Compare commits

..

1 Commits

Author SHA1 Message Date
agent-company ecc2c37bcd 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>
2026-03-26 10:08:34 +00:00
4 changed files with 65 additions and 61 deletions
+21 -11
View File
@@ -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)
@@ -131,10 +129,22 @@ class CompanyAnalyzer:
patent_path = f"patents/{patent_id}.pdf"
if not os.path.exists(patent_path):
raise FileNotFoundError(
f"Patent PDF not found at '{patent_path}'. "
f"Download the PDF first using SERP.save_patents() or the batch analysis pipeline."
)
# 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}' 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:
sections = SERP.parse_patent_pdf(patent_path)
+32
View File
@@ -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,
+9 -35
View File
@@ -9,38 +9,15 @@ const COLORS = ['#6366f1', '#0ea5e9', '#10b981', '#f59e0b', '#ef4444', '#8b5cf6'
export function AnalyticsPage() {
const [days, setDays] = useState(30);
const { data, isLoading, isError, refetch } = useQuery({
const { data, isLoading, isError } = useQuery({
queryKey: ['analytics', days],
queryFn: () => analyticsApi.getAnalytics(days),
});
if (isLoading) {
return (
<div className="space-y-6">
<div>
<h2 className="text-xl font-semibold text-text-primary border-b-2 border-primary/30 pb-2 mb-2">
Analytics Dashboard
</h2>
<p className="text-text-secondary">Loading analytics data...</p>
</div>
{/* Skeleton cards */}
<div className="grid grid-cols-1 md:grid-cols-3 gap-4">
{[1, 2, 3].map((i) => (
<div key={i} className="bg-gradient-to-br from-primary/10 to-secondary/10 border border-primary/20 rounded-xl p-5 text-center animate-pulse">
<div className="h-9 w-16 bg-primary/20 rounded mx-auto mb-2" />
<div className="h-4 w-24 bg-primary/10 rounded mx-auto" />
</div>
))}
</div>
{/* Skeleton charts */}
<div className="grid grid-cols-1 lg:grid-cols-2 gap-6">
{[1, 2].map((i) => (
<div key={i} className="bg-bg-card/60 border border-primary/15 rounded-2xl p-6 animate-pulse">
<div className="h-5 w-40 bg-primary/20 rounded mb-4" />
<div className="h-[300px] bg-primary/5 rounded" />
</div>
))}
</div>
<div className="flex items-center justify-center min-h-[400px]">
<div className="animate-spin rounded-full h-12 w-12 border-t-2 border-b-2 border-primary"></div>
</div>
);
}
@@ -56,18 +33,15 @@ export function AnalyticsPage() {
<div className="bg-gradient-to-br from-primary/10 to-secondary/5 border border-primary/20 rounded-xl p-6">
<div className="flex items-center gap-3 text-warning mb-2">
<Database size={24} />
<span className="font-semibold">Unable to Load Analytics</span>
<span className="font-semibold">Database Not Connected</span>
</div>
<p className="text-text-secondary">
Could not connect to the analytics database. Ensure PostgreSQL is running and
<code className="bg-bg-card px-2 py-1 rounded mx-1">DATABASE_URL</code> is configured correctly.
Set <code className="bg-bg-card px-2 py-1 rounded">USE_DATABASE=true</code> in your .env file to enable analytics tracking.
</p>
<button
onClick={() => refetch()}
className="mt-3 text-sm bg-primary/20 hover:bg-primary/30 text-primary font-medium px-4 py-2 rounded-lg transition-colors"
>
Retry
</button>
</div>
<div className="flex items-center gap-2 bg-secondary/10 border border-secondary/20 text-secondary rounded-xl px-4 py-3">
<AlertCircle size={18} />
<span>Analytics features require storing analysis results in PostgreSQL for historical tracking.</span>
</div>
</div>
);
+3 -15
View File
@@ -114,21 +114,9 @@ export function Batch() {
{/* Error */}
{mutation.isError && (
<div className="bg-error/10 border border-error/20 rounded-xl px-4 py-3">
<div className="flex items-center gap-2 text-error">
<AlertCircle size={18} />
<span className="font-semibold">Batch analysis failed</span>
</div>
<p className="text-text-secondary text-sm mt-1 ml-7">
{mutation.error instanceof Error ? mutation.error.message : 'An unexpected error occurred.'}
{' '}Check your connection and try again.
</p>
<button
onClick={() => mutation.reset()}
className="ml-7 mt-2 text-sm text-primary hover:text-primary-dark underline"
>
Dismiss
</button>
<div className="flex items-center gap-2 bg-error/10 border border-error/20 text-error rounded-xl px-4 py-3">
<AlertCircle size={18} />
<span>Batch analysis failed. Please try again.</span>
</div>
)}