fix: auto-download patent PDF in analyze_single_patent before reading

When the PDF is not on disk, analyze_single_patent now looks up the
cached PDF link from the database and downloads it automatically.
If no link is cached, a clear FileNotFoundError is raised. Also adds
a GET /analyze/patent/{patent_id} API endpoint that exposes this
functionality and returns 404 when the PDF cannot be obtained.

Closes leeworks-agents/SPARC#36

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

SPARC/analyzer.py

@@ -108,12 +108,10 @@ class CompanyAnalyzer:
     def analyze_single_patent(self, patent_id: str, company_name: str) -> str:
         """Analyze a single patent by ID.
 
-        Prerequisite:
-            The patent PDF must already exist at ``patents/{patent_id}.pdf``
-            before calling this method. PDFs are downloaded automatically when
-            using the batch analysis pipeline (``analyze_company`` or the
-            ``/analyze/batch`` API endpoint). For standalone usage, download
-            the PDF manually or call ``SERP.save_patents()`` first.
+        If the patent PDF is not already on disk, this method attempts to
+        download it automatically by looking up the PDF link in the database
+        cache. If the link is not cached either, a ``FileNotFoundError`` is
+        raised with instructions on how to obtain the PDF.
 
         Args:
           patent_id: Publication ID of the patent (e.g. "US-11234567-B2")
@@ -123,7 +121,7 @@ class CompanyAnalyzer:
           Analysis of the specific patent's innovation quality
 
         Raises:
-          FileNotFoundError: If the patent PDF is not found at the expected path.
+          FileNotFoundError: If the patent PDF cannot be found or downloaded.
         """
         import os
         logger.info("Analyzing patent %s for %s...", patent_id, company_name)
@@ -131,9 +129,21 @@ class CompanyAnalyzer:
         patent_path = f"patents/{patent_id}.pdf"
 
         if not os.path.exists(patent_path):
+            # Attempt to download the PDF automatically from cached metadata
+            cached = self.db.get_cached_patent(patent_id)
+            pdf_link = cached.get("pdf_link") if cached else None
+
+            if pdf_link:
+                logger.info("PDF not on disk; downloading %s from cached link", patent_id)
+                patent = SERP.save_patents(
+                    Patent(patent_id=patent_id, pdf_link=pdf_link)
+                )
+                patent_path = patent.pdf_path
+            else:
                 raise FileNotFoundError(
-                f"Patent PDF not found at '{patent_path}'. "
-                f"Download the PDF first using SERP.save_patents() or the batch analysis pipeline."
+                    f"Patent PDF not found at '{patent_path}' and no download link is "
+                    f"cached for '{patent_id}'. Run a company analysis first to populate "
+                    f"the cache, or call SERP.save_patents() with the patent's PDF link."
                 )
 
         try:

SPARC/api.py

@@ -429,6 +429,38 @@ async def analyze_company(
     return _convert_result(result)
 
 
+@app.get(
+    "/analyze/patent/{patent_id}",
+    tags=["Analysis"],
+)
+async def analyze_single_patent(
+    patent_id: str,
+    company_name: str = Query(description="Company name for analysis context"),
+    _: UserResponse = Depends(get_current_user),
+):
+    """Analyze a single patent by its publication ID.
+
+    If the patent PDF is not already cached locally, the system will attempt
+    to download it automatically from a previously cached link. If no link
+    is available, a 404 error is returned.
+
+    Args:
+        patent_id: Patent publication ID (e.g. "US-11234567-B2")
+        company_name: Company name for analysis context
+
+    Returns:
+        Analysis text for the patent
+    """
+    if not _analyzer:
+        raise HTTPException(status_code=503, detail="Analyzer not initialized")
+
+    try:
+        analysis = _analyzer.analyze_single_patent(patent_id, company_name)
+        return {"patent_id": patent_id, "company_name": company_name, "analysis": analysis}
+    except FileNotFoundError as e:
+        raise HTTPException(status_code=404, detail=str(e))
+
+
 @app.post(
     "/analyze/batch",
     response_model=BatchAnalysisResponse,

SPARC/serp_api.py

@@ -1,4 +1,3 @@
-import logging
 import os
 import re
 from datetime import datetime, timedelta
@@ -11,8 +10,6 @@ import serpapi
 from SPARC import config
 from SPARC.types import Patent, Patents
 
-logger = logging.getLogger(__name__)
-
 
 class SERP:
   def query(company: str, days_back: int = None) -> Patents:
@@ -47,7 +44,6 @@ class SERP:
       "tbs": date_filter,
       "api_key": config.api_key,
     }
-    logger.info("Querying Google Patents for '%s' (last %d days)", company, days_back)
     search = serpapi.search(params)
     # Convert results to Patent objects, skipping any without PDF links
     patent_ids = []
@@ -56,10 +52,8 @@ class SERP:
         pdf_link = patent.get("pdf")
         if pdf_link:
             patent_ids.append(Patent(patent_id=patent["publication_number"], pdf_link=pdf_link, summary=None))
-        else:
-            logger.debug("Skipping patent %s (no PDF link)", patent.get("publication_number", "unknown"))
+        # Patents without PDF links are skipped (see docstring for details)
 
-    logger.info("Found %d patents with PDF links for '%s'", len(patent_ids), company)
     return Patents(patents=patent_ids)
 
   def save_patents(patent: Patent) -> Patent:
@@ -76,13 +70,9 @@ class SERP:
     os.makedirs("patents", exist_ok=True)
 
     if not (os.path.exists(pdf_path) and os.path.getsize(pdf_path) > 0):
-      logger.info("Downloading PDF for %s", patent.patent_id)
       response = requests.get(patent.pdf_link)
       with open(pdf_path, "wb") as f:
         f.write(response.content)
-      logger.debug("Saved %d bytes to %s", len(response.content), pdf_path)
-    else:
-      logger.debug("Using cached PDF for %s at %s", patent.patent_id, pdf_path)
 
     patent.pdf_path = pdf_path
     return patent
@@ -100,13 +90,11 @@ class SERP:
       Dictionary containing all extracted sections
     """
 
-    logger.debug("Parsing patent PDF: %s", pdf_path)
     with pdfplumber.open(pdf_path) as pdf:
       # Extract all text
       full_text = ""
       for page in pdf.pages:
         full_text += page.extract_text() + "\n"
-      logger.debug("Extracted text from %d pages (%d chars)", len(pdf.pages), len(full_text))
 
     # Define section patterns (common in patents)
     sections = {