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:
+        If the patent PDF is not already on disk, this method attempts to
-            The patent PDF must already exist at ``patents/{patent_id}.pdf``
+        download it automatically by looking up the PDF link in the database
-            before calling this method. PDFs are downloaded automatically when
+        cache. If the link is not cached either, a ``FileNotFoundError`` is
-            using the batch analysis pipeline (``analyze_company`` or the
+        raised with instructions on how to obtain the PDF.
-            ``/analyze/batch`` API endpoint). For standalone usage, download
-            the PDF manually or call ``SERP.save_patents()`` first.
 
         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"Patent PDF not found at '{patent_path}' and no download link is "
-                f"Download the PDF first using SERP.save_patents() or the batch analysis pipeline."
+                    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

@@ -9,7 +9,7 @@ from typing import Annotated, List
 
 from fastapi import BackgroundTasks, Depends, FastAPI, HTTPException, Query, Request
 from fastapi.middleware.cors import CORSMiddleware
-from fastapi.responses import JSONResponse, StreamingResponse
+from fastapi.responses import JSONResponse
 from pydantic import BaseModel, EmailStr, Field
 from slowapi import Limiter
 from slowapi.errors import RateLimitExceeded
@@ -389,61 +389,6 @@ async def get_analytics(
     )
 
 
-# ============== Export Endpoints ==============
-
-
-@app.get("/export/{company_name}", tags=["Export"])
-async def export_company_csv(
-    company_name: str,
-    _: UserResponse = Depends(get_current_user),
-):
-    """Export analysis results for a company as a CSV file.
-
-    Returns all stored analysis records for the given company, including
-    analysis type, model used, response text, and timestamp.
-
-    Args:
-        company_name: Company name to export results for
-
-    Returns:
-        CSV file download
-    """
-    import csv
-    import io
-
-    db = get_db_client()
-    # Query all non-cached analysis results for this company
-    with db.get_conn() as conn:
-        with conn.cursor() as cur:
-            cur.execute(
-                """
-                SELECT company_name, analysis_type, model, response, timestamp
-                FROM llm_messages
-                WHERE LOWER(company_name) = LOWER(%s) AND is_cached = FALSE
-                ORDER BY timestamp DESC
-                """,
-                (company_name,),
-            )
-            rows = cur.fetchall()
-
-    if not rows:
-        raise HTTPException(status_code=404, detail=f"No analysis results found for '{company_name}'")
-
-    output = io.StringIO()
-    writer = csv.writer(output)
-    writer.writerow(["company_name", "analysis_type", "model", "analysis", "timestamp"])
-    for row in rows:
-        writer.writerow(row)
-
-    output.seek(0)
-    safe_name = company_name.replace(" ", "_").lower()
-    return StreamingResponse(
-        iter([output.getvalue()]),
-        media_type="text/csv",
-        headers={"Content-Disposition": f'attachment; filename="sparc_{safe_name}_export.csv"'},
-    )
-
-
 # ============== System Endpoints ==============
 
 
@@ -484,6 +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,

frontend/src/api/client.ts

@@ -126,23 +126,6 @@ export const analysisApi = {
   },
 };
 
-// Export API
-export const exportApi = {
-  exportCsv: async (companyName: string): Promise<void> => {
-    const response = await api.get(`/export/${encodeURIComponent(companyName)}`, {
-      responseType: 'blob',
-    });
-    const url = window.URL.createObjectURL(new Blob([response.data]));
-    const link = document.createElement('a');
-    link.href = url;
-    link.setAttribute('download', `sparc_${companyName.toLowerCase().replace(/\s+/g, '_')}_export.csv`);
-    document.body.appendChild(link);
-    link.click();
-    link.remove();
-    window.URL.revokeObjectURL(url);
-  },
-};
-
 // Analytics API
 export const analyticsApi = {
   getAnalytics: async (days = 30): Promise<Analytics> => {

frontend/src/pages/Analysis.tsx

@@ -1,7 +1,7 @@
 import { useState } from 'react';
 import { useMutation } from '@tanstack/react-query';
-import { analysisApi, exportApi } from '../api/client';
+import { analysisApi } from '../api/client';
-import { Search, CheckCircle, AlertCircle, Clock, FileText, Download } from 'lucide-react';
+import { Search, CheckCircle, AlertCircle, Clock, FileText } from 'lucide-react';
 import type { CompanyAnalysis } from '../types';
 
 export function Analysis() {
@@ -106,18 +106,9 @@ export function Analysis() {
           {/* Analysis Content */}
           {result.success && result.analysis && (
             <div className="bg-bg-card/60 backdrop-blur-lg border border-primary/15 rounded-2xl p-6">
-              <div className="flex items-center justify-between border-b-2 border-primary/30 pb-2 mb-4">
+              <h3 className="text-lg font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-4">
-                <h3 className="text-lg font-semibold text-text-primary">
                 AI Analysis Results
               </h3>
-                <button
-                  onClick={() => exportApi.exportCsv(result.company_name)}
-                  className="flex items-center gap-2 text-sm bg-primary/20 hover:bg-primary/30 text-primary font-medium px-3 py-1.5 rounded-lg transition-colors"
-                >
-                  <Download size={14} />
-                  Export CSV
-                </button>
-              </div>
               <div className="prose prose-invert max-w-none">
                 <div className="text-text-primary whitespace-pre-wrap leading-relaxed">
                   {result.analysis}