[WIP] Fix task harvest pidproviderxml

article/controller.py

@@ -369,7 +369,7 @@ def bulk_export_articles_to_articlemeta(
                         "article_id": article.id,
                         "article_pid": getattr(article, "pid", None),
                         "journal_acron": getattr(article, "journal_acron", None),
-                        "pub_year": getattr(article, "pub_year", None),
+                        "pub_date_year": getattr(article, "pub_date_year", None),
                         "force_update": force_update,
                     },
                 )
@@ -524,9 +524,9 @@ def _iter_from_article(self):
         if journal_id_list:
             filters["journal__in"] = journal_id_list
         if self.from_pub_year:
-            filters["pub_year__gte"] = self.from_pub_year
+            filters["pub_date_year__gte"] = self.from_pub_year
         if self.until_pub_year:
-            filters["pub_year__lte"] = self.until_pub_year
+            filters["pub_date_year__lte"] = self.until_pub_year
         if self.from_date:
             filters["updated__gte"] = self.from_date
         if self.until_date:
@@ -596,6 +596,7 @@ def _build_harvester(self, collection_acron):
             timeout=self.timeout,
         )
         if collection_acron == "scl":
-            return OPACHarvester(self.opac_url or "www.scielo.br", collection_acron, **kwargs)
+            domain = self.opac_url or Collection.get(collection_acron).base_url
+            return OPACHarvester(domain, collection_acron, **kwargs)
         return AMHarvester("article", collection_acron, **kwargs)
 

article/models.py

@@ -1812,6 +1812,7 @@ def create(cls, user, url=None, source_date=None, am_article=None, force_update=
             obj.source_date = source_date
             obj.am_article = am_article
             obj.status = cls.StatusChoices.PENDING
+            obj.save()
             obj.add_pid_provider(user, force_update, auto_solve_pid_conflict=auto_solve_pid_conflict)
             return obj
         except IntegrityError:

core/utils/harvesters.py

@@ -3,6 +3,8 @@
 from typing import Any, Dict, Generator, Optional
 from urllib.parse import urlencode
 
+from article.utils.url_builder import ArticleURLBuilder
+
 from core.utils.utils import fetch_data
 
 
@@ -224,7 +226,9 @@ def harvest_documents(self) -> Generator[Dict[str, Any], None, None]:
 
                     # Constrói URL do XML
                     journal_acron = item["journal_acronym"]
-                    xml_url = f"{self.domain}/j/{journal_acron}/a/{pid_v3}/?format=xml"
+                    xml_url = ArticleURLBuilder(
+                        self.domain, journal_acron, pid_v3=pid_v3
+                    ).get_xml_url()
 
                     # Extrai data de origem
                     origin_date = self._parse_gmt_date(
@@ -288,4 +292,3 @@ def _parse_gmt_date(self, date_str: Optional[str]) -> Optional[str]:
         except (ValueError, TypeError) as e:
             logging.warning(f"Failed to parse GMT date '{date_str}': {e}")
             return None
-