programasweights 0.2.3__tar.gz → 0.2.5__tar.gz

{programasweights-0.2.3 → programasweights-0.2.5}/AGENTS.md

@@ -81,6 +81,21 @@ Output: delete
 - Each function is stateless: one input, one output. No conversation history.
 - Write a few test inputs with expected outputs, then try different spec phrasings and pick the one that passes the most.
 
+## Chaining Functions
+
+Multiple PAW functions can be composed for multi-step tasks:
+
+```python
+classifier = paw.compile_and_load("Classify the bug type. Return ONLY one of: off-by-one, type-error, other")
+fixer = paw.compile_and_load("Fix the bug described in the first line. Return only the corrected code.")
+
+label = classifier(code_snippet)
+if label != "other":
+    fix = fixer(f"{label}: {code_snippet}")
+```
+
+Each function is independent -- chain them with regular Python logic.
+
 ## Browser / JavaScript SDK
 
 Programs compiled with `paw-4b-gpt2` run in the browser via WebAssembly.
@@ -143,7 +158,7 @@ fn = paw.function(program)              # accepts Program object, hash ID, or sl
 fn = paw.function("a6b454023d41ac9ca845")
 fn = paw.function("da03/my-classifier")
 
-result: str = fn(input_text: str, max_tokens=512, temperature=0.0)
+result: str = fn(input_text: str, max_tokens=None, temperature=0.0)
 
 fn = paw.compile_and_load(spec, compiler="paw-4b-qwen3-0.6b")
 
@@ -156,7 +171,7 @@ paw.login()
 
 | Error | Cause | Fix |
 |-------|-------|-----|
-| `httpx.HTTPStatusError: 404` on download | Program still uploading to CDN | SDK retries automatically. If persistent, recompile. |
+| `RuntimeError: assets not ready` on download | Program still generating after compile | SDK polls automatically for up to 30s. If persistent, recompile. |
 | `httpx.HTTPStatusError: 422` on compile | Spec too short (<10 chars) | Adjust spec length. |
 | `httpx.HTTPStatusError: 429` | Rate limit exceeded | Wait, or sign in for higher limits. |
 
@@ -168,6 +183,11 @@ paw.login()
 - **Cache**: `~/.cache/programasweights/`. Override with `PAW_CACHE_DIR`.
 - **Offline** after first download.
 
+## Limits
+
+- Spec + input + output share a ~2048 token context window, and inputs that exceed the context window will error out (not silently truncated).
+- Setting `max_tokens` high is safe -- generation stops at EOS or when the window is full.
+
 ## Browse Programs
 
 https://programasweights.com/hub

{programasweights-0.2.3 → programasweights-0.2.5}/CHANGELOG.md

@@ -1,5 +1,16 @@
 # Changelog
 
+## 0.2.4 (2026-04-01)
+
+- Download reliability: asset endpoints return 202 Retry-After when program is still generating, eliminating 404 race conditions
+- Guard local cleanup on HF upload success: files persist locally if upload fails
+- Stderr suppression now covers LoRA adapter loading (fixes CPU_REPACK warnings on 0.2.3)
+- Browser: fix LoRA switching bug (free old adapter + clear KV before loading new program)
+
+## 0.2.3 (2026-04-01)
+
+- Python 3.9 compatibility fix (`from __future__ import annotations`)
+
 ## 0.2.2 (2026-03-19)
 
 - Add `paw.compile_and_load()` convenience method (compile + load in one call)

{programasweights-0.2.3 → programasweights-0.2.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: programasweights
-Version: 0.2.3
+Version: 0.2.5
 Summary: Compile natural language specifications into neural programs that run locally via llama.cpp.
 Project-URL: Homepage, https://programasweights.com
 Project-URL: Repository, https://github.com/programasweights/programasweights-python

{programasweights-0.2.3 → programasweights-0.2.5}/docs/api-reference/python-sdk.md

@@ -32,15 +32,17 @@ Loads a compiled program and returns a callable. Downloads the program and base
 The returned callable:
 
 ```python
-output: str = fn(input_text, max_tokens=512, temperature=0.0)
+output: str = fn(input_text, max_tokens=None, temperature=0.0)
 ```
 
 | Parameter | Description |
 |-----------|-------------|
 | `input_text` | Input string for the program. |
-| `max_tokens` | Maximum tokens to generate (default `512`). |
+| `max_tokens` | Maximum tokens to generate. `None` (default) = use all remaining context window. |
 | `temperature` | Sampling temperature (default `0.0`). |
 
+**Context limits:** Spec + input + output share a ~2048 token window. If input + output together exceed the window, generation will error. For short inputs, high `max_tokens` is safe (stops at EOS). For long inputs, set `max_tokens` conservatively.
+
 ## `paw.compile`
 
 ```python

{programasweights-0.2.3 → programasweights-0.2.5}/docs/api-reference/rest-api.md

@@ -34,7 +34,11 @@ Compile a specification.
 | `name` | string | No | Display title. Auto-generated if omitted. |
 | `tags` | string[] | No | Tags for discovery. |
 
-**Response (JSON):** includes `job_id`, `status`, `program_id`, `slug` (if created), `pseudo_program`, and `timings`.
+**Response:**
+- **202 Accepted** on success — JSON with `job_id`, `status`, `program_id`, `slug` (if created), `pseudo_program`, and `timings`
+- **400** — unknown compiler
+- **422** — spec too short, too long, or token limit exceeded
+- **500** — compilation failed (server error)
 
 ### `POST /infer`
 
@@ -88,7 +92,11 @@ Owner-only fields: `public`, `name`, `tags`. Any authenticated user can set thei
 
 ### `GET /programs/{id_or_slug}/download`
 
-Redirects to the Hugging Face CDN for the `.paw` artifact. Private programs return 404 to non-owners.
+Downloads the `.paw` artifact. Returns one of:
+- **302** redirect to HuggingFace CDN (program already uploaded)
+- **200** file from server (freshly compiled, not yet on HF)
+- **202 Accepted** with `Retry-After` header (program assets still generating, retry after the indicated seconds)
+- **404** if program not found or private and not owned by the requester
 
 ### `GET /programs/resolve/{slug}`
 

{programasweights-0.2.3 → programasweights-0.2.5}/programasweights/__init__.py

@@ -23,7 +23,7 @@ API reference:
 """
 from __future__ import annotations
 
-__version__ = "0.2.3"
+__version__ = "0.2.5"
 
 from .config import get_api_url, get_api_key, set_api_key
 

{programasweights-0.2.3 → programasweights-0.2.5}/programasweights/client.py

@@ -108,25 +108,38 @@ class PAWClient:
         """Download a .paw bundle to the local cache.
 
         Returns the path to the extracted program directory.
-        Retries on 404 since freshly compiled programs may still be uploading.
+        Handles 202 Accepted (assets still generating) and 302 redirects (HF CDN).
         """
         program_dir = config.get_programs_dir() / program_id
         if (program_dir / "prompt_template.txt").exists():
             return program_dir
 
-        max_retries = 10
-        for attempt in range(max_retries):
+        max_wait = 30
+        elapsed = 0
+        resp = None
+        while elapsed < max_wait:
             resp = httpx.get(
                 f"{self._api_url}/api/v1/programs/{program_id}/download",
                 headers=self._headers(),
                 timeout=60.0,
                 follow_redirects=True,
             )
-            if resp.status_code == 404 and attempt < max_retries - 1:
+            if resp.status_code == 202:
+                retry_after = int(resp.headers.get("Retry-After", "3"))
+                time.sleep(retry_after)
+                elapsed += retry_after
+                continue
+            if resp.status_code == 404 and elapsed < max_wait - 3:
                 time.sleep(3)
+                elapsed += 3
                 continue
             resp.raise_for_status()
             break
+        else:
+            raise RuntimeError(
+                f"Program {program_id} assets not ready after {max_wait}s. "
+                "The program may still be generating. Try again shortly."
+            )
 
         program_dir.mkdir(parents=True, exist_ok=True)
         paw_path = program_dir / f"{program_id}.paw"

{programasweights-0.2.3 → programasweights-0.2.5}/programasweights/runtime_llamacpp.py

@@ -37,6 +37,7 @@ class PawFunction:
         program_dir = Path(program_dir)
         self._program_dir = program_dir
         self._verbose = verbose
+        self._n_ctx = n_ctx
 
         meta_path = program_dir / "meta.json"
         if meta_path.exists():
@@ -67,24 +68,24 @@ class PawFunction:
                 n_gpu_layers=n_gpu_layers,
                 verbose=verbose,
             )
+
+            adapter_path = program_dir / "adapter.gguf"
+            if adapter_path.exists():
+                self._adapter = llama_cpp.llama_adapter_lora_init(
+                    self._llm.model, str(adapter_path).encode("utf-8"),
+                )
+                if self._adapter:
+                    llama_cpp.llama_set_adapter_lora(self._llm.ctx, self._adapter, 1.0)
+                else:
+                    raise RuntimeError(f"Failed to load LoRA adapter: {adapter_path}")
+            else:
+                self._adapter = None
         finally:
             if not verbose:
                 _os.dup2(_old_stderr, _stderr_fd)
                 _os.close(_devnull)
                 _os.close(_old_stderr)
 
-        adapter_path = program_dir / "adapter.gguf"
-        if adapter_path.exists():
-            self._adapter = llama_cpp.llama_adapter_lora_init(
-                self._llm.model, str(adapter_path).encode("utf-8"),
-            )
-            if self._adapter:
-                llama_cpp.llama_set_adapter_lora(self._llm.ctx, self._adapter, 1.0)
-            else:
-                raise RuntimeError(f"Failed to load LoRA adapter: {adapter_path}")
-        else:
-            self._adapter = None
-
         placeholder = "{INPUT_PLACEHOLDER}"
         self._use_special = interpreter not in ("gpt2",)
 
@@ -151,20 +152,19 @@ class PawFunction:
     def __call__(
         self,
         input_text: str,
-        max_tokens: int = 512,
+        max_tokens: int | None = None,
         temperature: float = 0.0,
     ) -> str:
         """Run the program on an input.
 
         Args:
             input_text: The input to process.
-            max_tokens: Maximum output tokens.
+            max_tokens: Maximum output tokens. None = use all remaining context.
             temperature: Sampling temperature (0 = greedy).
 
         Returns:
             The program's output as a string.
         """
-        # Reset to prefix state: clear everything after the prefix
         self._llm.n_tokens = self._n_prefix
 
         input_with_suffix = input_text + self._suffix_text
@@ -174,10 +174,21 @@ class PawFunction:
             special=self._use_special,
         )
 
+        tokens_used = self._n_prefix + len(input_tokens)
+        remaining = self._n_ctx - tokens_used
+        if remaining <= 0:
+            raise ValueError(
+                f"Input too long: {tokens_used} tokens used (prefix={self._n_prefix}, "
+                f"input={len(input_tokens)}), context window={self._n_ctx}. "
+                f"Shorten your input."
+            )
+
+        gen_limit = remaining if max_tokens is None else min(max_tokens, remaining)
+
         self._llm.eval(input_tokens)
 
         output_tokens = []
-        for _ in range(max_tokens):
+        for _ in range(gen_limit):
             token = self._llm.sample(
                 temp=temperature if temperature > 0 else 0,
             )

{programasweights-0.2.3 → programasweights-0.2.5}/pyproject.toml

@@ -4,7 +4,7 @@ build-backend = "hatchling.build"
 
 [project]
 name = "programasweights"
-version = "0.2.3"
+version = "0.2.5"
 description = "Compile natural language specifications into neural programs that run locally via llama.cpp."
 readme = "PYPI_README.md"
 requires-python = ">=3.9"

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/main.py

@@ -57,6 +57,9 @@ class _FallbackLimiter:
         return True
 
 
+_API_VERSION = "0.4.0"
+
+
 def create_app() -> FastAPI:
     settings = get_settings()
     configure_logging(
@@ -64,10 +67,13 @@ def create_app() -> FastAPI:
         json_format=(settings.LOG_FORMAT == "json"),
     )
 
+    if settings.JWT_SECRET == "dev-secret-change-in-production":
+        raise RuntimeError("JWT_SECRET must be set in .env -- refusing to start with the default value")
+
     app = FastAPI(
         title="Programs as Weights API",
         description="Compile natural language specs into neural functions",
-        version="0.4.0",
+        version=_API_VERSION,
         docs_url="/api/docs",
         openapi_url="/api/openapi.json",
     )

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/routes/auth.py

@@ -21,10 +21,20 @@ from api.models.orm import User
 logger = structlog.get_logger()
 router = APIRouter()
 
-# In-memory session store (replace with Redis/DB for production)
+# In-memory session store (single-process; cleaned periodically)
 _sessions: dict[str, dict] = {}
 _github_states: dict[str, float] = {}
 
+_SESSION_MAX_AGE = 30 * 24 * 3600  # 30 days, matches cookie max_age
+
+
+def _cleanup_expired_sessions():
+    cutoff = time.time() - _SESSION_MAX_AGE
+    expired = [k for k, v in _sessions.items() if v.get("created_at", 0) < cutoff]
+    for k in expired:
+        del _sessions[k]
+
+
 GITHUB_AUTHORIZE_URL = "https://github.com/login/oauth/authorize"
 GITHUB_TOKEN_URL = "https://github.com/login/oauth/access_token"
 GITHUB_USER_URL = "https://api.github.com/user"
@@ -44,6 +54,8 @@ async def github_login(request: Request):
     for k in [k for k, v in _github_states.items() if v < cutoff]:
         del _github_states[k]
 
+    _cleanup_expired_sessions()
+
     params = (
         f"?client_id={settings.GITHUB_CLIENT_ID}"
         f"&redirect_uri=https://programasweights.com/api/v1/auth/github/callback"

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/routes/compile.py

@@ -2,6 +2,7 @@
 Compile endpoint: submit specs for compilation into neural programs.
 """
 
+import time
 import uuid
 
 import httpx
@@ -19,8 +20,18 @@ logger = structlog.get_logger()
 
 router = APIRouter()
 
-# In-memory job store (will be replaced with PostgreSQL + Celery)
+# In-memory job store (single-process; evicted after 1 hour)
 _jobs: dict[str, dict] = {}
+_JOBS_MAX_AGE = 3600
+
+
+def _evict_old_jobs():
+    import time as _time
+    cutoff = _time.time() - _JOBS_MAX_AGE
+    expired = [k for k, v in _jobs.items() if v.get("created_at", 0) < cutoff]
+    for k in expired:
+        del _jobs[k]
+
 
 # Compiler registry loaded from JSON file
 import json as _json
@@ -87,6 +98,8 @@ async def submit_compile(
             },
         )
 
+    _evict_old_jobs()
+
     job_id = str(uuid.uuid4())[:12]
     _jobs[job_id] = {
         "status": "queued",
@@ -94,6 +107,7 @@ async def submit_compile(
         "compiler_snapshot": snapshot,
         "program_id": None,
         "error": None,
+        "created_at": time.time(),
     }
 
     import asyncio
@@ -140,7 +154,7 @@ async def submit_compile(
                 public=req.public,
                 name=req.name,
                 tags=req.tags,
-                hf_url=f"https://huggingface.co/programasweights/paw-programs/resolve/main/{pid}.paw",
+                hf_url=None,
             ))
 
         db.add(CompileLog(
@@ -244,12 +258,11 @@ async def submit_compile(
         _jobs[job_id]["error"] = str(e)
         logger.exception("compile_failed", spec_len=len(req.spec), error=str(e))
 
-        return CompileResponse(
-            job_id=job_id,
-            status="failed",
-            compiler_snapshot=snapshot,
-            error=str(e),
-        )
+        raise HTTPException(status_code=500, detail={
+            "error": "compile_failed",
+            "message": "Compilation failed. Please try again.",
+            "request_id": getattr(request.state, "request_id", "unknown"),
+        })
 
 
 @router.get("/compile/{job_id}", response_model=CompileStatusResponse)

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/routes/health.py

@@ -8,6 +8,8 @@ from fastapi import APIRouter, Request
 
 from api.models.schemas import HealthResponse
 
+_API_VERSION = "0.4.0"
+
 router = APIRouter()
 
 _start_time = time.time()
@@ -21,7 +23,7 @@ async def health_check(request: Request):
 
     return HealthResponse(
         status="healthy" if not warnings else "degraded",
-        version="0.2.0",
+        version=_API_VERSION,
         gpu_services={},
         queue_depth=0,
         uptime_s=time.time() - _start_time,

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/routes/infer.py

@@ -107,9 +107,9 @@ async def run_inference(
     except HTTPException:
         raise
     except Exception as e:
-        logger.exception("infer_proxy_failed", program_id=canonical_id)
+        logger.exception("infer_proxy_failed", program_id=canonical_id, error=str(e))
         raise HTTPException(status_code=502, detail={
             "error": "inference_failed",
-            "message": str(e)[:200],
+            "message": "Inference failed. Please try again.",
             "request_id": request_id,
         })

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/routes/programs.py

@@ -6,7 +6,7 @@ import re
 from pathlib import Path
 
 from fastapi import APIRouter, Depends, HTTPException, Request
-from fastapi.responses import FileResponse, RedirectResponse
+from fastapi.responses import FileResponse, JSONResponse, RedirectResponse
 
 from api.config import get_settings
 from sqlalchemy import func, select, or_
@@ -125,6 +125,8 @@ async def list_programs(
 ):
     from sqlalchemy import text as sa_text
 
+    per_page = min(per_page, 100)
+
     if mine and user:
         aliased_ids = select(Alias.program_id).where(Alias.owner_id == user.id).scalar_subquery()
         query = select(Program).where(
@@ -201,7 +203,7 @@ async def list_programs(
                 use_fts = True
             else:
                 alias_match = select(Alias.program_id).where(
-                    Alias.slug.ilike(f"%{search_term}%")
+                    Alias.slug_name.ilike(f"%{search_term}%")
                 ).scalar_subquery()
                 conditions = [
                     Program.spec.ilike(f"%{search_term}%"),
@@ -414,6 +416,10 @@ async def download_program(
     program.downloads = (program.downloads or 0) + 1
     await db.commit()
 
+    if program.hf_url:
+        hf_url = f"https://huggingface.co/programasweights/paw-programs/resolve/main/{program.id}.paw"
+        return RedirectResponse(url=hf_url, status_code=302)
+
     settings = get_settings()
     local_paw = Path(settings.STORAGE_PATH) / program.id / f"{program.id}.paw"
     if local_paw.exists():
@@ -423,8 +429,11 @@ async def download_program(
             filename=f"{program.id}.paw",
         )
 
-    hf_url = f"https://huggingface.co/programasweights/paw-programs/resolve/main/{program.id}.paw"
-    return RedirectResponse(url=hf_url, status_code=302)
+    return JSONResponse(
+        status_code=202,
+        content={"status": "generating", "message": "Program assets are being prepared. Retry shortly."},
+        headers={"Retry-After": "3"},
+    )
 
 
 ALLOWED_ASSETS = {"adapter.gguf", "prompt_template.txt", "prefix_cache.bin", "prefix_tokens.json"}
@@ -443,18 +452,25 @@ async def serve_program_asset(
     if not program:
         raise HTTPException(status_code=404, detail="Asset not found")
 
+    if program.hf_url:
+        hf_url = f"https://huggingface.co/programasweights/paw-programs/resolve/main/{program.id}/{filename}"
+        return RedirectResponse(url=hf_url, status_code=302)
+
     settings = get_settings()
     asset_path = Path(settings.STORAGE_PATH) / program.id / filename
-    if not asset_path.exists():
-        raise HTTPException(status_code=404, detail="Asset not found on this server")
-
-    media_type = "application/octet-stream"
-    if filename.endswith(".txt"):
-        media_type = "text/plain"
-    elif filename.endswith(".json"):
-        media_type = "application/json"
-
-    return FileResponse(path=str(asset_path), media_type=media_type, filename=filename)
+    if asset_path.exists():
+        media_type = "application/octet-stream"
+        if filename.endswith(".txt"):
+            media_type = "text/plain"
+        elif filename.endswith(".json"):
+            media_type = "application/json"
+        return FileResponse(path=str(asset_path), media_type=media_type, filename=filename)
+
+    return JSONResponse(
+        status_code=202,
+        content={"status": "generating", "message": "Program assets are being prepared. Retry shortly."},
+        headers={"Retry-After": "3"},
+    )
 
 
 # --- Update program ---

{programasweights-0.2.3 → programasweights-0.2.5}/server/api/services/compile_provider.py

@@ -509,10 +509,22 @@ class CompileProvider:
                     upload_all_to_hf, program_id, adapter_dir,
                     self.interpreter, paw_path,
                 )
-                logger.info("hf_upload_done", program_id=program_id, hf_url=hf_url)
 
-                await asyncio.sleep(300)
-                cleanup_after_upload(adapter_dir, program_id)
+                if hf_url:
+                    from api.models.database import get_engine
+                    from sqlalchemy import text as sa_text
+                    engine = get_engine()
+                    with engine.connect() as conn:
+                        conn.execute(
+                            sa_text("UPDATE programs SET hf_url = :url WHERE id = :pid"),
+                            {"url": hf_url, "pid": program_id},
+                        )
+                        conn.commit()
+                    logger.info("hf_upload_done", program_id=program_id, hf_url=hf_url)
+                    await asyncio.sleep(300)
+                    cleanup_after_upload(adapter_dir, program_id)
+                else:
+                    logger.warning("hf_upload_failed_keeping_local", program_id=program_id)
             except Exception:
                 logger.exception("background_failed", program_id=program_id)
 

{programasweights-0.2.3 → programasweights-0.2.5}/web-app/frontend/src/App.tsx

@@ -1,4 +1,4 @@
-import { BrowserRouter as Router, Routes, Route, Navigate, useLocation } from 'react-router-dom';
+import { BrowserRouter as Router, Routes, Route, Navigate, useLocation, Link } from 'react-router-dom';
 import { Component, useEffect } from 'react';
 import type { ReactNode } from 'react';
 import { Toaster } from 'react-hot-toast';
@@ -55,6 +55,18 @@ class ErrorBoundary extends Component<{ children: ReactNode }, { hasError: boole
   }
 }
 
+function NotFoundPage() {
+  return (
+    <div className="min-h-[60vh] flex items-center justify-center">
+      <div className="text-center">
+        <h1 className="text-6xl font-bold text-gray-300 mb-4">404</h1>
+        <p className="text-xl text-gray-600 mb-6">Page not found</p>
+        <Link to="/" className="px-5 py-2.5 bg-gray-900 text-white rounded-lg hover:bg-gray-800 transition-colors text-sm font-medium">Back to home</Link>
+      </div>
+    </div>
+  );
+}
+
 function App() {
   return (
     <ErrorBoundary>
@@ -96,6 +108,8 @@ function App() {
             <Route path="/leaderboard" element={<Navigate to="/hub?sort=popular" replace />} />
             <Route path="/program/:id" element={<Navigate to="/hub" replace />} />
             <Route path="/publish" element={<Navigate to="/hub/upload" replace />} />
+
+            <Route path="*" element={<NotFoundPage />} />
           </Routes>
         </main>
         <Footer />