Handle pdf extraction error

src/app/domain/jobs/controllers/job_posts.py

@@ -165,8 +165,19 @@ async def create_job_post_from_url(
         # Generate pdf and save to s3
         if db_obj.url:
             pdf_content = await get_pdf(db_obj.url)
-            s3_client = boto3.client("s3")
-            s3_client.put_object(Bucket=app_s3_bucket_name, Key=f"job_posts/{db_obj.id}.pdf", Body=pdf_content)
+            if pdf_content:
+                s3_client = boto3.client("s3")
+                s3_client.put_object(Bucket=app_s3_bucket_name, Key=f"job_posts/{db_obj.id}.pdf", Body=pdf_content)
+            else:
+                error_msg = "Couldn't get pdf for a job post"
+                await logger.awarn(
+                    error_msg,
+                    job_id=db_obj.id,
+                    job_title=db_obj.title,
+                    company_id=company_db_obj.id,
+                    company_name=company_db_obj.name,
+                    company_url=company_db_obj.url,
+                )
 
         return job_posts_service.to_schema(schema_type=JobPost, data=db_obj)
 

src/app/lib/pdfshift.py

@@ -7,29 +7,35 @@
 pdfshift_api_key = os.environ["PDFSHIFT_API_KEY"]
 
 
-async def get_pdf(url: str, encode: bool = False) -> bytes | str:
+async def get_pdf(url: str, encode: bool = False) -> bytes | str | None:
     """Get pdf."""
     if not url:
         error_msg = "URL is required"
         raise ValueError(error_msg)
 
     body = {"source": url, "encode": encode, "timeout": 30}
 
-    async with httpx.AsyncClient() as client:
-        response = await client.post(
-            "https://api.pdfshift.io/v3/convert/pdf",
-            auth=("api", pdfshift_api_key),
-            json=body,
-        )
-
-        if response.status_code != 200 and not response.content:
-            error_msg = "Failed to extract pdf."
-            await logger.awarn(
-                error_msg,
-                status=response.status_code,
-                response=response.content,
-                url=url,
+    try:
+        async with httpx.AsyncClient() as client:
+            response = await client.post(
+                "https://api.pdfshift.io/v3/convert/pdf",
+                auth=("api", pdfshift_api_key),
+                json=body,
             )
-            raise LookupError(error_msg)
 
-        return response.content
+            if response.status_code != 200 and not response.content:
+                error_msg = "Failed to extract pdf."
+                await logger.awarn(
+                    error_msg,
+                    status=response.status_code,
+                    response=response.content,
+                    url=url,
+                )
+                raise LookupError(error_msg)
+
+            return response.content
+    except httpx.ReadTimeout as e:
+        error_msg = "Failed to get pdf"
+        await logger.awarn(error_msg, exc_info=e)
+
+    return None