update model about merging works

catalog/book/models.py

@@ -22,6 +22,7 @@ from os.path import exists
 from django.core.validators import MaxValueValidator, MinValueValidator
 from django.db import models
 from django.utils.translation import gettext_lazy as _
+from loguru import logger as _logger
 
 from catalog.common import (
     BaseSchema,
@@ -163,19 +164,39 @@ class Edition(Item):
             return detect_isbn_asin(lookup_id_value)
         return super().lookup_id_cleanup(lookup_id_type, lookup_id_value)
 
+    def merge_to(self, to_item):
+        super().merge_to(to_item)
+        for work in self.works.all():
+            to_item.works.add(work)
+        self.works.clear()
+
+    def delete(self, using=None, soft=True, *args, **kwargs):
+        if soft:
+            self.works.clear()
+        return super().delete(using, soft, *args, **kwargs)
+
     def update_linked_items_from_external_resource(self, resource):
         """add Work from resource.metadata['work'] if not yet"""
         links = resource.required_resources + resource.related_resources
         for w in links:
             if w.get("model") == "Work":
-                work = Work.objects.filter(
-                    primary_lookup_id_type=w["id_type"],
-                    primary_lookup_id_value=w["id_value"],
+                work_res = ExternalResource.objects.filter(
+                    id_type=w["id_type"], id_value=w["id_value"]
                 ).first()
+                if work_res:
+                    work = work_res.item
+                    if not work:
+                        _logger.warning(f"Unable to find work for {work_res}")
+                else:
+                    _logger.warning(
+                        f'Unable to find resource for {w["id_type"]}:{w["id_value"]}'
+                    )
+                    work = Work.objects.filter(
+                        primary_lookup_id_type=w["id_type"],
+                        primary_lookup_id_value=w["id_value"],
+                    ).first()
                 if work and work not in self.works.all():
                     self.works.add(work)
-                # if not work:
-                #     _logger.info(f'Unable to find link for {w["url"]}')
 
     def get_related_books(self):
         works = list(self.works.all())
@@ -224,6 +245,17 @@ class Work(Item):
     # a lazy fix is to remove smaller DoubanBook_Work ids
     # but ideally deal with 302 in scrape().
 
+    def merge_to(self, to_item):
+        super().merge_to(to_item)
+        for edition in self.editions.all():
+            to_item.editions.add(edition)
+        self.editions.clear()
+
+    def delete(self, using=None, soft=True, *args, **kwargs):
+        if soft:
+            self.editions.clear()
+        return super().delete(using, soft, *args, **kwargs)
+
 
 class Series(Item):
     category = ItemCategory.Book

catalog/book/tests.py

@@ -387,3 +387,46 @@ class MultiBookSitesTestCase(TestCase):
         self.assertEqual(w3e[1].title, "黄金时代")
         e = Edition.objects.get(primary_lookup_id_value=9781662601217)
         self.assertEqual(e.title, "Golden Age: A Novel")
+
+    @use_local_response
+    def test_works_merge(self):
+        # url1 and url4 has same ISBN, hence they share same Edition instance, which belongs to 2 Work instances
+        url1 = "https://book.douban.com/subject/1089243/"
+        url2 = "https://book.douban.com/subject/2037260/"
+        url3 = "https://www.goodreads.com/book/show/59952545-golden-age"
+        url4 = "https://www.goodreads.com/book/show/11798823"
+        p1 = SiteManager.get_site_by_url(
+            url1
+        ).get_resource_ready()  # lxml bug may break this
+        w1 = p1.item.works.all().first()
+        p2 = SiteManager.get_site_by_url(url2).get_resource_ready()
+        w2 = p2.item.works.all().first()
+        self.assertEqual(w1, w2)
+        self.assertEqual(p1.item.works.all().count(), 1)
+        p3 = SiteManager.get_site_by_url(url3).get_resource_ready()
+        w3 = p3.item.works.all().first()
+        self.assertNotEqual(w3, w2)
+        self.assertEqual(w2.external_resources.all().count(), 1)
+        self.assertEqual(w3.external_resources.all().count(), 1)
+        w3.merge_to(w2)
+        self.assertEqual(w2.external_resources.all().count(), 2)
+        self.assertEqual(w3.external_resources.all().count(), 0)
+        self.assertEqual(w2.editions.all().count(), 3)
+        self.assertEqual(w3.editions.all().count(), 0)
+        p4 = SiteManager.get_site_by_url(url4).get_resource_ready()
+        self.assertEqual(p4.item.id, p1.item.id)
+        self.assertEqual(p4.item.works.all().count(), 1)
+        self.assertEqual(p1.item.works.all().count(), 1)
+        w2e = w2.editions.all().order_by("title")
+        self.assertEqual(w2e.count(), 3)
+        self.assertEqual(w2e[0].title, "Golden Age: A Novel")
+        self.assertEqual(w2e[1].title, "Wang in Love and Bondage")
+        self.assertEqual(w2e[2].title, "黄金时代")
+        w3e = w3.editions.all().order_by("title")
+        self.assertEqual(w3e.count(), 0)
+        e = Edition.objects.get(primary_lookup_id_value=9781662601217)
+        self.assertEqual(e.title, "Golden Age: A Novel")
+        w2e[1].delete()
+        self.assertEqual(w2.editions.all().count(), 2)
+        w2e.delete()
+        self.assertEqual(p1.item.works.all().count(), 0)