code-coordinator 0.5.297__tar.gz → 0.5.299__tar.gz

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-coordinator
-Version: 0.5.297
+Version: 0.5.299
 Summary: Coordinate Claude Code workers across multiple machines and repos.
 Author-email: John Donaghy <johnfdonaghy@gmail.com>
 License-Expression: LicenseRef-FSL-1.1-MIT

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/code_coordinator.egg-info/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: code-coordinator
-Version: 0.5.297
+Version: 0.5.299
 Summary: Coordinate Claude Code workers across multiple machines and repos.
 Author-email: John Donaghy <johnfdonaghy@gmail.com>
 License-Expression: LicenseRef-FSL-1.1-MIT

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/code_coordinator.egg-info/SOURCES.txt

@@ -223,9 +223,11 @@ coord/health/registry.py
 coord/health/render.py
 coord/health/units.py
 coord/health/checks/__init__.py
+coord/health/checks/_ci_pin.py
 coord/health/checks/agent_install.py
 coord/health/checks/cargo_targets.py
 coord/health/checks/claude_binary.py
+coord/health/checks/coord_tui_ci_pin.py
 coord/health/checks/coord_web_ci_pin.py
 coord/health/checks/deploy_lane_facts.py
 coord/health/checks/disk.py
@@ -299,6 +301,7 @@ docs/ADR_COORD_TUI_CI.md
 docs/ADR_COORD_WEB_ACCEPTANCE_SUITE.md
 docs/ADR_COORD_WEB_CI.md
 docs/ADR_COORD_WEB_DIST.md
+docs/ADR_PY_RUST_MIRRORS.md
 docs/AGENT_OPERATIONS.md
 docs/ARCHITECTURE.md
 docs/ARCH_SECURITY_GATES.md
@@ -572,6 +575,7 @@ tests/test_gates.py
 tests/test_gather_fleet_tmux_sessions.py
 tests/test_gen_mock_index.py
 tests/test_generated_rust_fixture.py
+tests/test_generated_rust_requests.py
 tests/test_generated_types_fixture.py
 tests/test_github_ops.py
 tests/test_github_throttle.py
@@ -584,6 +588,7 @@ tests/test_health_cli.py
 tests/test_health_cli_graph.py
 tests/test_health_config.py
 tests/test_health_context.py
+tests/test_health_coord_tui_ci_pin.py
 tests/test_health_coord_web_ci_pin.py
 tests/test_health_deploy_lane_facts.py
 tests/test_health_fix.py
@@ -669,6 +674,7 @@ tests/test_provider_seam.py
 tests/test_providers.py
 tests/test_providers_config_labels.py
 tests/test_providers_pty.py
+tests/test_py_rust_mirror_inventory.py
 tests/test_quiet_hours.py
 tests/test_quiet_hours_store_2146.py
 tests/test_reap_merged_sessions.py

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/code_coordinator.egg-info/scm_file_list.json

@@ -140,9 +140,11 @@
     "coord/health/__init__.py",
     "coord/health/aggregate.py",
     "coord/health/checks/__init__.py",
+    "coord/health/checks/_ci_pin.py",
     "coord/health/checks/agent_install.py",
     "coord/health/checks/cargo_targets.py",
     "coord/health/checks/claude_binary.py",
+    "coord/health/checks/coord_tui_ci_pin.py",
     "coord/health/checks/coord_web_ci_pin.py",
     "coord/health/checks/deploy_lane_facts.py",
     "coord/health/checks/disk.py",
@@ -291,6 +293,7 @@
     "docs/ADR_COORD_WEB_ACCEPTANCE_SUITE.md",
     "docs/ADR_COORD_WEB_CI.md",
     "docs/ADR_COORD_WEB_DIST.md",
+    "docs/ADR_PY_RUST_MIRRORS.md",
     "docs/AGENT_OPERATIONS.md",
     "docs/ARCHITECTURE.md",
     "docs/ARCH_SECURITY_GATES.md",
@@ -598,6 +601,7 @@
     "tests/test_gather_fleet_tmux_sessions.py",
     "tests/test_gen_mock_index.py",
     "tests/test_generated_rust_fixture.py",
+    "tests/test_generated_rust_requests.py",
     "tests/test_generated_types_fixture.py",
     "tests/test_github_ops.py",
     "tests/test_github_throttle.py",
@@ -610,6 +614,7 @@
     "tests/test_health_cli_graph.py",
     "tests/test_health_config.py",
     "tests/test_health_context.py",
+    "tests/test_health_coord_tui_ci_pin.py",
     "tests/test_health_coord_web_ci_pin.py",
     "tests/test_health_deploy_lane_facts.py",
     "tests/test_health_fix.py",
@@ -695,6 +700,7 @@
     "tests/test_providers.py",
     "tests/test_providers_config_labels.py",
     "tests/test_providers_pty.py",
+    "tests/test_py_rust_mirror_inventory.py",
     "tests/test_quiet_hours.py",
     "tests/test_quiet_hours_store_2146.py",
     "tests/test_reap_merged_sessions.py",

code_coordinator-0.5.299/code_coordinator.egg-info/scm_version.json

@@ -0,0 +1,8 @@
+{
+  "tag": "0.5.299",
+  "distance": 0,
+  "node": "g472c4dff139719a7463904657ff555bbebb076af",
+  "dirty": false,
+  "branch": "HEAD",
+  "node_date": "2026-08-30"
+}

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/coord/deploy/coord-agent.service

@@ -74,15 +74,40 @@
 #   systemctl --user restart coord-agent
 #
 # See docs/AGENT_OPERATIONS.md → "Install a new agent (first time)".
-
+#
+# #2938: Restart=always, not on-failure. `POST /update` stops this unit's
+# own process so it can come back on the swapped venv — the agent IS its own
+# recovery channel here, via an explicit `systemctl --user restart` the
+# process fires at itself just before exiting (coord/agent_app.py's
+# `_default_exec_restart`). When that explicit restart lands, this process
+# exits 0 either way (a clean stop) and `Restart=on-failure` — correctly, by
+# its own definition — declines to act on a clean exit. If the explicit
+# restart request is ever lost (a dropped systemd/D-Bus call, a race with
+# the process's own exit, anything), NOTHING brings the unit back: it sits
+# `enabled`/`inactive` forever, port 7433 refused, and the fleet has lost
+# its only remote-control channel to that host. `Restart=always` makes
+# systemd itself the backstop for a clean exit, independent of whether the
+# self-restart call landed — dell64, 2026-08-29, ~23 minutes of that exact
+# outage, recovered only because someone could walk to the machine.
+#
+# StartLimitIntervalSec/StartLimitBurst (#2938) bound it: 5 restarts inside
+# 5 minutes and systemd stops trying, landing the unit in `failed` with
+# `Result: start-limit-hit` — visible in `systemctl --user status` and the
+# journal — rather than crash-looping forever. A binary that is genuinely
+# broken (not just "the explicit restart didn't land") must fail loud, not
+# spin silently; `Restart=always` without a limit would trade one silent
+# failure (never restarts) for another (restarts forever with no signal
+# anything is wrong).
 [Unit]
 Description=Coordinator agent server (port <PORT>)
 After=network-online.target
+StartLimitIntervalSec=300
+StartLimitBurst=5
 
 [Service]
 Type=simple
 ExecStart=%h/.coord-venv/bin/coord agent --machine <MACHINE_NAME> --port <PORT>
-Restart=on-failure
+Restart=always
 RestartSec=5
 Environment=PATH=%h/.coord-venv/bin:%h/.cargo/bin:%h/.local/bin:/usr/local/bin:/usr/bin:/bin
 

{code_coordinator-0.5.297 → code_coordinator-0.5.299/coord}/deploy/coord-serve.service

@@ -36,11 +36,23 @@
 #     put --token on ExecStart — it would leak via `ps`).
 #
 # See docs/AGENT_OPERATIONS.md → "Control-center daemon (coord serve)".
-
+#
+# #2938: Restart=always (was on-failure), + a StartLimitIntervalSec/
+# StartLimitBurst backstop — see deploy/coord-agent.service's header for the
+# full incident this closes (a clean stop that Restart=on-failure correctly
+# declines to act on, with no other recovery path, leaving the unit
+# `enabled`/`inactive` forever). coord-serve restarts through the same
+# `_restart_sibling_unit` caller as coord-agent's own self-restart
+# (coord/agent_app.py, #2069), so it is exposed to the identical failure
+# shape if that restart request is ever lost. 5 restarts / 5 minutes before
+# systemd gives up and lands the unit in `failed` (`start-limit-hit`) —
+# visible, not a silent crash-loop.
 [Unit]
 Description=coord serve — portable control-center daemon (#584)
 After=network-online.target
 Wants=network-online.target
+StartLimitIntervalSec=300
+StartLimitBurst=5
 
 [Service]
 Type=simple
@@ -78,7 +90,7 @@ Type=simple
 # ahead of ~/.local/bin or ~/.coord-venv/bin.
 Environment=PATH=%h/.cargo/bin:%h/.local/bin:%h/src/code-coordinator/.venv/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/snap/bin
 ExecStart=%h/.coord-venv/bin/coord serve --config %h/.coord/coordinator.yml --host 0.0.0.0 --port 7435
-Restart=on-failure
+Restart=always
 RestartSec=3
 # #2095: coord-web (its sibling on the same host, restarted through the same
 # `_restart_sibling_unit` caller) went down for the length of a fleet roll

{code_coordinator-0.5.297 → code_coordinator-0.5.299}/coord/deploy/coord-web.service

@@ -92,11 +92,24 @@
 # (#1830), even though the underlying reason differs (that one protects a
 # spawned tmux server from this unit's own cgroup teardown; this one avoids
 # ever having to signal a nonexistent auxiliary process group at all).
+#
+# #2938: Restart=always (was on-failure), + a StartLimitIntervalSec/
+# StartLimitBurst backstop — see deploy/coord-agent.service's header for the
+# full incident this closes (a clean stop that Restart=on-failure correctly
+# declines to act on, with no other recovery path, leaving the unit
+# `enabled`/`inactive` forever). coord-web restarts through the same
+# `_restart_sibling_unit` caller as coord-agent's own self-restart
+# (coord/agent_app.py, #2069), so it is exposed to the identical failure
+# shape if that restart request is ever lost. 5 restarts / 5 minutes before
+# systemd gives up and lands the unit in `failed` (`start-limit-hit`) —
+# visible, not a silent crash-loop.
 
 [Unit]
 Description=coord web — phone/web control-center dashboard (#700)
 After=network-online.target
 Wants=network-online.target
+StartLimitIntervalSec=300
+StartLimitBurst=5
 
 [Service]
 Type=simple
@@ -118,7 +131,7 @@ ExecStart=%h/.coord-venv/bin/coord web --config %h/.coord/coordinator.yml --host
 # setting it here too ensures any other subprocess spawned by coord web also
 # has a sane terminal type without requiring a code fix for each callsite.
 Environment=TERM=xterm-256color
-Restart=on-failure
+Restart=always
 RestartSec=3
 # #2095 — see the header comment above. Bounds the SSE-drain hang that took
 # the dashboard down during the 2026-08-10 fleet roll.

code_coordinator-0.5.299/coord/health/checks/_ci_pin.py

@@ -0,0 +1,236 @@
+"""Shared machinery for the cross-repo "which `coord` does *their* CI install?" checks.
+
+Extracted from :mod:`coord.health.checks.coord_web_ci_pin` (#2006) when #2900
+added its second consumer, :mod:`coord.health.checks.coord_tui_ci_pin`. Both
+answer the same question about a different downstream repo, so the parsing —
+walk a checkout's workflow YAML, find the ``pip install`` steps, pull the
+`coord` requirement out of each — is identical and lives here once. What is
+*not* shared is the grading: what counts as a fault depends on why that repo
+installs `coord` at all, and each check module owns its own verdicts.
+
+Nothing in here is a check. The module is underscore-prefixed on purpose:
+:func:`coord.health.registry.discover` skips ``_``-prefixed modules, so this
+cannot accidentally register anything.
+
+**Local-filesystem only.** Every function here reads a checkout already on
+this machine. No ``gh`` call, no network — cheap enough for the health poll
+tick, and absent on a machine with no such checkout, which each caller
+reports as OK ("not present on this machine") the same way every other
+checkout-derived lane reports absence.
+"""
+
+from __future__ import annotations
+
+import re
+from collections.abc import Iterator
+from dataclasses import dataclass
+from pathlib import Path
+
+import yaml
+
+#: The extra that carries uvicorn/Starlette. Since the base/`[server]` split
+#: (#1237) the bare distribution installs a client-only coord: it cannot
+#: serve anything, and it cannot even *import* ``coord.serve_app`` (Starlette
+#: is imported at module scope there). Both downstream repos need the extra —
+#: coord-web to boot ``coord web --fixture`` as its Playwright ``webServer``
+#: (#1818), coord-tui because ``scripts/codegen.py --rust`` reads
+#: ``coord.serve_app.openapi_spec()`` (#1941).
+SERVER_EXTRA = "server"
+
+#: PyPI cannot rename a project, so ``claude-coordinator`` is a permanent
+#: tombstone that will never gain another release (#2106). CI still asking
+#: for it does not get a stale coord — it gets whatever ancient version that
+#: tombstone last published, forever.
+TOMBSTONE_DIST = "claude-coordinator"
+
+#: The live distribution name.
+LIVE_DIST = "code-coordinator"
+
+_INSTALL_RE = re.compile(r"\b(?:pip3?|uv\s+pip|pipx)\s+install\b", re.IGNORECASE)
+
+# `code-coordinator[server]>=0.4.90` and friends. Deliberately tolerant of
+# quoting (`'code-coordinator[server]'`) and of the underscore spelling pip
+# normalises away, because this is reading someone else's hand-written YAML.
+_REQ_RE = re.compile(
+    r"(?P<dist>code[-_]coordinator|claude[-_]coordinator)"
+    r"(?:\s*\[(?P<extras>[^\]]*)\])?"
+    r"(?P<constraint>"
+    r"(?:\s*(?:===|[<>!~=]=|[<>])\s*[^\s'\",;\\]+)"
+    r"(?:\s*,\s*(?:===|[<>!~=]=|[<>])\s*[^\s'\",;\\]+)*"
+    r")?",
+    re.IGNORECASE,
+)
+
+
+@dataclass(frozen=True)
+class CoordPin:
+    """One ``pip install <coord>`` requirement found in a downstream repo's CI."""
+
+    workflow: str  # file name, e.g. "ci.yml"
+    job: str  # job id, e.g. "e2e"
+    spec: str  # the requirement as written, e.g. "code-coordinator[server]"
+    dist: str  # normalised distribution name
+    extras: tuple[str, ...] = ()
+    constraint: str | None = None  # ">=0.4.90", "==0.4.90", or None
+
+    @property
+    def has_server_extra(self) -> bool:
+        return SERVER_EXTRA in self.extras
+
+    @property
+    def is_tombstone(self) -> bool:
+        return self.dist == TOMBSTONE_DIST
+
+    @property
+    def is_exact_pin(self) -> bool:
+        """True for ``==``/``===``, the spec shape that rots silently.
+
+        ``~=`` is deliberately included: ``~=0.4.90`` freezes the minor
+        series just as effectively for a project on a single ``0.x`` line.
+        """
+        c = (self.constraint or "").lstrip()
+        return c.startswith(("==", "===", "~="))
+
+    @property
+    def floor(self) -> str | None:
+        """The ``>=`` version, when the spec declares one."""
+        for part in (self.constraint or "").split(","):
+            part = part.strip()
+            if part.startswith(">="):
+                return part[2:].strip()
+        return None
+
+
+def norm_dist(raw: str) -> str:
+    """``Coord_TUI`` -> ``coord-tui``; ``code_coordinator`` -> ``code-coordinator``."""
+    return raw.strip().lower().replace("_", "-")
+
+
+def _version_tuple(s: str) -> tuple[int, ...] | None:
+    parts = re.findall(r"\d+", s or "")
+    return tuple(int(p) for p in parts) if parts else None
+
+
+def release_tuple(s: str) -> tuple[int, ...] | None:
+    """A comparable tuple, but only for a version that is a real release.
+
+    ``coord.__version__`` falls back to ``0+unknown`` when it cannot resolve
+    a distribution (a source checkout with no install, a wheel-less CI box).
+    Comparing a CI floor against *that* would report every floor as "ahead of
+    this machine" — a fabricated finding, and fabricating findings is how a
+    check earns the right to be ignored. Unknown stays unknown.
+    """
+    if not s or "unknown" in s.lower():
+        return None
+    t = _version_tuple(s)
+    return t if t and len(t) >= 2 else None
+
+
+def floor_exceeds(floor: str, local: str) -> bool:
+    """True iff *floor* is strictly newer than *local*, both being releases.
+
+    Compares at equal precision — a floor of ``0.4`` is satisfied by a local
+    ``0.4.91``, so the shorter tuple is padded rather than compared as-is
+    (``(0, 4, 91) > (0, 4)`` would otherwise read as "ahead").
+    """
+    f, loc = release_tuple(floor), release_tuple(local)
+    if not f or not loc:
+        return False
+    width = max(len(f), len(loc))
+
+    def pad(t: tuple[int, ...]) -> tuple[int, ...]:
+        return t + (0,) * (width - len(t))
+
+    return pad(f) > pad(loc)
+
+
+def iter_run_steps(checkout_path: Path) -> Iterator[tuple[str, str, str]]:
+    """Yield ``(workflow_file_name, job_id, run_script)`` for every step.
+
+    A sibling of :func:`coord.health.checks.toolchain._iter_workflow_steps`,
+    but keeping the workflow/job labels: a finding here has to name *where*
+    in someone else's repo to go fix it, or it is just as much of a mystery
+    as the thing it replaced.
+    """
+    workflows_dir = checkout_path / ".github" / "workflows"
+    if not workflows_dir.is_dir():
+        return
+    try:
+        files = sorted(workflows_dir.glob("*.yml")) + sorted(workflows_dir.glob("*.yaml"))
+    except OSError:
+        return
+    for wf_file in files:
+        try:
+            doc = yaml.safe_load(wf_file.read_text(encoding="utf-8", errors="replace"))
+        except (OSError, yaml.YAMLError):
+            continue
+        if not isinstance(doc, dict):
+            continue
+        jobs = doc.get("jobs")
+        if not isinstance(jobs, dict):
+            continue
+        for job_id, job in jobs.items():
+            if not isinstance(job, dict):
+                continue
+            for step in job.get("steps") or []:
+                if not isinstance(step, dict):
+                    continue
+                run = step.get("run")
+                if isinstance(run, str) and run.strip():
+                    yield wf_file.name, str(job_id), run
+
+
+def find_coord_pins(checkout_path: Path) -> list[CoordPin]:
+    """Every ``pip install <coord>`` requirement in the checkout's workflows.
+
+    Only ``run:`` scripts that actually invoke an installer are scanned — a
+    comment or an echo naming the package is not a pin, and treating it as
+    one would grade prose.
+    """
+    found: list[CoordPin] = []
+    seen: set[tuple[str, str, str]] = set()
+    for workflow, job, run in iter_run_steps(checkout_path):
+        for line in run.splitlines():
+            if not _INSTALL_RE.search(line):
+                continue
+            for m in _REQ_RE.finditer(line):
+                extras = tuple(
+                    e.strip().lower() for e in (m.group("extras") or "").split(",") if e.strip()
+                )
+                constraint = (m.group("constraint") or "").strip() or None
+                spec = m.group(0).strip()
+                key = (workflow, job, spec)
+                if key in seen:
+                    continue
+                seen.add(key)
+                found.append(
+                    CoordPin(
+                        workflow=workflow,
+                        job=job,
+                        spec=spec,
+                        dist=norm_dist(m.group("dist")),
+                        extras=extras,
+                        constraint=constraint,
+                    )
+                )
+    return found
+
+
+def where(pin: CoordPin) -> str:
+    """``ci.yml:e2e`` — the workflow:job a finding should send you to."""
+    return f"{pin.workflow}:{pin.job}"
+
+
+def pin_values(pins: list[CoordPin]) -> list[dict]:
+    """The JSON-able ``values["pins"]`` payload both checks report."""
+    return [
+        {
+            "workflow": p.workflow,
+            "job": p.job,
+            "spec": p.spec,
+            "dist": p.dist,
+            "extras": list(p.extras),
+            "constraint": p.constraint,
+        }
+        for p in pins
+    ]