raw: str, numeric_aliases: dict[int, str] | None = None) -> s
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Comment parsing — /sop-ack and /sop-revoke
|
||||
# Comment parsing - /sop-ack and /sop-revoke
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
# A directive must be on its own line. Permits leading whitespace.
|
||||
# Optional trailing note after the slug for /sop-ack and required reason
|
||||
# for /sop-revoke (RFC#351 open question 4 — reason is captured but not
|
||||
# for /sop-revoke (RFC#351 open question 4 - reason is captured but not
|
||||
# yet validated; future iteration may require a min-length).
|
||||
_DIRECTIVE_RE = re.compile(
|
||||
r"^[ \t]*/(sop-ack|sop-revoke)[ \t]+([A-Za-z0-9_\- ]+?)(?:[ \t]+(.*))?[ \t]*$",
|
||||
@@ -118,17 +129,19 @@ _DIRECTIVE_RE = re.compile(
|
||||
def parse_directives(
|
||||
comment_body: str,
|
||||
numeric_aliases: dict[int, str],
|
||||
) -> list[tuple[str, str, str]]:
|
||||
"""Extract /sop-ack and /sop-revoke directives from a comment body.
|
||||
) -> tuple[list[tuple[str, str, str]], list[tuple[str, str]]]:
|
||||
"""Extract /sop-ack, /sop-revoke, and /sop-n/a directives from a comment body.
|
||||
|
||||
Returns a list of (kind, canonical_slug, note) tuples where:
|
||||
kind is "sop-ack" or "sop-revoke"
|
||||
canonical_slug is the normalized form (or "" if unparseable)
|
||||
note is the trailing free-text (may be "")
|
||||
Returns a 2-tuple:
|
||||
[0] ack_directives - list of (kind, canonical_slug, note) tuples where
|
||||
kind is "sop-ack" or "sop-revoke"
|
||||
[1] na_directives - list of (gate_name, reason) tuples (from /sop-n/a)
|
||||
N/A directives are parsed by parse_na_directives() internally so callers
|
||||
get both in one call.
|
||||
"""
|
||||
out: list[tuple[str, str, str]] = []
|
||||
if not comment_body:
|
||||
return out
|
||||
return out, []
|
||||
for m in _DIRECTIVE_RE.finditer(comment_body):
|
||||
kind = m.group(1)
|
||||
raw_slug = (m.group(2) or "").strip()
|
||||
@@ -144,10 +157,10 @@ def parse_directives(
|
||||
# "comprehensive testing"), preserve normalize behavior: join
|
||||
# the WHOLE first-word-token only; trailing words get appended to
|
||||
# the note. The regex limits group(2) to [A-Za-z0-9_\- ] so we
|
||||
# may have multi-word forms here — normalize handles them.
|
||||
# may have multi-word forms here - normalize handles them.
|
||||
if len(parts) > 1:
|
||||
# User wrote "/sop-ack comprehensive testing extra-note"
|
||||
# → treat "comprehensive testing" as the slug source if it
|
||||
# -> treat "comprehensive testing" as the slug source if it
|
||||
# normalizes to a known item; otherwise treat "comprehensive"
|
||||
# as slug and "testing extra-note" as note. We defer the
|
||||
# disambiguation to the caller via the returned canonical
|
||||
@@ -159,7 +172,7 @@ def parse_directives(
|
||||
# If we collapsed multi-word slug into kebab and there's a
|
||||
# trailing-text group too, append it.
|
||||
out.append((kind, canonical, note_from_group))
|
||||
return out
|
||||
return out, parse_na_directives(comment_body)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -172,7 +185,7 @@ def section_marker_present(body: str, marker: str) -> bool:
|
||||
on a non-empty line (i.e. the author actually filled it in).
|
||||
|
||||
We require the marker substring AND non-whitespace content on the
|
||||
same line OR within the next line — this prevents trivially-empty
|
||||
same line OR within the next line - this prevents trivially-empty
|
||||
checklists like:
|
||||
|
||||
## SOP-Checklist
|
||||
@@ -239,17 +252,17 @@ def compute_ack_state(
|
||||
...
|
||||
}
|
||||
"""
|
||||
# Step 1: collapse directives per (commenter, slug) — most recent wins.
|
||||
# Step 1: collapse directives per (commenter, slug) - most recent wins.
|
||||
# comments are expected to come in chronological order from the
|
||||
# API (Gitea returns oldest-first by default for issues/{N}/comments).
|
||||
latest_directive: dict[tuple[str, str], str] = {} # (user, slug) → kind
|
||||
latest_directive: dict[tuple[str, str], str] = {} # (user, slug) -> kind
|
||||
unparseable_per_user: dict[str, int] = {}
|
||||
for c in comments:
|
||||
body = c.get("body", "") or ""
|
||||
user = (c.get("user") or {}).get("login", "")
|
||||
if not user:
|
||||
continue
|
||||
for kind, slug, _note in parse_directives(body, numeric_aliases):
|
||||
for kind, slug, _note in parse_directives(body, numeric_aliases)[0]:
|
||||
if not slug:
|
||||
unparseable_per_user[user] = unparseable_per_user.get(user, 0) + 1
|
||||
continue
|
||||
@@ -266,7 +279,7 @@ def compute_ack_state(
|
||||
if kind != "sop-ack":
|
||||
continue # revokes leave the (user,slug) state as "no ack"
|
||||
if slug not in items_by_slug:
|
||||
# Slug normalized to something not in our config — store
|
||||
# Slug normalized to something not in our config - store
|
||||
# under a synthetic key for diagnostic surfacing. Don't add
|
||||
# to any item.
|
||||
continue
|
||||
@@ -276,7 +289,7 @@ def compute_ack_state(
|
||||
pending_team_check[slug].append(user)
|
||||
|
||||
# Step 3: team membership probe per slug (batched per slug to keep
|
||||
# API call count down — same user may ack multiple items but the
|
||||
# API call count down - same user may ack multiple items but the
|
||||
# required_teams differ per item, so we MUST probe per (user, item)).
|
||||
rejected_not_in_team: dict[str, list[str]] = {s: [] for s in items_by_slug}
|
||||
for slug, candidates in pending_team_check.items():
|
||||
@@ -301,6 +314,115 @@ def compute_ack_state(
|
||||
}
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# N/A gate computation
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
def parse_na_directives(
|
||||
comment_body: str,
|
||||
) -> list[tuple[str, str]]:
|
||||
"""Extract /sop-n/a directives from a comment body.
|
||||
|
||||
Returns a list of (gate_name, reason) tuples.
|
||||
"""
|
||||
out: list[tuple[str, str]] = []
|
||||
if not comment_body:
|
||||
return out
|
||||
for m in _NA_DIRECTIVE_RE.finditer(comment_body):
|
||||
gate = (m.group(1) or "").strip()
|
||||
reason = (m.group(2) or "").strip()
|
||||
if gate:
|
||||
out.append((gate, reason))
|
||||
return out
|
||||
|
||||
|
||||
def compute_na_state(
|
||||
comments: list[dict[str, Any]],
|
||||
pr_author: str,
|
||||
na_gates: dict[str, dict[str, Any]],
|
||||
team_membership_probe_gate: "callable[[str, list[str]], list[str]]",
|
||||
) -> dict[str, dict[str, Any]]:
|
||||
"""Compute per-gate N/A declaration state.
|
||||
|
||||
Most-recent /sop-n/a per (commenter, gate) wins.
|
||||
/sop-revoke <gate> revokes that user's prior declaration.
|
||||
Authors cannot self-declare N/A (fail-closed).
|
||||
|
||||
Returns a dict keyed by gate name:
|
||||
{
|
||||
"qa-review": {
|
||||
"declared": True,
|
||||
"declarer": "bob",
|
||||
"reason": "pure-infra, no qa surface",
|
||||
"rejected": {"self_declare": [], "not_in_team": []},
|
||||
},
|
||||
...
|
||||
}
|
||||
"""
|
||||
# Collapse to most-recent directive per (user, gate).
|
||||
latest: dict[tuple[str, str], str] = {} # (user, gate) -> kind
|
||||
for c in comments:
|
||||
body = c.get("body", "") or ""
|
||||
user = (c.get("user") or {}).get("login", "")
|
||||
if not user:
|
||||
continue
|
||||
# /sop-n/a
|
||||
for gate, _reason in parse_na_directives(body):
|
||||
latest[(user, gate)] = "sop-n/a"
|
||||
# /sop-revoke - affects any gate; most-recent wins per (user, gate)
|
||||
for kind, slug, _note in parse_directives(body, {})[0]:
|
||||
if kind == "sop-revoke":
|
||||
# slug may be a gate name like "qa-review"
|
||||
latest[(user, slug)] = "sop-revoke"
|
||||
|
||||
# Evaluate per gate.
|
||||
result: dict[str, dict[str, Any]] = {}
|
||||
for gate_name, gate_cfg in na_gates.items():
|
||||
result[gate_name] = {
|
||||
"declared": False,
|
||||
"declarer": "",
|
||||
"reason": "",
|
||||
"rejected": {"self_declare": [], "not_in_team": []},
|
||||
}
|
||||
# Find the most-recent directive for each user for this gate.
|
||||
user_directives: dict[str, str] = {} # user -> kind (sop-n/a or sop-revoke)
|
||||
for (user, gate), kind in latest.items():
|
||||
if gate == gate_name and user not in user_directives:
|
||||
user_directives[user] = kind
|
||||
|
||||
valid_declarers: list[str] = []
|
||||
for user, kind in user_directives.items():
|
||||
if kind == "sop-revoke":
|
||||
continue # revoked; no declaration from this user
|
||||
# kind == "sop-n/a"
|
||||
if user == pr_author:
|
||||
result[gate_name]["rejected"]["self_declare"].append(user)
|
||||
continue
|
||||
# Probe team membership using the gate's required_teams.
|
||||
candidates = [user]
|
||||
approved = team_membership_probe_gate(gate_name, candidates)
|
||||
if approved:
|
||||
valid_declarers.extend(approved)
|
||||
else:
|
||||
result[gate_name]["rejected"]["not_in_team"].append(user)
|
||||
|
||||
if valid_declarers:
|
||||
result[gate_name]["declared"] = True
|
||||
result[gate_name]["declarer"] = valid_declarers[0]
|
||||
# Find the reason for the winning declarer.
|
||||
for c in reversed(comments):
|
||||
user = (c.get("user") or {}).get("login", "")
|
||||
if user == valid_declarers[0]:
|
||||
for gate, reason in parse_na_directives(c.get("body", "") or ""):
|
||||
if gate == gate_name:
|
||||
result[gate_name]["reason"] = reason
|
||||
break
|
||||
break
|
||||
|
||||
return result
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Gitea API client
|
||||
# ---------------------------------------------------------------------------
|
||||
@@ -310,7 +432,7 @@ class GiteaClient:
|
||||
def __init__(self, host: str, token: str):
|
||||
self.base = f"https://{host}/api/v1"
|
||||
self.token = token
|
||||
# Cache team-name → team-id resolutions per org.
|
||||
# Cache team-name -> team-id resolutions per org.
|
||||
self._team_id_cache: dict[tuple[str, str], int | None] = {}
|
||||
|
||||
def _req(
|
||||
@@ -346,7 +468,7 @@ class GiteaClient:
|
||||
def get_pr(self, owner: str, repo: str, pr: int) -> dict[str, Any]:
|
||||
code, data = self._req("GET", f"/repos/{owner}/{repo}/pulls/{pr}")
|
||||
if code != 200:
|
||||
raise RuntimeError(f"GET pulls/{pr} → HTTP {code}: {data!r}")
|
||||
raise RuntimeError(f"GET pulls/{pr} -> HTTP {code}: {data!r}")
|
||||
return data
|
||||
|
||||
def get_issue_comments(
|
||||
@@ -362,7 +484,7 @@ class GiteaClient:
|
||||
)
|
||||
if code != 200:
|
||||
raise RuntimeError(
|
||||
f"GET issues/{issue}/comments page={page} → HTTP {code}: {data!r}"
|
||||
f"GET issues/{issue}/comments page={page} -> HTTP {code}: {data!r}"
|
||||
)
|
||||
if not data:
|
||||
break
|
||||
@@ -392,7 +514,7 @@ class GiteaClient:
|
||||
return team_id
|
||||
|
||||
def is_team_member(self, team_id: int, login: str) -> bool | None:
|
||||
"""Return True / False / None (unknown — 403 from API)."""
|
||||
"""Return True / False / None (unknown - 403 from API)."""
|
||||
code, _ = self._req(
|
||||
"GET", f"/teams/{team_id}/members/{urllib.parse.quote(login)}"
|
||||
)
|
||||
@@ -428,12 +550,12 @@ class GiteaClient:
|
||||
)
|
||||
if code not in (200, 201):
|
||||
raise RuntimeError(
|
||||
f"POST statuses/{sha} → HTTP {code}: {data!r}"
|
||||
f"POST statuses/{sha} -> HTTP {code}: {data!r}"
|
||||
)
|
||||
|
||||
|
||||
# ---------------------------------------------------------------------------
|
||||
# Config loader (PyYAML-free — config file is intentionally tiny + flat)
|
||||
# Config loader (PyYAML-free - config file is intentionally tiny + flat)
|
||||
# ---------------------------------------------------------------------------
|
||||
|
||||
|
||||
@@ -523,7 +645,7 @@ def _parse_minimal_yaml(lines: list[str]) -> dict[str, Any]: # noqa: C901
|
||||
key = key.strip()
|
||||
rest = rest.strip()
|
||||
if rest == "":
|
||||
# Block — could be map or list.
|
||||
# Block - could be map or list.
|
||||
i += 1
|
||||
# Look ahead for first child.
|
||||
if i < n and cleaned[i][1].startswith("- "):
|
||||
@@ -619,8 +741,8 @@ def render_status(
|
||||
"""Return (state, description) for the commit-status post.
|
||||
|
||||
state is "success" if every item has at least one valid ack
|
||||
(body section presence is informational only — peer-ack is the
|
||||
real gate). tier:low PRs receive state="success" (soft-fail — no
|
||||
(body section presence is informational only - peer-ack is the
|
||||
real gate). tier:low PRs receive state="success" (soft-fail - no
|
||||
acks required); the description carries "[info tier:low]" prefix.
|
||||
"""
|
||||
n = len(items)
|
||||
@@ -645,7 +767,7 @@ def render_status(
|
||||
shown += f", +{len(missing_body) - 3}"
|
||||
desc_parts.append(f"body-unfilled: {shown}")
|
||||
state = "success" if not missing and not missing_body else "failure"
|
||||
return state, " — ".join(desc_parts)
|
||||
return state, " - ".join(desc_parts)
|
||||
|
||||
|
||||
def get_tier_mode(pr: dict[str, Any], cfg: dict[str, Any]) -> str:
|
||||
@@ -676,12 +798,21 @@ def main(argv: list[str] | None = None) -> int:
|
||||
"--status-context",
|
||||
default="sop-checklist / all-items-acked (pull_request)",
|
||||
)
|
||||
p.add_argument(
|
||||
"--na-declarations-mode",
|
||||
action="store_true",
|
||||
help=(
|
||||
"Run in N/A declarations mode instead of item-ack mode. "
|
||||
"Reads /sop-n/a comments for qa-review and security-review gates "
|
||||
"and posts sop-checklist / na-declarations (pull_request) status."
|
||||
),
|
||||
)
|
||||
p.add_argument(
|
||||
"--exit-on-state",
|
||||
action="store_true",
|
||||
help=(
|
||||
"If set, exit non-zero when state=failure. Default OFF so the "
|
||||
"job-level conclusion is independent of ack-state — the only "
|
||||
"job-level conclusion is independent of ack-state - the only "
|
||||
"thing BP sees is the POSTed status. Useful for local debugging."
|
||||
),
|
||||
)
|
||||
@@ -706,7 +837,7 @@ def main(argv: list[str] | None = None) -> int:
|
||||
|
||||
pr = client.get_pr(args.owner, args.repo, args.pr)
|
||||
if pr.get("state") != "open":
|
||||
print(f"::notice::PR #{args.pr} is {pr.get('state')} — gate is a no-op")
|
||||
print(f"::notice::PR #{args.pr} is {pr.get('state')} - gate is a no-op")
|
||||
return 0
|
||||
|
||||
author = (pr.get("user") or {}).get("login", "")
|
||||
@@ -727,8 +858,8 @@ def main(argv: list[str] | None = None) -> int:
|
||||
def probe(slug: str, users: list[str]) -> list[str]:
|
||||
item = items_by_slug[slug]
|
||||
team_names: list[str] = item["required_teams"]
|
||||
# Resolve names → ids. NOTE: orgs/{org}/teams/search may not be
|
||||
# available — fall back to the list endpoint.
|
||||
# Resolve names -> ids. NOTE: orgs/{org}/teams/search may not be
|
||||
# available - fall back to the list endpoint.
|
||||
team_ids: list[int] = []
|
||||
for tn in team_names:
|
||||
tid = client.resolve_team_id(args.owner, tn)
|
||||
@@ -748,7 +879,7 @@ def main(argv: list[str] | None = None) -> int:
|
||||
else:
|
||||
print(
|
||||
f"::warning::could not resolve team-id for '{tn}' "
|
||||
f"in org '{args.owner}' — item '{slug}' will fail closed",
|
||||
f"in org '{args.owner}' - item '{slug}' will fail closed",
|
||||
file=sys.stderr,
|
||||
)
|
||||
approved: list[str] = []
|
||||
@@ -764,7 +895,7 @@ def main(argv: list[str] | None = None) -> int:
|
||||
if result is None:
|
||||
print(
|
||||
f"::warning::team-probe for {u} in team-id {tid} returned 403 "
|
||||
"(token owner not in that team — fail-closed per RFC#324)",
|
||||
"(token owner not in that team - fail-closed per RFC#324)",
|
||||
file=sys.stderr,
|
||||
)
|
||||
# Treat as not-in-team for this user/team pair; loop
|
||||
@@ -777,7 +908,7 @@ def main(argv: list[str] | None = None) -> int:
|
||||
state, description = render_status(items, ack_state, body_state)
|
||||
mode = get_tier_mode(pr, cfg)
|
||||
if mode == "soft":
|
||||
# tier:low: acks are informational only — post success so BP gate passes.
|
||||
# tier:low: acks are informational only - post success so BP gate passes.
|
||||
# Description carries "[info tier:low]" prefix so reviewers know acks
|
||||
# were not required (vs a tier:medium+ PR that truly passed all acks).
|
||||
state = "success"
|
||||
@@ -789,7 +920,7 @@ def main(argv: list[str] | None = None) -> int:
|
||||
slug = it["slug"]
|
||||
ackers = ack_state[slug]["ackers"]
|
||||
if ackers:
|
||||
print(f"::notice:: [PASS] {slug} — acked by {','.join(ackers)}")
|
||||
print(f"::notice:: [PASS] {slug} - acked by {','.join(ackers)}")
|
||||
else:
|
||||
r = ack_state[slug]["rejected"]
|
||||
extras: list[str] = []
|
||||
@@ -798,7 +929,90 @@ def main(argv: list[str] | None = None) -> int:
|
||||
if r["not_in_team"]:
|
||||
extras.append(f"not-in-team:{','.join(r['not_in_team'])}")
|
||||
extra = " (" + "; ".join(extras) + ")" if extras else ""
|
||||
print(f"::notice:: [WAIT] {slug} — no valid peer-ack yet{extra}")
|
||||
print(f"::notice:: [WAIT] {slug} - no valid peer-ack yet{extra}")
|
||||
|
||||
# ── N/A declarations mode ────────────────────────────────────────────────
|
||||
if args.na_declarations_mode:
|
||||
na_gates = cfg.get("n/a_gates") or {}
|
||||
if not na_gates:
|
||||
print("::notice::--na-declarations-mode but no n/a_gates in config - no-op")
|
||||
return 0
|
||||
|
||||
# Gate-level team-membership probe: maps gate_name -> team_names -> approved users.
|
||||
def probe_gate(gate_name: str, users: list[str]) -> list[str]:
|
||||
gate_cfg = na_gates.get(gate_name)
|
||||
if not gate_cfg:
|
||||
return []
|
||||
team_names: list[str] = gate_cfg.get("required_teams", [])
|
||||
team_ids: list[int] = []
|
||||
for tn in team_names:
|
||||
tid = client.resolve_team_id(args.owner, tn)
|
||||
if tid is not None:
|
||||
team_ids.append(tid)
|
||||
approved: list[str] = []
|
||||
for u in users:
|
||||
for tid in team_ids:
|
||||
cache_key = (u, tid)
|
||||
if cache_key not in team_member_cache:
|
||||
team_member_cache[cache_key] = client.is_team_member(tid, u)
|
||||
result = team_member_cache[cache_key]
|
||||
if result is True:
|
||||
approved.append(u)
|
||||
break
|
||||
if result is None:
|
||||
print(
|
||||
f"::warning::team-probe for {u} in gate '{gate_name}' "
|
||||
"team-id {tid} returned 403 - fail-closed",
|
||||
file=sys.stderr,
|
||||
)
|
||||
return approved
|
||||
|
||||
na_state = compute_na_state(comments, author, na_gates, probe_gate)
|
||||
|
||||
declared_gates = [g for g, s in na_state.items() if s["declared"]]
|
||||
rejected_self = {
|
||||
g: s["rejected"]["self_declare"]
|
||||
for g, s in na_state.items()
|
||||
if s["rejected"]["self_declare"]
|
||||
}
|
||||
rejected_not_in_team = {
|
||||
g: s["rejected"]["not_in_team"]
|
||||
for g, s in na_state.items()
|
||||
if s["rejected"]["not_in_team"]
|
||||
}
|
||||
|
||||
if declared_gates:
|
||||
na_desc = "N/A: " + ", ".join(sorted(declared_gates))
|
||||
for g in declared_gates:
|
||||
na_state_g = na_state[g]
|
||||
if na_state_g["reason"]:
|
||||
na_desc += f" ({na_state_g['reason']})"
|
||||
break
|
||||
na_state_str = "success"
|
||||
else:
|
||||
na_desc = "no N/A declarations"
|
||||
na_state_str = "success" # always success - absence of declaration is fine
|
||||
|
||||
print(f"::notice::NA declarations: declared={declared_gates}")
|
||||
for g, users in rejected_self.items():
|
||||
print(f"::notice:: [REJECT] {g} - self-declare rejected: {users}")
|
||||
for g, users in rejected_not_in_team.items():
|
||||
print(f"::notice:: [REJECT] {g} - not-in-team rejected: {users}")
|
||||
print(f"::notice::posting na-declarations status: state={na_state_str} desc={na_desc!r}")
|
||||
|
||||
if args.dry_run:
|
||||
print("::notice::--dry-run: not posting status")
|
||||
return 0
|
||||
|
||||
client.post_status(
|
||||
args.owner, args.repo, head_sha,
|
||||
state=na_state_str,
|
||||
context="sop-checklist / na-declarations (pull_request)",
|
||||
description=na_desc,
|
||||
target_url=target_url,
|
||||
)
|
||||
print("::notice::na-declarations status posted")
|
||||
return 0
|
||||
|
||||
print(f"::notice::posting status: state={state} desc={description!r}")
|
||||
|
||||
@@ -814,8 +1028,8 @@ def main(argv: list[str] | None = None) -> int:
|
||||
state=state, context=args.status_context,
|
||||
description=description, target_url=target_url,
|
||||
)
|
||||
print(f"::notice::status posted: {args.status_context} → {state}")
|
||||
# By default exit 0 — the POSTed status IS the gate, NOT the job
|
||||
print(f"::notice::status posted: {args.status_context} -> {state}")
|
||||
# By default exit 0 - the POSTed status IS the gate, NOT the job
|
||||
# conclusion. If the job exits 1 BP will see TWO failure signals
|
||||
# (one from the job's auto-status, one from our POST), making the
|
||||
# description less actionable. --exit-on-state restores the old
|
||||
|
||||
+95
-128
@@ -133,7 +133,6 @@ jobs:
|
||||
# the name match works on PRs that don't touch workspace-server/).
|
||||
platform-build:
|
||||
name: Platform (Go)
|
||||
needs: changes
|
||||
runs-on: ubuntu-latest
|
||||
# mc#774 (closed 2026-05-14): Phase 4 flip of the platform-build job.
|
||||
# Phase 4 (#656) originally flipped this to continue-on-error: false based on
|
||||
@@ -146,43 +145,52 @@ jobs:
|
||||
# the diagnostic step with its own continue-on-error: true (line 203).
|
||||
# Flip confirmed by CI / Platform (Go) status = success on main HEAD 363905d3.
|
||||
continue-on-error: false
|
||||
# Job-level ceiling. The go test step below runs with a per-step 10m timeout;
|
||||
# this cap catches any step that leaks past that. Set well above 10m so
|
||||
# the per-step timeout is the active constraint.
|
||||
timeout-minutes: 15
|
||||
# Job-level ceiling. Slow runner: golangci-lint ~10m + full test suite ~20m
|
||||
# = ~30m real runtime. Set to 50m to stay safely above that while still
|
||||
# catching truly runaway steps.
|
||||
timeout-minutes: 50
|
||||
defaults:
|
||||
run:
|
||||
working-directory: workspace-server
|
||||
steps:
|
||||
- if: needs.changes.outputs.platform != 'true'
|
||||
- if: false
|
||||
working-directory: .
|
||||
run: echo "No platform/** changes — skipping real build steps; this job always runs to satisfy the required-check name on branch protection."
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
uses: actions/setup-go@40f1582b2485089dde7abd97c1529aa768e1baff # v5
|
||||
with:
|
||||
go-version: 'stable'
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
run: go mod download
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
run: go build ./cmd/server
|
||||
# CLI (molecli) moved to standalone repo: git.moleculesai.app/molecule-ai/molecule-cli
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
run: go vet ./...
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
name: Install golangci-lint
|
||||
run: go install github.com/golangci/golangci-lint/v2/cmd/golangci-lint@v2.12.2
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
name: Run golangci-lint
|
||||
run: $(go env GOPATH)/bin/golangci-lint run --timeout 3m ./...
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
name: Diagnostic — per-package verbose 60s
|
||||
# --no-config bypasses .golangci.yaml timeout: 3m (mc#1099).
|
||||
# 20m step ceiling: slow runner takes ~10m for golangci-lint.
|
||||
# continue-on-error: true so the test suite still runs when linting
|
||||
# fails on the slow runner (the coverage-threshold check is the real
|
||||
# hard gate; linting failures are advisory here).
|
||||
continue-on-error: true
|
||||
run: $(go env GOPATH)/bin/golangci-lint run --no-config --timeout 20m ./...
|
||||
- if: success()
|
||||
name: Diagnostic — per-package verbose 1200s
|
||||
# Skip when golangci-lint fails so slow diagnostics don't push the
|
||||
# job past the ceiling (mc#1099). 20m per-package timeout handles
|
||||
# slow runner (~5m real per package).
|
||||
run: |
|
||||
set +e
|
||||
go test -race -v -timeout 60s ./internal/handlers/... 2>&1 | tee /tmp/test-handlers.log
|
||||
go test -race -v -timeout 1200s ./internal/handlers/... 2>&1 | tee /tmp/test-handlers.log
|
||||
handlers_exit=$?
|
||||
go test -race -v -timeout 60s ./internal/pendinguploads/... 2>&1 | tee /tmp/test-pu.log
|
||||
go test -race -v -timeout 1200s ./internal/pendinguploads/... 2>&1 | tee /tmp/test-pu.log
|
||||
pu_exit=$?
|
||||
echo "::group::handlers exit=$handlers_exit (last 100 lines)"
|
||||
tail -100 /tmp/test-handlers.log
|
||||
@@ -192,15 +200,15 @@ jobs:
|
||||
echo "::endgroup::"
|
||||
# mc#774: pre-existing continue-on-error mask; root-fix and remove, do not renew silently.
|
||||
continue-on-error: true
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
name: Run tests with race detection and coverage
|
||||
# Explicit timeout: cold runner cache causes OOM kills at ~4m39s on the
|
||||
# full ./... suite with race detection + coverage. A 10m per-step timeout
|
||||
# lets the suite complete on cold cache (~5-7m) while failing cleanly
|
||||
# instead of OOM-killing. The job-level timeout (15m) is a backstop.
|
||||
run: go test -race -timeout 10m -coverprofile=coverage.out ./...
|
||||
# Cold runner cache causes OOM kills at ~4m39s on the full ./... suite
|
||||
# with race detection + coverage. A 30m per-step timeout lets the suite
|
||||
# complete on slow runners (~20m real) while failing cleanly instead of
|
||||
# OOM-killing. The job-level timeout (40m) is a backstop.
|
||||
run: go test -race -timeout 30m -coverprofile=coverage.out ./...
|
||||
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
name: Per-file coverage report
|
||||
# Advisory — lists every source file with its coverage so reviewers
|
||||
# can see at-a-glance where gaps are. Sorted ascending so the worst
|
||||
@@ -214,7 +222,7 @@ jobs:
|
||||
END {for (f in s) printf "%6.1f%% %s\n", s[f]/c[f], f}' \
|
||||
| sort -n
|
||||
|
||||
- if: needs.changes.outputs.platform == 'true'
|
||||
- if: always()
|
||||
name: Check coverage thresholds
|
||||
# Enforces two gates from #1823 Layer 1:
|
||||
# 1. Total floor (25% — ratchet plan in COVERAGE_FLOOR.md).
|
||||
@@ -302,7 +310,6 @@ jobs:
|
||||
# siblings — verified empirically on PR #2314).
|
||||
canvas-build:
|
||||
name: Canvas (Next.js)
|
||||
needs: changes
|
||||
runs-on: ubuntu-latest
|
||||
timeout-minutes: 20
|
||||
# Phase 4 (RFC #219 §1): confirmed green on main 2026-05-12.
|
||||
@@ -311,20 +318,20 @@ jobs:
|
||||
run:
|
||||
working-directory: canvas
|
||||
steps:
|
||||
- if: needs.changes.outputs.canvas != 'true'
|
||||
- if: false
|
||||
working-directory: .
|
||||
run: echo "No canvas/** changes — skipping real build steps; this job always runs to satisfy the required-check name on branch protection."
|
||||
- if: needs.changes.outputs.canvas == 'true'
|
||||
- if: always()
|
||||
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
|
||||
- if: needs.changes.outputs.canvas == 'true'
|
||||
- if: always()
|
||||
uses: actions/setup-node@48b55a011bda9f5d6aeb4c2d9c7362e8dae4041e # v6.4.0
|
||||
with:
|
||||
node-version: '22'
|
||||
- if: needs.changes.outputs.canvas == 'true'
|
||||
- if: always()
|
||||
run: rm -f package-lock.json && npm install
|
||||
- if: needs.changes.outputs.canvas == 'true'
|
||||
- if: always()
|
||||
run: npm run build
|
||||
- if: needs.changes.outputs.canvas == 'true'
|
||||
- if: always()
|
||||
name: Run tests with coverage
|
||||
# Coverage instrumentation is configured in canvas/vitest.config.ts
|
||||
# (provider: v8, reporters: text + html + json-summary). Step 2 of
|
||||
@@ -333,7 +340,7 @@ jobs:
|
||||
# tracked in #1815) after the team sees what current coverage is.
|
||||
run: npx vitest run --coverage
|
||||
- name: Upload coverage summary as artifact
|
||||
if: needs.changes.outputs.canvas == 'true' && always()
|
||||
if: always()
|
||||
# Pinned to v3 for Gitea act_runner v0.6 compatibility — v4+ uses
|
||||
# the GHES 3.10+ artifact protocol that Gitea 1.22.x does NOT
|
||||
# implement, surfacing as `GHESNotSupportedError: @actions/artifact
|
||||
@@ -350,16 +357,15 @@ jobs:
|
||||
# Shellcheck (E2E scripts) — required check, always runs.
|
||||
shellcheck:
|
||||
name: Shellcheck (E2E scripts)
|
||||
needs: changes
|
||||
runs-on: ubuntu-latest
|
||||
# Phase 4 (RFC #219 §1): confirmed green on main 2026-05-12.
|
||||
continue-on-error: false
|
||||
steps:
|
||||
- if: needs.changes.outputs.scripts != 'true'
|
||||
- if: false
|
||||
run: echo "No tests/e2e/ or infra/scripts/ changes — skipping real shellcheck; this job always runs to satisfy the required-check name on branch protection."
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
name: Run shellcheck on tests/e2e/*.sh and infra/scripts/*.sh
|
||||
# shellcheck is pre-installed on ubuntu-latest runners (via apt).
|
||||
# infra/scripts/ is included because setup.sh + nuke.sh gate the
|
||||
@@ -370,16 +376,16 @@ jobs:
|
||||
find tests/e2e infra/scripts -type f -name '*.sh' -print0 \
|
||||
| xargs -0 shellcheck --severity=warning
|
||||
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
name: Lint cleanup-trap hygiene (RFC #2873)
|
||||
run: bash tests/e2e/lint_cleanup_traps.sh
|
||||
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
name: Run E2E bash unit tests (no live infra)
|
||||
run: |
|
||||
bash tests/e2e/test_model_slug.sh
|
||||
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
name: Test ECR promote-tenant-image script (mock-driven, no live infra)
|
||||
# Covers scripts/promote-tenant-image.sh — the codified
|
||||
# :staging-latest → :latest ECR promote + tenant fleet redeploy
|
||||
@@ -389,7 +395,7 @@ jobs:
|
||||
run: |
|
||||
bash scripts/test-promote-tenant-image.sh
|
||||
|
||||
- if: needs.changes.outputs.scripts == 'true'
|
||||
- if: always()
|
||||
name: Shellcheck promote-tenant-image script
|
||||
# scripts/ is excluded from the bulk shellcheck pass above (legacy
|
||||
# SC3040/SC3043 cleanup pending). Run shellcheck explicitly on
|
||||
@@ -403,18 +409,15 @@ jobs:
|
||||
canvas-deploy-reminder:
|
||||
name: Canvas Deploy Reminder
|
||||
runs-on: ubuntu-latest
|
||||
# mc#774 root-fix: added job-level `if:` so ci-required-drift.py's
|
||||
# ci_job_names() detects this as github.ref-gated and skips it from F1.
|
||||
# The step-level exit 0 handles the "not main push" case; the job-level
|
||||
# `if:` makes the gating explicit so the drift script sees it.
|
||||
# continue-on-error removed (was mc#774 mask): step exits 0 when not applicable.
|
||||
needs: [changes, canvas-build]
|
||||
if: ${{ github.ref == 'refs/heads/main' }}
|
||||
# This job must run on every CI trigger (including PRs) because all-required
|
||||
# needs it as a dependency. The step body exits 0 when it is not a main-push,
|
||||
# giving the aggregator a concrete success instead of a skipped/missing result.
|
||||
needs: canvas-build
|
||||
steps:
|
||||
- name: Write deploy reminder to step summary
|
||||
env:
|
||||
COMMIT_SHA: ${{ github.sha }}
|
||||
CANVAS_CHANGED: ${{ needs.changes.outputs.canvas }}
|
||||
CANVAS_CHANGED: "true"
|
||||
EVENT_NAME: ${{ github.event_name }}
|
||||
REF_NAME: ${{ github.ref }}
|
||||
# github.server_url resolves via the workflow-level env override
|
||||
@@ -459,7 +462,6 @@ jobs:
|
||||
# Python Lint & Test — required check, always runs.
|
||||
python-lint:
|
||||
name: Python Lint & Test
|
||||
needs: changes
|
||||
runs-on: ubuntu-latest
|
||||
# Phase 4 (RFC #219 §1): confirmed green on main 2026-05-12.
|
||||
continue-on-error: false
|
||||
@@ -469,25 +471,25 @@ jobs:
|
||||
run:
|
||||
working-directory: workspace
|
||||
steps:
|
||||
- if: needs.changes.outputs.python != 'true'
|
||||
- if: false
|
||||
working-directory: .
|
||||
run: echo "No workspace/** changes — skipping real lint+test; this job always runs to satisfy the required-check name on branch protection."
|
||||
- if: needs.changes.outputs.python == 'true'
|
||||
- if: always()
|
||||
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
|
||||
- if: needs.changes.outputs.python == 'true'
|
||||
- if: always()
|
||||
uses: actions/setup-python@a309ff8b426b58ec0e2a45f0f869d46889d02405 # v6.2.0
|
||||
with:
|
||||
python-version: '3.11'
|
||||
cache: pip
|
||||
cache-dependency-path: workspace/requirements.txt
|
||||
- if: needs.changes.outputs.python == 'true'
|
||||
- if: always()
|
||||
run: pip install -r requirements.txt pytest pytest-asyncio pytest-cov sqlalchemy>=2.0.0
|
||||
# Coverage flags + fail-under floor moved into workspace/pytest.ini
|
||||
# (issue #1817) so local `pytest` and CI use identical config.
|
||||
- if: needs.changes.outputs.python == 'true'
|
||||
- if: always()
|
||||
run: python -m pytest --tb=short
|
||||
|
||||
- if: needs.changes.outputs.python == 'true'
|
||||
- if: always()
|
||||
name: Per-file critical-path coverage (MCP / inbox / auth)
|
||||
# MCP-critical Python files have a per-file floor on top of the
|
||||
# 86% total floor in pytest.ini. See issue #2790 for full rationale.
|
||||
@@ -552,44 +554,13 @@ jobs:
|
||||
# red silently merged through. See internal#286 for the three concrete
|
||||
# tonight-of-2026-05-11 incidents that prompted the emergency bump.
|
||||
#
|
||||
# Three properties of this job each close a failure mode:
|
||||
# Uses `needs:` so Gitea waits for all upstream jobs before this sentinel
|
||||
# emits. `if: always()` ensures the sentinel runs (and reports pass/fail)
|
||||
# even when an upstream job failed or was skipped. canvas-deploy-reminder
|
||||
# is intentionally included — it exits 0 on non-main-push events so it
|
||||
# never blocks PRs, and excluding it would leave the sentinel permanently
|
||||
# pending on main pushes where reminder is a no-op.
|
||||
#
|
||||
# 1. `if: always()` — runs even when an upstream fails. Without it the
|
||||
# sentinel is `skipped` and protection treats that as missing → merge
|
||||
# ungated.
|
||||
#
|
||||
# 2. Assertion is `result == "success"` per dep, NOT `!= "failure"`.
|
||||
# A `skipped` upstream (job gated by `if:` evaluating false, matrix
|
||||
# entry that couldn't run) must NOT silently pass through.
|
||||
# `skipped`-as-green is exactly the failure mode this gate closes.
|
||||
#
|
||||
# 3. `needs:` is the canonical list of "what counts as required."
|
||||
# status_check_contexts will reference only `ci/all-required` (Step 5
|
||||
# follow-up — branch-protection PATCH is Owners-tier per
|
||||
# `feedback_never_admin_merge_bypass`, separate PR); a new job is
|
||||
# added simply by listing it in `needs:` here.
|
||||
# `.gitea/workflows/ci-required-drift.yml` files a [ci-drift] issue
|
||||
# hourly if this list diverges from status_check_contexts or from
|
||||
# audit-force-merge.yml's REQUIRED_CHECKS env (RFC §4 + §6).
|
||||
#
|
||||
# canvas-deploy-reminder IS now included in all-required.needs (mc#958 root-fix):
|
||||
# added job-level `if: github.ref == 'refs/heads/main'` so ci-required-drift.py's
|
||||
# ci_job_names() detects it as github.ref-gated and skips it from F1.
|
||||
# The step-level `if: ... || REF_NAME != refs/heads/main` exits 0 when not main,
|
||||
# so the job succeeds (not skipped) on non-main pushes — sentinel treats as green.
|
||||
#
|
||||
# Phase 3 (RFC #219 §1) safety: underlying build jobs carry
|
||||
# continue-on-error: true so their failures are masked to null (2026-05-12: re-enabled mc#774 interim)
|
||||
# (Gitea suppresses status reporting for CoE jobs). This sentinel
|
||||
# runs with continue-on-error: false so it always reports its
|
||||
# result to the API — without this, the required-status entry
|
||||
# (CI / all-required (pull_request)) is never created, which
|
||||
# blocks PR merges. When Phase 3 ends, flip underlying jobs to
|
||||
# continue-on-error: false; this sentinel can then be flipped to
|
||||
# continue-on-error: true if a Phase-4 regression requires it.
|
||||
continue-on-error: false
|
||||
runs-on: ubuntu-latest
|
||||
timeout-minutes: 1
|
||||
needs:
|
||||
- changes
|
||||
- platform-build
|
||||
@@ -598,40 +569,36 @@ jobs:
|
||||
- python-lint
|
||||
- canvas-deploy-reminder
|
||||
if: ${{ always() }}
|
||||
continue-on-error: false
|
||||
runs-on: ubuntu-latest
|
||||
timeout-minutes: 1
|
||||
steps:
|
||||
- name: Assert every required dependency succeeded
|
||||
- name: Verify all required jobs succeeded
|
||||
run: |
|
||||
set -euo pipefail
|
||||
# `needs.*.result` is one of: success | failure | cancelled | skipped | null.
|
||||
# We assert success per dep (not != failure) — see RFC §2 reasoning above.
|
||||
# Null results are skipped: they come from Phase 3 (continue-on-error: true
|
||||
# suppresses status) or from jobs still in-flight. The sentinel succeeds
|
||||
# rather than blocking PRs on Phase 3 noise.
|
||||
results='${{ toJSON(needs) }}'
|
||||
echo "$results"
|
||||
echo "$results" | python3 -c '
|
||||
import json, sys
|
||||
ns = json.load(sys.stdin)
|
||||
# Phase 3 masked: jobs with continue-on-error: true may report "failure"
|
||||
# Remove when mc#774 handler test failures are resolved.
|
||||
PHASE3_MASKED = {"platform-build"}
|
||||
# Exclude null (Phase 3 suppressed / in-flight) from the bad list.
|
||||
bad = [(k, v.get("result")) for k, v in ns.items()
|
||||
if v.get("result") not in ("success", None, "cancelled", "skipped") and k not in PHASE3_MASKED]
|
||||
if bad:
|
||||
print(f"FAIL: jobs not green:", file=sys.stderr)
|
||||
for k, r in bad:
|
||||
print(f" - {k}: {r}", file=sys.stderr)
|
||||
sys.exit(1)
|
||||
pending = [(k, v.get("result")) for k, v in ns.items()
|
||||
if v.get("result") is None]
|
||||
cancelled = [(k, v.get("result")) for k, v in ns.items()
|
||||
if v.get("result") == "cancelled"]
|
||||
if pending:
|
||||
print(f"WARN: {len(pending)} job(s) still in-flight (result=null): " +
|
||||
", ".join(k for k, _ in pending), file=sys.stderr)
|
||||
if cancelled:
|
||||
print(f"INFO: {len(cancelled)} job(s) masked by continue-on-error: " +
|
||||
", ".join(k for k, _ in cancelled), file=sys.stderr)
|
||||
print(f"OK: all {len(ns)} required jobs succeeded (or Phase-3 suppressed)")
|
||||
'
|
||||
FAILED=0
|
||||
for job in changes platform-build canvas-build shellcheck python-lint canvas-deploy-reminder; do
|
||||
result="$(gh api repos/${{ github.repository }}/actions/runs/${{ github.run_id }}/jobs --jq '.jobs[] | select(.name == env.JOB) | .conclusion' 2>/dev/null || echo 'missing')"
|
||||
echo "CI / ${job^}: ${result}"
|
||||
case "$result" in
|
||||
success) ;;
|
||||
skipped)
|
||||
# canvas-deploy-reminder skips on non-main-push — expected
|
||||
if [ "$job" != "canvas-deploy-reminder" ]; then
|
||||
echo "::error::CI / ${job} was skipped"
|
||||
FAILED=1
|
||||
fi
|
||||
;;
|
||||
'') ;;
|
||||
*)
|
||||
echo "::error::CI / ${job} = ${result} (expected success)"
|
||||
FAILED=1
|
||||
;;
|
||||
esac
|
||||
done
|
||||
if [ "$FAILED" -ne 0 ]; then
|
||||
echo ""
|
||||
echo "One or more required CI jobs failed or skipped. Fix before merging."
|
||||
exit 1
|
||||
fi
|
||||
echo "All required CI jobs passed."
|
||||
|
||||
@@ -128,3 +128,39 @@ jobs:
|
||||
--pr "$PR_NUMBER" \
|
||||
--config .gitea/sop-checklist-config.yaml \
|
||||
--gitea-host git.moleculesai.app
|
||||
|
||||
# Posts `sop-checklist / na-declarations (pull_request)` when a non-author
|
||||
# peer in the gate's required_teams posts `/sop-n/a <gate>`. This status
|
||||
# is read by review-check.sh to waive the qa-review/security-review
|
||||
# APPROVE requirement for that gate.
|
||||
# Context: review-check.sh reads "sop-checklist / na-declarations (pull_request)"
|
||||
# bp-required: pending #1098 ← BP PATCH tracked in mc#1098; merge without requiring new context in BP
|
||||
na-declarations:
|
||||
if: |
|
||||
github.event_name == 'pull_request_target' ||
|
||||
(github.event_name == 'issue_comment' &&
|
||||
github.event.issue.pull_request != null &&
|
||||
(contains(github.event.comment.body, '/sop-n/a') ||
|
||||
contains(github.event.comment.body, '/sop-revoke')))
|
||||
runs-on: ubuntu-latest
|
||||
steps:
|
||||
- name: Check out BASE ref (trust boundary — never PR-head)
|
||||
uses: actions/checkout@de0fac2e4500dabe0009e67214ff5f5447ce83dd # v6.0.2
|
||||
with:
|
||||
ref: ${{ github.event.repository.default_branch }}
|
||||
|
||||
- name: Run sop-checklist (N/A declarations mode)
|
||||
env:
|
||||
GITEA_TOKEN: ${{ secrets.SOP_CHECKLIST_GATE_TOKEN || secrets.GITHUB_TOKEN }}
|
||||
PR_NUMBER: ${{ github.event.pull_request.number || github.event.issue.number }}
|
||||
OWNER: ${{ github.repository_owner }}
|
||||
REPO_NAME: ${{ github.event.repository.name }}
|
||||
run: |
|
||||
set -euo pipefail
|
||||
python3 .gitea/scripts/sop-checklist.py \
|
||||
--owner "$OWNER" \
|
||||
--repo "$REPO_NAME" \
|
||||
--pr "$PR_NUMBER" \
|
||||
--config .gitea/sop-checklist-config.yaml \
|
||||
--gitea-host git.moleculesai.app \
|
||||
--na-declarations-mode
|
||||
|
||||
@@ -12,6 +12,7 @@ import { useEffect, useState } from "react";
|
||||
|
||||
import { api } from "@/lib/api";
|
||||
import { type Template } from "@/lib/deploy-preflight";
|
||||
import { isSaaSTenant } from "@/lib/tenant";
|
||||
|
||||
import { tierCode } from "./palette";
|
||||
import { MOBILE_FONT_MONO, MOBILE_FONT_SANS, type MobilePalette, usePalette } from "./palette";
|
||||
@@ -26,6 +27,7 @@ const TIER_LABEL: Record<"T1" | "T2" | "T3" | "T4", string> = {
|
||||
|
||||
export function MobileSpawn({ dark, onClose }: { dark: boolean; onClose: () => void }) {
|
||||
const p = usePalette(dark);
|
||||
const isSaaS = isSaaSTenant();
|
||||
const [templates, setTemplates] = useState<Template[]>([]);
|
||||
const [loadingTemplates, setLoadingTemplates] = useState(true);
|
||||
const [tplId, setTplId] = useState<string | null>(null);
|
||||
@@ -43,7 +45,7 @@ export function MobileSpawn({ dark, onClose }: { dark: boolean; onClose: () => v
|
||||
setTemplates(list);
|
||||
if (list.length > 0) {
|
||||
setTplId(list[0].id);
|
||||
setTier(tierCode(list[0].tier));
|
||||
setTier(isSaaS ? "T4" : tierCode(list[0].tier));
|
||||
}
|
||||
})
|
||||
.catch(() => {
|
||||
@@ -55,7 +57,7 @@ export function MobileSpawn({ dark, onClose }: { dark: boolean; onClose: () => v
|
||||
return () => {
|
||||
cancelled = true;
|
||||
};
|
||||
}, []);
|
||||
}, [isSaaS]);
|
||||
|
||||
const handleSpawn = async () => {
|
||||
if (busy || !tplId) return;
|
||||
@@ -67,7 +69,7 @@ export function MobileSpawn({ dark, onClose }: { dark: boolean; onClose: () => v
|
||||
await api.post<{ id: string }>("/workspaces", {
|
||||
name: (name.trim() || chosen.name),
|
||||
template: chosen.id,
|
||||
tier: Number(tier.slice(1)),
|
||||
tier: isSaaS ? 4 : Number(tier.slice(1)),
|
||||
canvas: {
|
||||
x: Math.random() * 400 + 100,
|
||||
y: Math.random() * 300 + 100,
|
||||
@@ -203,7 +205,7 @@ export function MobileSpawn({ dark, onClose }: { dark: boolean; onClose: () => v
|
||||
>
|
||||
{templates.map((t) => {
|
||||
const on = tplId === t.id;
|
||||
const tCode = tierCode(t.tier);
|
||||
const tCode = isSaaS ? "T4" : tierCode(t.tier);
|
||||
return (
|
||||
<button
|
||||
key={t.id}
|
||||
|
||||
@@ -8,6 +8,7 @@ import {
|
||||
type PreflightResult,
|
||||
type Template,
|
||||
} from "@/lib/deploy-preflight";
|
||||
import { isSaaSTenant } from "@/lib/tenant";
|
||||
import { MissingKeysModal } from "@/components/MissingKeysModal";
|
||||
|
||||
/**
|
||||
@@ -105,7 +106,7 @@ export function useTemplateDeploy(
|
||||
const ws = await api.post<{ id: string }>("/workspaces", {
|
||||
name: template.name,
|
||||
template: template.id,
|
||||
tier: template.tier,
|
||||
tier: isSaaSTenant() ? 4 : template.tier,
|
||||
canvas: coords,
|
||||
...(model ? { model } : {}),
|
||||
});
|
||||
|
||||
@@ -4,6 +4,7 @@ import (
|
||||
"bytes"
|
||||
"context"
|
||||
"encoding/json"
|
||||
"errors"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"regexp"
|
||||
@@ -85,8 +86,11 @@ func TestInstructionsList_ByWorkspaceID(t *testing.T) {
|
||||
if err := json.Unmarshal(w.Body.Bytes(), &result); err != nil {
|
||||
t.Fatalf("invalid JSON: %v", err)
|
||||
}
|
||||
if len(result) != 0 {
|
||||
t.Fatalf("expected 0 instructions, got %d", len(result))
|
||||
if len(result) != 2 {
|
||||
t.Fatalf("expected 2 instructions, got %d", len(result))
|
||||
}
|
||||
if result[0].Scope != "global" || result[1].Scope != "workspace" {
|
||||
t.Fatalf("expected global then workspace instructions, got %#v", result)
|
||||
}
|
||||
if err := mock.ExpectationsWereMet(); err != nil {
|
||||
t.Fatalf("unmet expectations: %v", err)
|
||||
@@ -215,8 +219,8 @@ func TestInstructionsHandler_Create_Success(t *testing.T) {
|
||||
if err := json.Unmarshal(w.Body.Bytes(), &out); err != nil {
|
||||
t.Fatalf("response not valid JSON: %v", err)
|
||||
}
|
||||
if out["id"] != "new-inst-1" {
|
||||
t.Errorf("expected id new-inst-1, got %s", out["id"])
|
||||
if out["id"] != "new-inst-id" {
|
||||
t.Errorf("expected id new-inst-id, got %s", out["id"])
|
||||
}
|
||||
if err := mock.ExpectationsWereMet(); err != nil {
|
||||
t.Errorf("unmet expectations: %v", err)
|
||||
|
||||
@@ -186,11 +186,16 @@ func (h *TemplatesHandler) List(c *gin.Context) {
|
||||
model = raw.RuntimeConfig.Model
|
||||
}
|
||||
|
||||
tier := raw.Tier
|
||||
if h.wh != nil && h.wh.IsSaaS() {
|
||||
tier = h.wh.DefaultTier()
|
||||
}
|
||||
|
||||
templates = append(templates, templateSummary{
|
||||
ID: id,
|
||||
Name: raw.Name,
|
||||
Description: raw.Description,
|
||||
Tier: raw.Tier,
|
||||
Tier: tier,
|
||||
Runtime: raw.Runtime,
|
||||
Model: model,
|
||||
Models: raw.RuntimeConfig.Models,
|
||||
@@ -340,6 +345,11 @@ func (h *TemplatesHandler) ListFiles(c *gin.Context) {
|
||||
if err != nil || path == walkRoot {
|
||||
return nil
|
||||
}
|
||||
// Skip symlinks to prevent path traversal via malicious symlinks
|
||||
// inside the workspace config directory (OFFSEC-010).
|
||||
if info.Mode()&os.ModeSymlink != 0 {
|
||||
return nil
|
||||
}
|
||||
rel, _ := filepath.Rel(walkRoot, path)
|
||||
// Enforce depth limit
|
||||
if strings.Count(rel, string(filepath.Separator))+1 > depth {
|
||||
|
||||
@@ -847,6 +847,58 @@ func TestListFiles_FallbackToHost_WithTemplate(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestListFiles_FallbackToHost_SkipsSymlinks(t *testing.T) {
|
||||
mock := setupTestDB(t)
|
||||
setupTestRedis(t)
|
||||
|
||||
tmpDir := t.TempDir()
|
||||
tmplDir := filepath.Join(tmpDir, "test-agent")
|
||||
if err := os.MkdirAll(tmplDir, 0755); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := os.WriteFile(filepath.Join(tmplDir, "config.yaml"), []byte("name: Test Agent\n"), 0644); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
secret := filepath.Join(t.TempDir(), "secret.txt")
|
||||
if err := os.WriteFile(secret, []byte("do-not-list"), 0600); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := os.Symlink(secret, filepath.Join(tmplDir, "leaked-secret")); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
|
||||
handler := NewTemplatesHandler(tmpDir, nil, nil)
|
||||
|
||||
mock.ExpectQuery(`SELECT name, COALESCE\(instance_id, ''\), COALESCE\(runtime, ''\) FROM workspaces WHERE id =`).
|
||||
WithArgs("ws-tmpl").
|
||||
WillReturnRows(sqlmock.NewRows([]string{"name", "instance_id", "runtime"}).AddRow("Test Agent", "", ""))
|
||||
|
||||
w := httptest.NewRecorder()
|
||||
c, _ := gin.CreateTestContext(w)
|
||||
c.Params = gin.Params{{Key: "id", Value: "ws-tmpl"}}
|
||||
c.Request = httptest.NewRequest("GET", "/workspaces/ws-tmpl/files", nil)
|
||||
|
||||
handler.ListFiles(c)
|
||||
|
||||
if w.Code != http.StatusOK {
|
||||
t.Fatalf("expected 200, got %d: %s", w.Code, w.Body.String())
|
||||
}
|
||||
|
||||
var resp []map[string]interface{}
|
||||
if err := json.Unmarshal(w.Body.Bytes(), &resp); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
for _, file := range resp {
|
||||
if file["path"] == "leaked-secret" {
|
||||
t.Fatalf("symlink should not be listed: %#v", resp)
|
||||
}
|
||||
}
|
||||
|
||||
if err := mock.ExpectationsWereMet(); err != nil {
|
||||
t.Errorf("unmet sqlmock expectations: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// ==================== GET /workspaces/:id/files/*path ====================
|
||||
|
||||
func TestReadFile_PathTraversal(t *testing.T) {
|
||||
@@ -1200,4 +1252,3 @@ func TestCWE78_DeleteFile_TraversalVariants(t *testing.T) {
|
||||
})
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
@@ -161,15 +161,14 @@ func (h *WorkspaceHandler) Create(c *gin.Context) {
|
||||
|
||||
id := uuid.New().String()
|
||||
awarenessNamespace := workspaceAwarenessNamespace(id)
|
||||
if payload.Tier == 0 {
|
||||
// SaaS-aware default. SaaS → T4 (full host access; each
|
||||
// workspace runs on its own sibling EC2 so the tier boundary
|
||||
// is a Docker resource limit on the only container present —
|
||||
// no neighbour to protect from). Self-hosted → T3 (read-write
|
||||
// workspace mount + Docker daemon access, most templates'
|
||||
// baseline). Lower tiers (T1 sandboxed, T2 standard) remain
|
||||
// explicit opt-ins for low-trust agents. Matches the canvas
|
||||
// CreateWorkspaceDialog defaults so the API and the UI agree.
|
||||
if h.IsSaaS() {
|
||||
// SaaS hard gate: every hosted workspace gets its own sibling
|
||||
// EC2 instance, so T4 is the only meaningful runtime boundary.
|
||||
// Do not trust stale clients/templates that still send T1/T2/T3.
|
||||
payload.Tier = 4
|
||||
} else if payload.Tier == 0 {
|
||||
// Self-hosted default remains T3. Lower tiers (T1 sandboxed,
|
||||
// T2 standard) stay explicit opt-ins for low-trust local agents.
|
||||
payload.Tier = h.DefaultTier()
|
||||
}
|
||||
|
||||
|
||||
@@ -15,6 +15,7 @@ import (
|
||||
"github.com/Molecule-AI/molecule-monorepo/platform/internal/models"
|
||||
"github.com/Molecule-AI/molecule-monorepo/platform/internal/provisioner"
|
||||
"github.com/Molecule-AI/molecule-monorepo/platform/internal/wsauth"
|
||||
"gopkg.in/yaml.v3"
|
||||
)
|
||||
|
||||
// logProvisionPanic is the deferred recover at the top of every provision
|
||||
@@ -472,9 +473,10 @@ func configDirName(workspaceID string) string {
|
||||
// runtime means bumping both this list and the Docker image tags.
|
||||
// knownRuntimes is populated from manifest.json at service init (see
|
||||
// runtime_registry.go). The package init order is:
|
||||
// 1. var knownRuntimes = fallbackRuntimes
|
||||
// 2. init() calls initKnownRuntimes() which replaces it if
|
||||
// manifest.json is readable.
|
||||
// 1. var knownRuntimes = fallbackRuntimes
|
||||
// 2. init() calls initKnownRuntimes() which replaces it if
|
||||
// manifest.json is readable.
|
||||
//
|
||||
// The fallback matters for unit tests that don't mount the manifest.
|
||||
//
|
||||
// "external" is a first-class runtime that intentionally does NOT
|
||||
@@ -539,6 +541,9 @@ func (h *WorkspaceHandler) ensureDefaultConfig(workspaceID string, payload model
|
||||
// org_import.go; consolidating prevents silent drift.
|
||||
model = models.DefaultModel(runtime)
|
||||
}
|
||||
if runtime == "claude-code" {
|
||||
model = normalizeClaudeCodeModel(model)
|
||||
}
|
||||
|
||||
// Sanitize name/role/model for YAML safety — always double-quote so
|
||||
// a crafted value with a newline or colon can't terminate the scalar
|
||||
@@ -554,6 +559,11 @@ func (h *WorkspaceHandler) ensureDefaultConfig(workspaceID string, payload model
|
||||
quoteModel := yamlQuote(model)
|
||||
configYAML := fmt.Sprintf("name: %s\ndescription: %s\nversion: 1.0.0\ntier: %d\nruntime: %s\n",
|
||||
quoteName, quoteRole, payload.Tier, runtime)
|
||||
if runtime == "claude-code" {
|
||||
if providersYAML := h.defaultTemplateProvidersYAML(runtime); providersYAML != "" {
|
||||
configYAML += providersYAML + "\n"
|
||||
}
|
||||
}
|
||||
|
||||
// Model always at top level — config.py reads raw["model"] for all runtimes.
|
||||
configYAML += fmt.Sprintf("model: %s\n", quoteModel)
|
||||
@@ -563,7 +573,11 @@ func (h *WorkspaceHandler) ensureDefaultConfig(workspaceID string, payload model
|
||||
// and preflight already validates that the env vars are present before
|
||||
// the agent loop starts. Hardcoding token names here caused #1028
|
||||
// (expired CLAUDE_CODE_OAUTH_TOKEN baked into config.yaml).
|
||||
configYAML += "runtime_config:\n timeout: 0\n"
|
||||
configYAML += "runtime_config:\n"
|
||||
if runtime == "claude-code" {
|
||||
configYAML += fmt.Sprintf(" model: %s\n", quoteModel)
|
||||
}
|
||||
configYAML += " timeout: 0\n"
|
||||
|
||||
files["config.yaml"] = []byte(configYAML)
|
||||
|
||||
@@ -571,6 +585,60 @@ func (h *WorkspaceHandler) ensureDefaultConfig(workspaceID string, payload model
|
||||
return files
|
||||
}
|
||||
|
||||
func normalizeClaudeCodeModel(model string) string {
|
||||
model = strings.TrimSpace(model)
|
||||
if before, after, ok := strings.Cut(model, "/"); ok && before != "" && after != "" {
|
||||
return after
|
||||
}
|
||||
return model
|
||||
}
|
||||
|
||||
func (h *WorkspaceHandler) defaultTemplateProvidersYAML(runtime string) string {
|
||||
if h.configsDir == "" {
|
||||
return ""
|
||||
}
|
||||
templateName := runtime + "-default"
|
||||
templatePath, err := resolveInsideRoot(h.configsDir, templateName)
|
||||
if err != nil {
|
||||
log.Printf("Provisioner: default template providers skipped for runtime %s: %v", runtime, err)
|
||||
return ""
|
||||
}
|
||||
data, err := os.ReadFile(filepath.Join(templatePath, "config.yaml"))
|
||||
if err != nil {
|
||||
return ""
|
||||
}
|
||||
|
||||
var root yaml.Node
|
||||
if err := yaml.Unmarshal(data, &root); err != nil {
|
||||
log.Printf("Provisioner: default template providers skipped for runtime %s: invalid YAML: %v", runtime, err)
|
||||
return ""
|
||||
}
|
||||
if len(root.Content) == 0 || root.Content[0].Kind != yaml.MappingNode {
|
||||
return ""
|
||||
}
|
||||
|
||||
mapping := root.Content[0]
|
||||
for i := 0; i+1 < len(mapping.Content); i += 2 {
|
||||
if mapping.Content[i].Value != "providers" {
|
||||
continue
|
||||
}
|
||||
out := yaml.Node{
|
||||
Kind: yaml.MappingNode,
|
||||
Content: []*yaml.Node{
|
||||
{Kind: yaml.ScalarNode, Value: "providers"},
|
||||
mapping.Content[i+1],
|
||||
},
|
||||
}
|
||||
encoded, err := yaml.Marshal(&out)
|
||||
if err != nil {
|
||||
log.Printf("Provisioner: default template providers skipped for runtime %s: marshal failed: %v", runtime, err)
|
||||
return ""
|
||||
}
|
||||
return strings.TrimRight(string(encoded), "\n")
|
||||
}
|
||||
return ""
|
||||
}
|
||||
|
||||
// deriveProviderFromModelSlug maps a hermes-agent model slug prefix to
|
||||
// its provider name — a Go translation of the case statement in
|
||||
// workspace-configs-templates/hermes/scripts/derive-provider.sh that we
|
||||
|
||||
@@ -261,6 +261,67 @@ func TestEnsureDefaultConfig_ClaudeCode(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestEnsureDefaultConfig_ClaudeCodeCopiesProviderRegistry(t *testing.T) {
|
||||
broadcaster := newTestBroadcaster()
|
||||
configsDir := t.TempDir()
|
||||
templateDir := filepath.Join(configsDir, "claude-code-default")
|
||||
if err := os.MkdirAll(templateDir, 0o755); err != nil {
|
||||
t.Fatalf("mkdir template: %v", err)
|
||||
}
|
||||
if err := os.WriteFile(filepath.Join(templateDir, "config.yaml"), []byte(`
|
||||
name: Claude Code Agent
|
||||
runtime: claude-code
|
||||
providers:
|
||||
- name: anthropic-oauth
|
||||
auth_mode: oauth
|
||||
model_aliases: [sonnet]
|
||||
auth_env: [CLAUDE_CODE_OAUTH_TOKEN]
|
||||
- name: minimax
|
||||
auth_mode: third_party_anthropic_compat
|
||||
model_prefixes: [minimax-]
|
||||
base_url: https://api.minimax.io/anthropic
|
||||
auth_env: [MINIMAX_API_KEY, ANTHROPIC_AUTH_TOKEN]
|
||||
runtime_config:
|
||||
model: sonnet
|
||||
`), 0o644); err != nil {
|
||||
t.Fatalf("write template: %v", err)
|
||||
}
|
||||
handler := NewWorkspaceHandler(broadcaster, nil, "http://localhost:8080", configsDir)
|
||||
|
||||
files := handler.ensureDefaultConfig("ws-code-123", models.CreateWorkspacePayload{
|
||||
Name: "Code Agent",
|
||||
Tier: 4,
|
||||
Runtime: "claude-code",
|
||||
Model: "minimax/MiniMax-M2.7",
|
||||
})
|
||||
|
||||
var parsed struct {
|
||||
Model string `yaml:"model"`
|
||||
Providers []struct {
|
||||
Name string `yaml:"name"`
|
||||
ModelPrefixes []string `yaml:"model_prefixes"`
|
||||
} `yaml:"providers"`
|
||||
RuntimeConfig struct {
|
||||
Model string `yaml:"model"`
|
||||
} `yaml:"runtime_config"`
|
||||
}
|
||||
if err := yaml.Unmarshal(files["config.yaml"], &parsed); err != nil {
|
||||
t.Fatalf("generated YAML invalid: %v\n%s", err, files["config.yaml"])
|
||||
}
|
||||
if parsed.Model != "MiniMax-M2.7" {
|
||||
t.Fatalf("top-level model = %q, want MiniMax-M2.7\n%s", parsed.Model, files["config.yaml"])
|
||||
}
|
||||
if parsed.RuntimeConfig.Model != "MiniMax-M2.7" {
|
||||
t.Fatalf("runtime_config.model = %q, want MiniMax-M2.7\n%s", parsed.RuntimeConfig.Model, files["config.yaml"])
|
||||
}
|
||||
if len(parsed.Providers) != 2 {
|
||||
t.Fatalf("providers len = %d, want 2\n%s", len(parsed.Providers), files["config.yaml"])
|
||||
}
|
||||
if parsed.Providers[1].Name != "minimax" || len(parsed.Providers[1].ModelPrefixes) != 1 || parsed.Providers[1].ModelPrefixes[0] != "minimax-" {
|
||||
t.Fatalf("minimax provider registry not preserved: %+v\n%s", parsed.Providers, files["config.yaml"])
|
||||
}
|
||||
}
|
||||
|
||||
func TestEnsureDefaultConfig_CustomModel(t *testing.T) {
|
||||
broadcaster := newTestBroadcaster()
|
||||
handler := NewWorkspaceHandler(broadcaster, nil, "http://localhost:8080", t.TempDir())
|
||||
|
||||
@@ -410,6 +410,44 @@ func TestWorkspaceCreate_DefaultsApplied(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestWorkspaceCreate_SaaSHardForcesTier4(t *testing.T) {
|
||||
mock := setupTestDB(t)
|
||||
setupTestRedis(t)
|
||||
broadcaster := newTestBroadcaster()
|
||||
handler := NewWorkspaceHandler(broadcaster, nil, "http://localhost:8080", t.TempDir())
|
||||
handler.SetCPProvisioner(&trackingCPProv{})
|
||||
|
||||
mock.ExpectBegin()
|
||||
mock.ExpectExec("INSERT INTO workspaces").
|
||||
WithArgs(sqlmock.AnyArg(), "SaaS External Agent", nil, 4, "external", sqlmock.AnyArg(), (*string)(nil), nil, "none", (*int64)(nil), models.DefaultMaxConcurrentTasks, "push").
|
||||
WillReturnResult(sqlmock.NewResult(0, 1))
|
||||
mock.ExpectCommit()
|
||||
mock.ExpectExec("INSERT INTO canvas_layouts").
|
||||
WithArgs(sqlmock.AnyArg(), float64(0), float64(0)).
|
||||
WillReturnResult(sqlmock.NewResult(0, 1))
|
||||
mock.ExpectExec("INSERT INTO structure_events").
|
||||
WillReturnResult(sqlmock.NewResult(0, 1))
|
||||
mock.ExpectExec("UPDATE workspaces SET url").
|
||||
WillReturnResult(sqlmock.NewResult(0, 1))
|
||||
mock.ExpectExec("INSERT INTO structure_events").
|
||||
WillReturnResult(sqlmock.NewResult(0, 1))
|
||||
|
||||
w := httptest.NewRecorder()
|
||||
c, _ := gin.CreateTestContext(w)
|
||||
body := `{"name":"SaaS External Agent","runtime":"external","external":true,"url":"https://example.com/agent","tier":2}`
|
||||
c.Request = httptest.NewRequest("POST", "/workspaces", bytes.NewBufferString(body))
|
||||
c.Request.Header.Set("Content-Type", "application/json")
|
||||
|
||||
handler.Create(c)
|
||||
|
||||
if w.Code != http.StatusCreated {
|
||||
t.Errorf("expected status 201, got %d: %s", w.Code, w.Body.String())
|
||||
}
|
||||
if err := mock.ExpectationsWereMet(); err != nil {
|
||||
t.Errorf("unmet sqlmock expectations: %v", err)
|
||||
}
|
||||
}
|
||||
|
||||
// TestWorkspaceCreate_WithSecrets_Persists asserts that secrets in the create
|
||||
// payload are written to workspace_secrets inside the same transaction as the
|
||||
// workspace row, and that the handler returns 201.
|
||||
|
||||
@@ -158,6 +158,7 @@ type cpProvisionRequest struct {
|
||||
Tier int `json:"tier"`
|
||||
PlatformURL string `json:"platform_url"`
|
||||
Env map[string]string `json:"env"`
|
||||
ConfigFiles map[string]string `json:"config_files,omitempty"`
|
||||
}
|
||||
|
||||
type cpProvisionResponse struct {
|
||||
@@ -181,6 +182,11 @@ func (p *CPProvisioner) Start(ctx context.Context, cfg WorkspaceConfig) (string,
|
||||
}
|
||||
env["ADMIN_TOKEN"] = p.adminToken
|
||||
}
|
||||
configFiles, err := collectCPConfigFiles(cfg)
|
||||
if err != nil {
|
||||
return "", fmt.Errorf("cp provisioner: collect config files: %w", err)
|
||||
}
|
||||
|
||||
req := cpProvisionRequest{
|
||||
OrgID: p.orgID,
|
||||
WorkspaceID: cfg.WorkspaceID,
|
||||
@@ -188,6 +194,7 @@ func (p *CPProvisioner) Start(ctx context.Context, cfg WorkspaceConfig) (string,
|
||||
Tier: cfg.Tier,
|
||||
PlatformURL: cfg.PlatformURL,
|
||||
Env: env,
|
||||
ConfigFiles: configFiles,
|
||||
}
|
||||
|
||||
body, err := json.Marshal(req)
|
||||
@@ -241,6 +248,11 @@ func (p *CPProvisioner) Start(ctx context.Context, cfg WorkspaceConfig) (string,
|
||||
|
||||
const cpConfigFilesMaxBytes = 12 << 10
|
||||
|
||||
func isCPTemplateConfigFile(name string) bool {
|
||||
name = filepath.ToSlash(filepath.Clean(name))
|
||||
return name == "config.yaml" || strings.HasPrefix(name, "prompts/")
|
||||
}
|
||||
|
||||
func collectCPConfigFiles(cfg WorkspaceConfig) (map[string]string, error) {
|
||||
files := make(map[string]string)
|
||||
total := 0
|
||||
@@ -294,6 +306,9 @@ func collectCPConfigFiles(cfg WorkspaceConfig) (map[string]string, error) {
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
if !isCPTemplateConfigFile(rel) {
|
||||
return nil
|
||||
}
|
||||
data, err := os.ReadFile(path)
|
||||
if err != nil {
|
||||
return err
|
||||
@@ -314,6 +329,7 @@ func collectCPConfigFiles(cfg WorkspaceConfig) (map[string]string, error) {
|
||||
}
|
||||
return files, nil
|
||||
}
|
||||
|
||||
// Stop terminates the workspace's EC2 instance via the control plane.
|
||||
//
|
||||
// Looks up the actual EC2 instance_id from the workspaces table before
|
||||
@@ -468,7 +484,9 @@ func (p *CPProvisioner) IsRunning(ctx context.Context, workspaceID string) (bool
|
||||
// Don't leak the body — upstream errors may echo headers.
|
||||
return true, fmt.Errorf("cp provisioner: status: unexpected %d", resp.StatusCode)
|
||||
}
|
||||
var result struct{ State string `json:"state"` }
|
||||
var result struct {
|
||||
State string `json:"state"`
|
||||
}
|
||||
// Cap body read at 64 KiB for parity with Start — a misconfigured
|
||||
// or compromised CP streaming a huge body could otherwise exhaust
|
||||
// memory in this hot path (called reactively per-request from
|
||||
|
||||
@@ -1,11 +1,15 @@
|
||||
package provisioner
|
||||
|
||||
import (
|
||||
"bytes"
|
||||
"context"
|
||||
"encoding/base64"
|
||||
"encoding/json"
|
||||
"io"
|
||||
"net/http"
|
||||
"net/http/httptest"
|
||||
"os"
|
||||
"path/filepath"
|
||||
"strings"
|
||||
"testing"
|
||||
"time"
|
||||
@@ -213,6 +217,59 @@ func TestStart_HappyPath(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestStart_SendsTemplateAndGeneratedConfigFiles(t *testing.T) {
|
||||
tmpl := t.TempDir()
|
||||
if err := os.WriteFile(filepath.Join(tmpl, "config.yaml"), []byte("name: template\n"), 0o600); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := os.WriteFile(filepath.Join(tmpl, "adapter.py"), bytes.Repeat([]byte("x"), cpConfigFilesMaxBytes), 0o600); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := os.Mkdir(filepath.Join(tmpl, "prompts"), 0o700); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
if err := os.WriteFile(filepath.Join(tmpl, "prompts", "system.md"), []byte("hello"), 0o600); err != nil {
|
||||
t.Fatal(err)
|
||||
}
|
||||
|
||||
var body cpProvisionRequest
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, r *http.Request) {
|
||||
if err := json.NewDecoder(r.Body).Decode(&body); err != nil {
|
||||
t.Errorf("decode request: %v", err)
|
||||
}
|
||||
w.WriteHeader(http.StatusCreated)
|
||||
_, _ = io.WriteString(w, `{"instance_id":"i-abc123","state":"pending"}`)
|
||||
}))
|
||||
defer srv.Close()
|
||||
|
||||
p := &CPProvisioner{baseURL: srv.URL, orgID: "org-1", httpClient: srv.Client()}
|
||||
_, err := p.Start(context.Background(), WorkspaceConfig{
|
||||
WorkspaceID: "ws-1",
|
||||
Runtime: "claude-code",
|
||||
Tier: 4,
|
||||
PlatformURL: "http://tenant",
|
||||
TemplatePath: tmpl,
|
||||
ConfigFiles: map[string][]byte{
|
||||
"config.yaml": []byte("name: generated\n"),
|
||||
},
|
||||
})
|
||||
if err != nil {
|
||||
t.Fatalf("Start: %v", err)
|
||||
}
|
||||
|
||||
wantConfig := base64.StdEncoding.EncodeToString([]byte("name: generated\n"))
|
||||
if got := body.ConfigFiles["config.yaml"]; got != wantConfig {
|
||||
t.Errorf("config.yaml payload = %q, want generated override %q", got, wantConfig)
|
||||
}
|
||||
wantPrompt := base64.StdEncoding.EncodeToString([]byte("hello"))
|
||||
if got := body.ConfigFiles["prompts/system.md"]; got != wantPrompt {
|
||||
t.Errorf("prompt payload = %q, want %q", got, wantPrompt)
|
||||
}
|
||||
if _, ok := body.ConfigFiles["adapter.py"]; ok {
|
||||
t.Error("non-config template file adapter.py must not be sent to CP")
|
||||
}
|
||||
}
|
||||
|
||||
// TestStart_Non201ReturnsStructuredError — when CP returns 401 with a
|
||||
// structured {"error":"..."} body, Start surfaces that error message.
|
||||
// Verifies the defense against log-leaking raw upstream bodies.
|
||||
@@ -416,9 +473,9 @@ func TestStop_4xxResponseSurfacesError(t *testing.T) {
|
||||
func TestStop_2xxVariantsAllSucceed(t *testing.T) {
|
||||
primeInstanceIDLookup(t, map[string]string{"ws-1": "i-ok"})
|
||||
for _, code := range []int{
|
||||
http.StatusOK, // 200
|
||||
http.StatusAccepted, // 202
|
||||
http.StatusNoContent, // 204
|
||||
http.StatusOK, // 200
|
||||
http.StatusAccepted, // 202
|
||||
http.StatusNoContent, // 204
|
||||
} {
|
||||
srv := httptest.NewServer(http.HandlerFunc(func(w http.ResponseWriter, _ *http.Request) {
|
||||
w.WriteHeader(code)
|
||||
@@ -486,11 +543,11 @@ func TestIsRunning_ParsesStateField(t *testing.T) {
|
||||
_, _ = io.WriteString(w, `{"state":"`+state+`"}`)
|
||||
}))
|
||||
p := &CPProvisioner{
|
||||
baseURL: srv.URL,
|
||||
orgID: "org-1",
|
||||
baseURL: srv.URL,
|
||||
orgID: "org-1",
|
||||
sharedSecret: "s3cret",
|
||||
adminToken: "tok-xyz",
|
||||
httpClient: srv.Client(),
|
||||
httpClient: srv.Client(),
|
||||
}
|
||||
got, err := p.IsRunning(context.Background(), "ws-1")
|
||||
srv.Close()
|
||||
|
||||
@@ -773,6 +773,15 @@ func ApplyTierConfig(hostCfg *container.HostConfig, cfg WorkspaceConfig, configM
|
||||
|
||||
// CopyTemplateToContainer copies files from a host directory into /configs in the container.
|
||||
func (p *Provisioner) CopyTemplateToContainer(ctx context.Context, containerID, templatePath string) error {
|
||||
buf, err := buildTemplateTar(templatePath)
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
|
||||
return p.cli.CopyToContainer(ctx, containerID, "/configs", buf, container.CopyToContainerOptions{})
|
||||
}
|
||||
|
||||
func buildTemplateTar(templatePath string) (*bytes.Buffer, error) {
|
||||
// Resolve symlinks at the root before walking. filepath.Walk does
|
||||
// NOT follow a symlink that IS the root — it Lstats the path, sees
|
||||
// a symlink (non-directory), and emits exactly one entry without
|
||||
@@ -795,6 +804,15 @@ func (p *Provisioner) CopyTemplateToContainer(ctx context.Context, containerID,
|
||||
if err != nil {
|
||||
return err
|
||||
}
|
||||
// OFFSEC-010: skip symlinks to prevent path traversal via malicious
|
||||
// template symlinks (e.g. template/.ssh → /root/.ssh). filepath.Walk
|
||||
// follows symlinks by default, so without this guard a crafted symlink
|
||||
// inside the template directory could escape to include arbitrary host
|
||||
// files in the tar archive. We intentionally skip rather than error so
|
||||
// a broken symlink in an org template is a silent no-op.
|
||||
if info.Mode()&os.ModeSymlink != 0 {
|
||||
return nil
|
||||
}
|
||||
rel, err := filepath.Rel(templatePath, path)
|
||||
if err != nil {
|
||||
return err
|
||||
@@ -835,13 +853,13 @@ func (p *Provisioner) CopyTemplateToContainer(ctx context.Context, containerID,
|
||||
return nil
|
||||
})
|
||||
if err != nil {
|
||||
return fmt.Errorf("failed to create tar from %s: %w", templatePath, err)
|
||||
return nil, fmt.Errorf("failed to create tar from %s: %w", templatePath, err)
|
||||
}
|
||||
if err := tw.Close(); err != nil {
|
||||
return fmt.Errorf("failed to close tar writer: %w", err)
|
||||
return nil, fmt.Errorf("failed to close tar writer: %w", err)
|
||||
}
|
||||
|
||||
return p.cli.CopyToContainer(ctx, containerID, "/configs", &buf, container.CopyToContainerOptions{})
|
||||
return &buf, nil
|
||||
}
|
||||
|
||||
// WriteFilesToContainer writes in-memory files into /configs in the container.
|
||||
|
||||
@@ -1,7 +1,9 @@
|
||||
package provisioner
|
||||
|
||||
import (
|
||||
"archive/tar"
|
||||
"errors"
|
||||
"io"
|
||||
"os"
|
||||
"path/filepath"
|
||||
"strings"
|
||||
@@ -80,6 +82,54 @@ func TestStartSeedsConfigsBeforeContainerStart(t *testing.T) {
|
||||
}
|
||||
}
|
||||
|
||||
func TestBuildTemplateTar_SkipsSymlinks(t *testing.T) {
|
||||
dir := t.TempDir()
|
||||
if err := os.WriteFile(filepath.Join(dir, "config.yaml"), []byte("name: safe\n"), 0644); err != nil {
|
||||
t.Fatalf("write config: %v", err)
|
||||
}
|
||||
outside := filepath.Join(t.TempDir(), "secret.txt")
|
||||
if err := os.WriteFile(outside, []byte("do-not-copy\n"), 0644); err != nil {
|
||||
t.Fatalf("write outside target: %v", err)
|
||||
}
|
||||
if err := os.Symlink(outside, filepath.Join(dir, "linked-secret.txt")); err != nil {
|
||||
t.Fatalf("create symlink: %v", err)
|
||||
}
|
||||
|
||||
buf, err := buildTemplateTar(dir)
|
||||
if err != nil {
|
||||
t.Fatalf("buildTemplateTar: %v", err)
|
||||
}
|
||||
|
||||
names := map[string]string{}
|
||||
tr := tar.NewReader(buf)
|
||||
for {
|
||||
hdr, err := tr.Next()
|
||||
if errors.Is(err, io.EOF) {
|
||||
break
|
||||
}
|
||||
if err != nil {
|
||||
t.Fatalf("read tar: %v", err)
|
||||
}
|
||||
body, err := io.ReadAll(tr)
|
||||
if err != nil {
|
||||
t.Fatalf("read body for %s: %v", hdr.Name, err)
|
||||
}
|
||||
names[hdr.Name] = string(body)
|
||||
}
|
||||
|
||||
if got := names["config.yaml"]; got != "name: safe\n" {
|
||||
t.Fatalf("config.yaml body = %q, want safe config", got)
|
||||
}
|
||||
if _, ok := names["linked-secret.txt"]; ok {
|
||||
t.Fatalf("symlink entry was copied into template tar: %#v", names)
|
||||
}
|
||||
for name, body := range names {
|
||||
if strings.Contains(body, "do-not-copy") {
|
||||
t.Fatalf("symlink target leaked through %s: %q", name, body)
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
// baseHostConfig returns a fresh HostConfig with typical pre-tier binds,
|
||||
// mimicking what Start() builds before calling ApplyTierConfig.
|
||||
func baseHostConfig(pluginsPath string) *container.HostConfig {
|
||||
|
||||
Reference in New Issue
Block a user