Intermediate commit

app/services/domains/publications/pdf_resolution_pipeline.py

@@ -5,6 +5,7 @@ import logging
 from typing import Any
 
 from app.services.domains.arxiv.application import ArxivRateLimitError
+from app.services.domains.arxiv.guards import arxiv_skip_reason_for_item
 from app.services.domains.openalex.client import OpenAlexBudgetExhaustedError
 from app.services.domains.publications.types import PublicationListItem
 from app.services.domains.unpaywall.application import OaResolutionOutcome, resolve_publication_oa_outcomes
@@ -17,6 +18,7 @@ logger = logging.getLogger(__name__)
 class PipelineOutcome:
     outcome: OaResolutionOutcome | None
     scholar_candidates: Any | None  # Kept for backward compatibility with calling signatures
+    arxiv_rate_limited: bool = False
 
 
 async def resolve_publication_pdf_outcome_for_row(
@@ -24,6 +26,7 @@ async def resolve_publication_pdf_outcome_for_row(
     row: PublicationListItem,
     request_email: str | None,
     openalex_api_key: str | None = None,
+    allow_arxiv_lookup: bool = True,
 ) -> PipelineOutcome:
     # 1. OpenAlex OA — raises OpenAlexBudgetExhaustedError if budget is gone
     openalex_outcome = await _openalex_outcome(row, request_email=request_email, openalex_api_key=openalex_api_key)
@@ -31,13 +34,29 @@ async def resolve_publication_pdf_outcome_for_row(
         return PipelineOutcome(openalex_outcome, None)
 
     # 2. arXiv
-    arxiv_outcome = await _arxiv_outcome(row, request_email=request_email)
+    arxiv_rate_limited = False
+    try:
+        arxiv_outcome = await _arxiv_outcome(
+            row,
+            request_email=request_email,
+            allow_lookup=allow_arxiv_lookup,
+        )
+    except ArxivRateLimitError:
+        arxiv_rate_limited = True
+        arxiv_outcome = None
+        logger.warning(
+            "publications.pdf_resolution.arxiv_rate_limited",
+            extra={
+                "event": "publications.pdf_resolution.arxiv_rate_limited",
+                "publication_id": int(row.publication_id),
+            },
+        )
     if arxiv_outcome and arxiv_outcome.pdf_url:
-        return PipelineOutcome(arxiv_outcome, None)
+        return PipelineOutcome(arxiv_outcome, None, arxiv_rate_limited=arxiv_rate_limited)
 
     # 3. Unpaywall (which falls back to Crossref)
     oa_outcome = await _oa_outcome(row=row, request_email=request_email)
-    return PipelineOutcome(oa_outcome, None)
+    return PipelineOutcome(oa_outcome, None, arxiv_rate_limited=arxiv_rate_limited)
 
 
 async def _openalex_outcome(
@@ -87,9 +106,23 @@ async def _openalex_outcome(
     return None
 
 
-async def _arxiv_outcome(row: PublicationListItem, request_email: str | None) -> OaResolutionOutcome | None:
+async def _arxiv_outcome(
+    row: PublicationListItem,
+    *,
+    request_email: str | None,
+    allow_lookup: bool = True,
+) -> OaResolutionOutcome | None:
     from app.services.domains.arxiv.application import discover_arxiv_id_for_publication
 
+    if not allow_lookup:
+        _log_arxiv_skip(row=row, skip_reason="batch_arxiv_cooldown_active")
+        return None
+
+    skip_reason = arxiv_skip_reason_for_item(item=row)
+    if skip_reason is not None:
+        _log_arxiv_skip(row=row, skip_reason=skip_reason)
+        return None
+
     try:
         arxiv_id = await discover_arxiv_id_for_publication(item=row, request_email=request_email)
         if arxiv_id:
@@ -103,7 +136,7 @@ async def _arxiv_outcome(row: PublicationListItem, request_email: str | None) ->
                 used_crossref=False,
             )
     except ArxivRateLimitError:
-        raise  # propagate so the batch loop can stop
+        raise  # propagate so orchestration can switch to non-arXiv fallback
     except Exception as exc:
         logger.warning(
             "publications.pdf_resolution.arxiv_failed",
@@ -112,6 +145,17 @@ async def _arxiv_outcome(row: PublicationListItem, request_email: str | None) ->
     return None
 
 
+def _log_arxiv_skip(*, row: PublicationListItem, skip_reason: str) -> None:
+    logger.info(
+        "publications.pdf_resolution.arxiv_skipped",
+        extra={
+            "event": "publications.pdf_resolution.arxiv_skipped",
+            "publication_id": int(row.publication_id),
+            "skip_reason": skip_reason,
+        },
+    )
+
+
 async def _oa_outcome(
     *,
     row: PublicationListItem,