From a490e141267c52e0f6ba278ba69d6d2ee15f14b5 Mon Sep 17 00:00:00 2001
From: Justin Visser <justinfvisser@gmail.com>
Date: Mon, 2 Mar 2026 18:01:42 +0100
Subject: [PATCH] openalex cooldown

---
 .claude/settings.json                             |  3 ++-
 app/services/ingestion/scheduler.py               |  4 ++++
 app/services/publications/pdf_queue_resolution.py | 15 +++++++++++++++
 3 files changed, 21 insertions(+), 1 deletion(-)

diff --git a/.claude/settings.json b/.claude/settings.json
index ec4aab3..e801a7c 100644
--- a/.claude/settings.json
+++ b/.claude/settings.json
@@ -23,7 +23,8 @@
       "Bash(gh run list --branch openalex-size-reduction-scholar-hydration --limit 3 --json databaseId,status,conclusion,name,event)",
       "Bash(gh run list --limit 10 --json databaseId,status,conclusion,name,event,headBranch)",
       "Bash(gh run view 22554585841 --log-failed)",
-      "Bash(gh run rerun 22554585841 --failed)"
+      "Bash(gh run rerun 22554585841 --failed)",
+      "Bash(bash scripts/premerge.sh)"
     ],
     "additionalDirectories": [
       "/home/jv/src/personal/playground/scholar_scraper/scholar-scraper/frontend/src"
diff --git a/app/services/ingestion/scheduler.py b/app/services/ingestion/scheduler.py
index bf4e9e7..6d2bec0 100644
--- a/app/services/ingestion/scheduler.py
+++ b/app/services/ingestion/scheduler.py
@@ -289,6 +289,10 @@ class SchedulerService:
 
     async def _drain_pdf_queue(self) -> None:
         from app.services.publications.pdf_queue import drain_ready_jobs
+        from app.services.publications.pdf_queue_resolution import is_budget_cooldown_active
+
+        if is_budget_cooldown_active():
+            return
 
         session_factory = get_session_factory()
         async with session_factory() as session:
diff --git a/app/services/publications/pdf_queue_resolution.py b/app/services/publications/pdf_queue_resolution.py
index ec4422e..26e4b0b 100644
--- a/app/services/publications/pdf_queue_resolution.py
+++ b/app/services/publications/pdf_queue_resolution.py
@@ -2,6 +2,7 @@ from __future__ import annotations
 
 import asyncio
 import logging
+from datetime import UTC, datetime, timedelta
 
 from app.db.models import Publication, PublicationPdfJob
 from app.db.session import get_session_factory
@@ -29,8 +30,20 @@ PDF_EVENT_ATTEMPT_STARTED = "attempt_started"
 PDF_EVENT_RESOLVED = "resolved"
 PDF_EVENT_FAILED = "failed"
 
+_BUDGET_COOLDOWN_MINUTES = 15
+
 logger = logging.getLogger(__name__)
 _scheduled_tasks: set[asyncio.Task[None]] = set()
+_budget_cooldown_until: datetime | None = None
+
+
+def is_budget_cooldown_active() -> bool:
+    return _budget_cooldown_until is not None and datetime.now(UTC) < _budget_cooldown_until
+
+
+def _enter_budget_cooldown() -> None:
+    global _budget_cooldown_until
+    _budget_cooldown_until = datetime.now(UTC) + timedelta(minutes=_BUDGET_COOLDOWN_MINUTES)
 
 
 async def _mark_attempt_started(
@@ -233,11 +246,13 @@ async def _run_resolution_task(
                     detail="arXiv temporarily disabled for remaining batch after rate limit",
                 )
         except OpenAlexBudgetExhaustedError:
+            _enter_budget_cooldown()
             structured_log(
                 logger,
                 "warning",
                 "pdf_queue.budget_exhausted",
                 detail="Stopping PDF resolution batch — OpenAlex daily budget exhausted",
+                cooldown_minutes=_BUDGET_COOLDOWN_MINUTES,
             )
             break
         except Exception: