coderouter-cli 2.9.2__py3-none-any.whl → 2.9.4__py3-none-any.whl

coderouter/launcher_swap.py

@@ -85,9 +85,18 @@ class _ModelState:
 def _pick_ephemeral_port() -> int:
     """Ask the OS for a free port. Best-effort only (§6.6 known-trap #4):
 
-    there is an unavoidable TOCTOU window between this call returning
-    and the child actually binding the port. Fixed ``port:`` in the
-    catalog entry avoids the race entirely (§10 Q2, recommended).
+    there is an unavoidable TOCTOU window between this call's bind+close
+    returning a "free" port and the child actually binding that same
+    port a moment later — nothing stops another process (or another
+    concurrent swap spawn) from grabbing it first. Retrying on a fresh
+    port when the child fails to come up
+    (``LauncherSwapConfig.port_retry_attempts``, default 2 additional
+    attempts) narrows the *impact* of losing that race but does NOT
+    close the window itself — a genuinely race-free allocation would
+    need the OS to hand the child an already-bound socket fd, which the
+    current spawn path (argv ``--port N``) does not support. Fixed
+    ``port:`` in the catalog entry avoids the race entirely (§10 Q2,
+    recommended) and remains the only way to eliminate it outright.
     """
     with socket.socket(socket.AF_INET, socket.SOCK_STREAM) as s:
         s.bind(("127.0.0.1", 0))
@@ -126,6 +135,13 @@ class SwapManager:
         self._readiness_timeout_s = config.readiness_timeout_s
         self._sweep_interval_s = config.sweep_interval_s
         self._sweeper_task: asyncio.Task[None] | None = None
+        # [Unreleased] per-model TTL override: the sweeper must run
+        # whenever EITHER the global TTL is enabled OR at least one
+        # catalog entry sets its own ``ttl_seconds`` (even if the global
+        # value is None/disabled) — see _effective_ttl / start().
+        self._any_ttl_configured = self._ttl_seconds is not None or any(
+            m.ttl_seconds is not None for m in config.models
+        )
 
     # ------------------------------------------------------------------
     # Catalog matching
@@ -214,24 +230,40 @@ class SwapManager:
         async with state.lock:
             await self._unload_locked(state, spec, reason)
 
+    def _effective_ttl(self, spec: SwapModelSpec) -> float | None:
+        """``spec.ttl_seconds`` if set, else the global ``self._ttl_seconds``.
+
+        [Unreleased] per-model TTL override (docs/backends/launcher.md
+        "launcher.swap fields" table). ``None`` at either level keeps
+        its usual meaning ("TTL disabled") — a spec-level override only
+        takes effect when it's not ``None``.
+        """
+        return spec.ttl_seconds if spec.ttl_seconds is not None else self._ttl_seconds
+
     async def sweep_once(self) -> None:
         """One TTL sweep pass: stop every idle, un-leased, expired model.
 
         Takes each model's own lock independently (never more than one
         per-model lock at a time — §6.6 known-trap #7) so a slow unload
-        for model A never delays the sweep of model B.
+        for model A never delays the sweep of model B. Each model's TTL
+        is resolved individually via :meth:`_effective_ttl` — a model
+        whose global TTL is disabled can still expire on its own
+        override, and vice versa.
         """
-        if self._ttl_seconds is None:
+        if not self._any_ttl_configured:
             return
         now = time.monotonic()
         for spec in list(self._catalog.values()):
+            ttl = self._effective_ttl(spec)
+            if ttl is None:
+                continue
             state = self._states.get(spec.name)
             if state is None:
                 continue
             async with state.lock:
                 if state.status != "ready" or state.in_flight > 0:
                     continue
-                if (now - state.last_used) < self._ttl_seconds:
+                if (now - state.last_used) < ttl:
                     continue
                 await self._unload_locked(state, spec, reason="ttl")
 
@@ -240,8 +272,14 @@ class SwapManager:
     # ------------------------------------------------------------------
 
     async def start(self) -> None:
-        """Start the TTL sweeper background task. No-op when TTL is disabled."""
-        if self._ttl_seconds is None or self._sweeper_task is not None:
+        """Start the TTL sweeper background task.
+
+        No-op when TTL is disabled both globally AND on every catalog
+        entry (``self._any_ttl_configured`` — [Unreleased] per-model
+        override means a per-model ``ttl_seconds`` alone is enough to
+        need the sweeper, even with the global TTL off).
+        """
+        if not self._any_ttl_configured or self._sweeper_task is not None:
             return
         self._sweeper_task = asyncio.create_task(self._sweeper_loop())
 
@@ -285,6 +323,35 @@ class SwapManager:
     def _engine(self) -> Any | None:
         return getattr(self._app.state, "engine", None)
 
+    def _remove_from_registry(self, proc_id: str) -> None:
+        """Drop a swap-managed ``ManagedProcess`` from the launcher registry.
+
+        [Unreleased] registry-litter fix: ``stop_process`` only sets
+        ``status="stopped"`` — it never removes the entry, because for
+        MANUALLY started processes the stopped row doubles as visible
+        history in the /launcher UI (with logs and an explicit ✕ delete
+        button). Swap-managed processes have no such contract: every
+        failed readiness attempt (times ``1 + port_retry_attempts``) and
+        every TTL unload would otherwise leave a permanent "stopped" row
+        accumulating unboundedly in ``GET /api/launcher/processes``. So
+        SwapManager removes ITS OWN processes right after it stops them,
+        guarded on ``swap_managed`` so a manual process can never be
+        swept up by mistake. Crash leftovers (a swap process that died
+        on its own, status "error"/"stopped" without SwapManager
+        stopping it) are deliberately NOT removed here — their log tail
+        is the only crash forensics an operator has; the ✕ button still
+        applies.
+        """
+        reg = _registry_for_app(self._app)
+        try:
+            proc = reg.get(proc_id)
+        except KeyError:
+            return
+        if not proc.swap_managed:
+            return
+        with contextlib.suppress(KeyError):
+            reg.remove(proc_id)
+
     def _resolve_options(self, spec: SwapModelSpec) -> dict[str, Any]:
         """Resolve ``spec.option_profile`` into the launcher options dict."""
         if not spec.option_profile or self._launcher_cfg is None:
@@ -346,11 +413,20 @@ class SwapManager:
     async def _spawn_with_retry(
         self, spec: SwapModelSpec
     ) -> tuple[ManagedProcess | None, Exception | None]:
-        """Spawn + wait for readiness. §10 Q2: one retry on a fresh ephemeral
-        port when ``spec.port`` is unset and the first attempt doesn't come
-        up (fixed ports never retry — a second attempt would collide again).
+        """Spawn + wait for readiness.
+
+        §10 Q2: when ``spec.port`` is unset, up to
+        ``1 + LauncherSwapConfig.port_retry_attempts`` attempts total
+        (default: 1 initial + 2 retries = 3), each on a freshly picked
+        ephemeral port via :func:`_pick_ephemeral_port` — see that
+        function's docstring for the residual pick-then-bind TOCTOU
+        window this does NOT close, only bounds the impact of. Fixed
+        ports never retry — a second attempt on the same port would
+        just collide again.
         """
-        attempts = 1 if spec.port is not None else 2
+        attempts = (
+            1 if spec.port is not None else 1 + self._config.port_retry_attempts
+        )
         last_exc: Exception | None = None
         for _attempt in range(attempts):
             port = spec.port if spec.port is not None else _pick_ephemeral_port()
@@ -368,6 +444,10 @@ class SwapManager:
             )
             with contextlib.suppress(Exception):
                 await stop_process(self._app, proc.id)
+            # [Unreleased] registry-litter fix: without this, every failed
+            # readiness attempt leaves one "stopped" row in the registry
+            # forever (1 + port_retry_attempts rows per failed load).
+            self._remove_from_registry(proc.id)
         return None, last_exc
 
     async def _spawn(self, spec: SwapModelSpec, port: int) -> ManagedProcess:
@@ -394,6 +474,10 @@ class SwapManager:
             # provider). H-2: exempt from launcher auto-restart (crash
             # recovery = next-request re-spawn under the per-model lock).
             swap_managed=True,
+            # [Unreleased]: catalog model name, surfaced by
+            # GET /api/launcher/processes as "swap_model" so the /launcher
+            # UI can show which swap catalog entry a process backs.
+            swap_model=spec.name,
         )
 
     async def _await_ready(self, proc: ManagedProcess) -> bool:
@@ -457,6 +541,10 @@ class SwapManager:
             with contextlib.suppress(Exception):
                 await stop_process(self._app, proc_id)
             await self._deregister(spec)
+            # [Unreleased] registry-litter fix: a TTL-unloaded swap process
+            # is not "history" the way a manually stopped one is — leaving
+            # it would grow the registry by one row per load/unload cycle.
+            self._remove_from_registry(proc_id)
         logger.info(
             "swap-unload",
             extra={"model": spec.name, "reason": reason, "proc_id": proc_id},

coderouter/translation/anthropic.py

@@ -143,9 +143,18 @@ def normalize_message_roles(payload: dict[str, Any]) -> dict[str, Any]:
     (see anthropics/claude-code#63469, vllm-project/vllm#44000).
 
     Policy:
-        - ``role: "system"`` → text content merged into the top-level
-          ``system`` field (appended after any existing system prompt;
-          same join rule as ``convert.to_anthropic_request``).
+        - leading ``role: "system"`` (before any user/assistant turn) →
+          text content merged into the top-level ``system`` field
+          (appended after any existing system prompt; same join rule as
+          ``convert.to_anthropic_request``). This is a genuine system
+          prompt from an OpenAI-style client, and it stays at position 0.
+        - mid-conversation ``role: "system"`` → coerced to ``user`` **in
+          place**, like any other non-spec role. Claude Code's
+          system-reminders are volatile per-turn text; hoisting them to
+          the top-level field moves content that changes every turn ahead
+          of the whole conversation, which invalidates the prefix cache of
+          local backends (llama.cpp / LM Studio) and forces a full prompt
+          reprocess on every request.
         - any other non-spec role (``ctx``, ``msg``, ...) → coerced to
           ``user`` so conversation position is preserved. Anthropic
           merges consecutive same-role turns, so this is safe.
@@ -175,9 +184,21 @@ def normalize_message_roles(payload: dict[str, Any]) -> dict[str, Any]:
             continue
         if role == "system":
             text = _content_as_text(msg.get("content"))
-            if text:
-                system_texts.append(text)
             coerced_roles.append("system")
+            if text:
+                if not messages_out:
+                    # Leading system message — a genuine system prompt from an
+                    # OpenAI-style client. Hoisting to the top-level field is
+                    # correct and prefix-safe (it stays at position 0).
+                    system_texts.append(text)
+                else:
+                    # Mid-conversation system message — volatile per-turn text
+                    # (Claude Code >= 2.1.154 system-reminders). Keep it where
+                    # the client put it. Hoisting moves text that changes every
+                    # turn ahead of the entire conversation, which invalidates
+                    # the prefix cache of local backends (llama.cpp / LM Studio)
+                    # and forces a full prompt reprocess on every request.
+                    messages_out.append({"role": "user", "content": text})
             continue
         # Unknown role (ctx / msg / future surprises): keep its position
         # in the conversation as a user turn; drop if nothing salvageable.

{coderouter_cli-2.9.2.dist-info → coderouter_cli-2.9.4.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: coderouter-cli
-Version: 2.9.2
+Version: 2.9.4
 Summary: Local-first, free-first, fallback-built-in LLM router. Claude Code / OpenAI compatible.
 Project-URL: Homepage, https://github.com/zephel01/CodeRouter
 Project-URL: Repository, https://github.com/zephel01/CodeRouter
@@ -40,7 +40,7 @@ Requires-Dist: types-pyyaml>=6.0.12; extra == 'dev'
 Provides-Extra: doctor
 Requires-Dist: ruamel-yaml>=0.18.6; extra == 'doctor'
 Provides-Extra: repair
-Requires-Dist: json-repair>=0.30; extra == 'repair'
+Requires-Dist: json-repair>=0.60.1; extra == 'repair'
 Description-Content-Type: text/markdown
 
 <h1 align="center">CodeRouter</h1>

{coderouter_cli-2.9.2.dist-info → coderouter_cli-2.9.4.dist-info}/RECORD

@@ -10,8 +10,9 @@ coderouter/errors.py,sha256=Xmq67lheyw8iv3Ox39jh2c4tvNI5RcUR4QkoxVDN6l4,1130
 coderouter/gguf_introspect.py,sha256=KeE00CfbYRa4gSTrzuwC6zVuHi20IzLC4nUQa98BEXI,10387
 coderouter/hardware.py,sha256=gn3_9qbVcGRR81yKMn1lJE_8-YDRau0LxIH_M-f7pxE,8356
 coderouter/language_tax.py,sha256=LTbE3tIfoJuV2O3T0NixRKhzq_dEOTUuPEerJv2q9uk,9360
+coderouter/launcher_devices.py,sha256=SoGpzknRqvnmvg_IeXy1TkeE7aLc_MOFqLzXdd3CjVs,19275
 coderouter/launcher_speculative.py,sha256=kWaHNmhzBsbWuM_lRGDlbWkgJ1suuxBWQNTKhJOC-yg,12644
-coderouter/launcher_swap.py,sha256=Oc9itCKzjSFbE_RWxqcjIJyiYax8o9zZo7czmEI2s-U,19150
+coderouter/launcher_swap.py,sha256=W7qpdB-8Npr8yHGfx0N3SgZ2vDJ6BR9JuxW0I3TaZ-4,23783
 coderouter/logging.py,sha256=5rfi4_R4aoeeUYDc642dgs61wA0aBX7wvubyB-k3Uyo,58074
 coderouter/output_filters.py,sha256=0ry_rPiS_kC-FnHgaNVP6v7e6Al2djxzu9vBzZ8kEkE,25314
 coderouter/token_estimation.py,sha256=iz22vZEEW2P7uKLB2pYvPNpIbZGbgXRO5MtfkS_-9Sk,7531
@@ -25,7 +26,7 @@ coderouter/config/__init__.py,sha256=FODEn74fN-qZnt4INPSHswqhOlEgpL6-_onxsitSx8g
 coderouter/config/capability_registry.py,sha256=RZgzHF54dXy0CCDocRCz2ee21qXGbbgTSKTkRI9iLL4,16936
 coderouter/config/env_file.py,sha256=CoMK27fuAXm-NtoLzXb8yN2E-wDFjHQuFwiIlmgTBQw,10356
 coderouter/config/loader.py,sha256=FUEe8m4Tnmj_aul0vSctD8vKvNW-oLRoMRbTpSKqSmc,4077
-coderouter/config/schemas.py,sha256=zMv6UjL2ebcxc3dMIOa9lQjOGvvuHMWCPqoOVA35CEw,117181
+coderouter/config/schemas.py,sha256=wHaOqbehppjtL-01eLh8DNmm0A3rZXdJ-cNbBcS2yM8,120984
 coderouter/data/__init__.py,sha256=uNyfD9jaCvTWsBAWtaw1Fr25OSxzv3psGMfBjT1z0Cc,328
 coderouter/data/model-capabilities.yaml,sha256=S9jt6SC6-3s2-icZ_n-a14iEMnc2yB1C2R6q-N_tZWQ,19309
 coderouter/guards/__init__.py,sha256=5qliYBqygvVPneej7nx0uSjxDKsz7t8VzvrDgVBJlvU,1170
@@ -43,7 +44,7 @@ coderouter/ingress/__init__.py,sha256=WQsCH2CGJCAhy0mS6GSEdeYZRkkQu2OHDsP4CJWTLu
 coderouter/ingress/anthropic_routes.py,sha256=1ThXzx8VgxRV7UCi3NF5ey7kbr0ADPKTXyQuYS5gK1w,27063
 coderouter/ingress/app.py,sha256=6e_qNlYZIyjLKhY20t0ZNL-x5Fv8R8oaIH_HCdxa9Ic,20921
 coderouter/ingress/dashboard_routes.py,sha256=1SqNNrVTfQiZcx5n2MvjAzWirittX0e61tCt-MKp4Ag,25617
-coderouter/ingress/launcher_routes.py,sha256=kVorBfnWL9Q0a6h-gad2YxgeoZjHFHbL8C4DcHtkigg,82031
+coderouter/ingress/launcher_routes.py,sha256=lU2EILVAi-lvHBHIxyixsMTRZlp-Rv7-2sw-eW4ccYg,119687
 coderouter/ingress/metrics_routes.py,sha256=M22dwOGn24P05Ge4W3c7d7mYytSGWjIR-pPSPOAiHJY,3965
 coderouter/ingress/openai_routes.py,sha256=TrugsQNJfNPgKKLzfT1aXzrs4emWYGE4XebEBuaZoWk,12505
 coderouter/metrics/__init__.py,sha256=7Es351DPS7yLM0yVF_F0eesmiD83n7Zzhie44chht38,1465
@@ -66,11 +67,11 @@ coderouter/state/request_log.py,sha256=mNxQJ9IDGpUtT5rZbCo_-CTyhxmvig7C2Gfvdn9rc
 coderouter/state/store.py,sha256=h-rsMJq8GILsOfCP94nI40cuHaj4Vqycsm9UNN77REI,7445
 coderouter/state/suggest_rules.py,sha256=FvdhEvao5NvdKp9zs8AkcoFKHY4yqqXY2HekvSjpDFA,16670
 coderouter/translation/__init__.py,sha256=PYXN7XVEwpG1uC8RLy6fvnGbzEZhhrEuUapH8IYOtG8,1788
-coderouter/translation/anthropic.py,sha256=aZkcYH4x82b0x7efJgJb9RWn9Hbyc9pEOthXe4vjUdU,11113
+coderouter/translation/anthropic.py,sha256=uEScq-icLn6ZXFWaxsKROv7feaq3zD0ZogjbTJ3Y7gQ,12578
 coderouter/translation/convert.py,sha256=VV4kpu4dvLmGBmJfQnCLY5ryy_AnQcV2NuwjqRtaR8Q,52633
 coderouter/translation/tool_repair.py,sha256=DGOKArM53yfVbIfiAveoul1vGq9NEKmjagf43Qh4IkM,57630
-coderouter_cli-2.9.2.dist-info/METADATA,sha256=dcDACrTxRsla8y8mcY05IC4KilzZD-co6xkEz9xC_H4,15994
-coderouter_cli-2.9.2.dist-info/WHEEL,sha256=lCkmxWfQsSc9CfIClYeavTdQeEX2toPqufh9gI35EQA,87
-coderouter_cli-2.9.2.dist-info/entry_points.txt,sha256=-dnLfD1YZ2WjH2zSdNCvlO65wYltM9bsHt9Fhg3yGss,51
-coderouter_cli-2.9.2.dist-info/licenses/LICENSE,sha256=wkEzoR86jFw33jvfOHjULqmkGEfxTFMgMaJnpR8mPRw,1065
-coderouter_cli-2.9.2.dist-info/RECORD,,
+coderouter_cli-2.9.4.dist-info/METADATA,sha256=E7EiSxxEYfkchQLvqlvdQzADFsnwbReGwi0Patjfk8I,15996
+coderouter_cli-2.9.4.dist-info/WHEEL,sha256=lCkmxWfQsSc9CfIClYeavTdQeEX2toPqufh9gI35EQA,87
+coderouter_cli-2.9.4.dist-info/entry_points.txt,sha256=-dnLfD1YZ2WjH2zSdNCvlO65wYltM9bsHt9Fhg3yGss,51
+coderouter_cli-2.9.4.dist-info/licenses/LICENSE,sha256=wkEzoR86jFw33jvfOHjULqmkGEfxTFMgMaJnpR8mPRw,1065
+coderouter_cli-2.9.4.dist-info/RECORD,,