Update existing record

scoap3/articles/api/views.py

@@ -63,7 +63,6 @@ def create(self, request, *args, **kwargs):
         return Response(
             serializer.data, status=status.HTTP_201_CREATED, headers=headers
         )
-        article_id = data.get("id")
 
 
 class ArticleWorkflowImportView(ViewSet):

scoap3/articles/tests/data/record_failing_on_airflow.json

@@ -0,0 +1,116 @@
+{
+    "dois":[
+       {
+          "value":"10.1103/PhysRevD.109.046003"
+       }
+    ],
+    "page_nr":[
+       17
+    ],
+    "arxiv_eprints":[
+       {
+          "value":"2312.06750",
+          "categories":[
+             "hep-th",
+             "gr-qc",
+             "math-ph",
+             "math.MP"
+          ]
+       }
+    ],
+    "authors":[
+       {
+          "full_name":"Chatwin-Davies, Aidan",
+          "given_names":"Aidan",
+          "surname":"Chatwin-Davies",
+          "affiliations":[
+             {
+                "value":"Okinawa Institute of Science and Technology, 1919-1 Tancha, Onna-son, Kunigami-gun, Okinawa, 904-0495, Japan",
+                "organization":"Okinawa Institute of Science and Technology, 1919-1 Tancha, Onna-son, Kunigami-gun, Okinawa, 904-0495",
+                "country":"Japan"
+             }
+          ]
+       },
+       {
+          "full_name":"Leung, Pompey",
+          "given_names":"Pompey",
+          "surname":"Leung",
+          "affiliations":[
+             {
+                "value":"Department of Physics and Astronomy, University of British Columbia, 6224 Agricultural Road, Vancouver, British Columbia, V6T 1Z1, Canada",
+                "organization":"Department of Physics and Astronomy, University of British Columbia, 6224 Agricultural Road, Vancouver, British Columbia, V6T 1Z1",
+                "country":"Canada"
+             }
+          ]
+       },
+       {
+          "full_name":"Remmen, Grant N.",
+          "given_names":"Grant N.",
+          "surname":"Remmen",
+          "affiliations":[
+             {
+                "value":"Center for Cosmology and Particle Physics, Department of Physics, New York University, New York, New York, 10003, USA",
+                "organization":"Center for Cosmology and Particle Physics, Department of Physics, New York University, New York, New York, 10003",
+                "country":"USA"
+             }
+          ]
+       }
+    ],
+    "license":[
+       {
+          "url":"https://creativecommons.org/licenses/by/4.0/",
+          "license":"CC-BY-4.0"
+       }
+    ],
+    "collections":[
+       {
+          "primary":"HEP"
+       },
+       {
+          "primary":"Citeable"
+       },
+       {
+          "primary":"Published"
+       }
+    ],
+    "publication_info":[
+       {
+          "journal_title":"Physical Review D",
+          "journal_volume":"109",
+          "year":2024,
+          "journal_issue":"4",
+          "material":"article"
+       }
+    ],
+    "abstracts":[
+       {
+          "value":"<p>Holographic screens are codimension-one hypersurfaces that extend the notion of apparent horizons to general (non-black hole) spacetimes and that display interesting thermodynamic properties. We show that if a spacetime contains a codimension-two, boundary-homologous, minimal extremal spacelike surface <math xmlns=\"http://www.w3.org/1998/Math/MathML\" display=\"inline\"><mi>X</mi></math> (known as an HRT surface in <math xmlns=\"http://www.w3.org/1998/Math/MathML\" display=\"inline\"><mrow><mi>AdS</mi><mo>/</mo><mi>CFT</mi></mrow></math>), then any holographic screens are sequestered to the causal wedges of <math xmlns=\"http://www.w3.org/1998/Math/MathML\" display=\"inline\"><mi>X</mi></math>. That is, any single connected component of a holographic screen can be located in at most one of the causal future, causal past, inner wedge, or outer wedge of <math xmlns=\"http://www.w3.org/1998/Math/MathML\" display=\"inline\"><mi>X</mi></math>. We comment on how this result informs possible coarse grained entropic interpretations of generic holographic screens, as well as on connections to semiclassical objects such as quantum extremal surfaces.</p>",
+          "source":"APS"
+       }
+    ],
+    "acquisition_source":{
+       "source":"APS",
+       "method":"APS",
+       "date":"2024-02-08T00:00:16.954712"
+    },
+    "copyright":[
+       {
+          "year":2024,
+          "statement":"Published by the American Physical Society"
+       }
+    ],
+    "imprints":[
+       {
+          "date":"2024-02-07",
+          "publisher":"APS"
+       }
+    ],
+    "record_creation_date":"2024-02-08T00:00:16.954712",
+    "titles":[
+       {
+          "title":"Holographic screen sequestration",
+          "source":"APS"
+       }
+    ],
+    "$schema":"https://repo.qa.scoap3.org/schemas/hep.json"
+ }

scoap3/articles/tests/data/workflow_record.json

@@ -10,7 +10,6 @@
       "statement": "The Author"
     }
   ],
-  "control_number": "81204",
   "_oai": {
     "updated": "2023-10-31T08:21:45Z",
     "id": "oai:repo.scoap3.org:81204",

scoap3/articles/tests/test_article_views.py

@@ -1,10 +1,14 @@
 import json
+import os
+from pathlib import Path
 
 import pytest
+from django.test import TestCase
 from django.urls import reverse
 from rest_framework import status
 
-from scoap3.articles.models import Article
+from scoap3.articles.models import Article, ArticleIdentifier
+from scoap3.users.tests.factories import UserFactory
 
 pytestmark = pytest.mark.django_db
 
@@ -53,7 +57,6 @@ def test_update_article_from_workflow(self, client, user, shared_datadir):
         )
 
         data["titles"][0]["title"] = "New title"
-        data["dois"].append({"value": "10.5506/APhysPolB.54.10-A5"})
         response = client.post(
             reverse("api:article-workflow-import-list"),
             data,
@@ -69,11 +72,13 @@ def test_update_article_from_workflow(self, client, user, shared_datadir):
         ]
 
         assert article.title == "New title"
-        assert len(expected_dois) == 2
-        assert "10.5506/APhysPolB.54.10-A5" in expected_dois
+        assert len(expected_dois) == 1
         assert "10.5506/APhysPolB.54.10-A3" in expected_dois
 
 
+pytestmark = pytest.mark.django_db
+
+
 class TestArticleIdentifierViewSet:
     def test_get_article_identifier(self, client):
         url = reverse("api:articleidentifier-list")
@@ -83,3 +88,43 @@ def test_get_article_identifier(self, client):
         url = reverse("api:articleidentifier-detail", kwargs={"pk": 0})
         response = client.get(url)
         assert response.status_code == status.HTTP_404_NOT_FOUND
+
+
+pytestmark = pytest.mark.django_db
+
+
+class TestTheSameArticleCreationTwice(TestCase):
+    @staticmethod
+    def shared_datadir():
+        return os.path.join(
+            Path(__file__).parent.resolve(), "data", "record_failing_on_airflow.json"
+        )
+
+    def test_create_article_from_workflow(self):
+        user = UserFactory()
+        self.client.force_login(user)
+        with open(self.shared_datadir()) as file:
+            contents = file.read()
+        data = json.loads(contents)
+        response = self.client.post(
+            reverse("api:article-workflow-import-list"),
+            data,
+            content_type="application/json",
+        )
+        assert response.status_code == status.HTTP_200_OK
+
+        response = self.client.post(
+            reverse("api:article-workflow-import-list"),
+            data,
+            content_type="application/json",
+        )
+        assert response.status_code == status.HTTP_200_OK
+        doi = data.get("dois")[0].get("value")
+        assert (
+            len(
+                ArticleIdentifier.objects.filter(
+                    identifier_type="DOI", identifier_value=doi
+                )
+            )
+            == 1
+        )

scoap3/exports/views.py

@@ -21,7 +21,6 @@ def generate_csv_response(data, action_name, write_header=True):
     writer = csv.writer(response)
     if write_header:
         writer.writerow(data.get("header"))
-    print(data)
     for row in data.get("data", []):
         writer.writerow(row)
 

scoap3/tasks.py

@@ -71,17 +71,21 @@ def _create_licenses(data):
 
 def _create_article(data, licenses):
     article_data = {
-        "id": data.get("control_number"),
         "publication_date": data["imprints"][0].get("date"),
         "title": data["titles"][0].get("title"),
         "subtitle": data["titles"][0].get("subtitle", ""),
         "abstract": data["abstracts"][0].get("value", ""),
     }
+    doi = data.get("dois")[0].get("value")
     if (
-        article_data.get("id")
-        and Article.objects.filter(pk=article_data["id"]).exists()
+        doi
+        and ArticleIdentifier.objects.filter(
+            identifier_type="DOI", identifier_value=doi
+        ).exists()
     ):
-        article = Article.objects.get(pk=article_data["id"])
+        article = ArticleIdentifier.objects.get(
+            identifier_type="DOI", identifier_value=doi
+        ).article_id
         article.__dict__.update(**article_data)
     else:
         article = Article.objects.create(**article_data)
@@ -110,14 +114,22 @@ def _create_article_identifier(data, article):
             "identifier_value": doi.get("value"),
         }
         ArticleIdentifier.objects.get_or_create(**article_identifier_data)
-
     for arxiv in data.get("arxiv_eprints", []):
         article_identifier_data = {
             "article_id": article,
             "identifier_type": "arXiv",
             "identifier_value": arxiv.get("value"),
         }
-        ArticleIdentifier.objects.get_or_create(**article_identifier_data)
+        doi = data.get("dois")[0].get("value")
+        if ArticleIdentifier.objects.filter(
+            identifier_type="arXiv", article_id=article
+        ).exists():
+            article_identifier = ArticleIdentifier.objects.get(
+                article_id=article, identifier_type="arXiv"
+            )
+            article_identifier.__dict__.update(**article_identifier_data)
+        else:
+            ArticleIdentifier.objects.get_or_create(**article_identifier_data)
 
 
 def _create_copyright(data, article):
@@ -128,7 +140,11 @@ def _create_copyright(data, article):
             "holder": copyright.get("holder", ""),
             "year": copyright.get("year"),
         }
-        Copyright.objects.get_or_create(**copyright_data)
+        if Copyright.objects.filter(article_id=article).exists():
+            copyright = Copyright.objects.get(article_id=article)
+            copyright.__dict__.update(**copyright_data)
+        else:
+            Copyright.objects.get_or_create(**copyright_data)
 
 
 def _create_article_arxiv_category(data, article):
@@ -141,8 +157,15 @@ def _create_article_arxiv_category(data, article):
                 "category": arxiv_category,
                 "primary": True if idx == 0 else False,
             }
-
-            ArticleArxivCategory.objects.get_or_create(**article_arxiv_category_data)
+            if ArticleArxivCategory.objects.filter(article_id=article).exists():
+                article_arxiv_category = ArticleArxivCategory.objects.get(
+                    article_id=article
+                )
+                article_arxiv_category.__dict__.update(**article_arxiv_category_data)
+            else:
+                ArticleArxivCategory.objects.get_or_create(
+                    **article_arxiv_category_data
+                )
 
 
 def _create_publisher(data):
@@ -170,7 +193,11 @@ def _create_publication_info(data, article, publishers):
             "journal_issue_date": publication_info.get("journal_issue_date"),
             "publisher_id": publishers[idx].id,
         }
-        PublicationInfo.objects.get_or_create(**publication_info_data)
+        if PublicationInfo.objects.filter(article_id=article).exists():
+            publication_info = PublicationInfo.objects.get(article_id=article)
+            publication_info.__dict__.update(**publication_info_data)
+        else:
+            PublicationInfo.objects.get_or_create(**publication_info_data)
 
 
 def _create_experimental_collaborations(data):