Merge branch 'main' into user-export

.github/workflows/codeql-analysis.yml

@@ -40,7 +40,7 @@ jobs:
 
     # Initializes the CodeQL tools for scanning.
     - name: Initialize CodeQL
-      uses: github/codeql-action/init@v2
+      uses: github/codeql-action/init@v3
       with:
         languages: ${{ matrix.language }}
         # If you wish to specify custom queries, you can do so here or in a config file.
@@ -51,7 +51,7 @@ jobs:
     # Autobuild attempts to build any compiled languages  (C/C++, C#, or Java).
     # If this step fails, then you should remove it and run the build manually (see below)
     - name: Autobuild
-      uses: github/codeql-action/autobuild@v2
+      uses: github/codeql-action/autobuild@v3
 
     # ℹ️ Command-line programs to run using the OS shell.
     # 📚 https://git.io/JvXDl
@@ -65,4 +65,4 @@ jobs:
     #   make release
 
     - name: Perform CodeQL Analysis
-      uses: github/codeql-action/analyze@v2
+      uses: github/codeql-action/analyze@v3

.github/workflows/lint-frontend.yaml

@@ -22,7 +22,8 @@ jobs:
       - uses: actions/checkout@v4
 
       - name: Install modules
-        run: npm install stylelint stylelint-config-recommended stylelint-config-standard stylelint-order eslint
+      #  run: npm install stylelint stylelint-config-recommended stylelint-config-standard stylelint-order eslint
+        run: npm install eslint@^8.9.0
 
       # See .stylelintignore for files that are not linted.
       # - name: Run stylelint

bookwyrm/activitystreams.py

@@ -139,14 +139,14 @@ def _get_audience(self, status):  # pylint: disable=no-self-use
                 | (
                     Q(following=status.user) & Q(following=status.reply_parent.user)
                 )  # if the user is following both authors
-            ).distinct()
+            )
 
         # only visible to the poster's followers and tagged users
         elif status.privacy == "followers":
             audience = audience.filter(
                 Q(following=status.user)  # if the user is following the author
             )
-        return audience.distinct()
+        return audience.distinct("id")
 
     @tracer.start_as_current_span("ActivityStream.get_audience")
     def get_audience(self, status):
@@ -156,7 +156,7 @@ def get_audience(self, status):
         status_author = models.User.objects.filter(
             is_active=True, local=True, id=status.user.id
         ).values_list("id", flat=True)
-        return list(set(list(audience) + list(status_author)))
+        return list(set(audience) | set(status_author))
 
     def get_stores_for_users(self, user_ids):
         """convert a list of user ids into redis store ids"""
@@ -183,15 +183,13 @@ class HomeStream(ActivityStream):
     def get_audience(self, status):
         trace.get_current_span().set_attribute("stream_id", self.key)
         audience = super()._get_audience(status)
-        if not audience:
-            return []
         # if the user is following the author
         audience = audience.filter(following=status.user).values_list("id", flat=True)
         # if the user is the post's author
         status_author = models.User.objects.filter(
             is_active=True, local=True, id=status.user.id
         ).values_list("id", flat=True)
-        return list(set(list(audience) + list(status_author)))
+        return list(set(audience) | set(status_author))
 
     def get_statuses_for_user(self, user):
         return models.Status.privacy_filter(
@@ -239,9 +237,7 @@ def _get_audience(self, status):
         )
 
         audience = super()._get_audience(status)
-        if not audience:
-            return models.User.objects.none()
-        return audience.filter(shelfbook__book__parent_work=work).distinct()
+        return audience.filter(shelfbook__book__parent_work=work)
 
     def get_audience(self, status):
         # only show public statuses on the books feed,

bookwyrm/management/commands/deduplicate_book_data.py

@@ -1,13 +1,14 @@
 """ PROCEED WITH CAUTION: uses deduplication fields to permanently
 merge book data objects """
+
 from django.core.management.base import BaseCommand
 from django.db.models import Count
 from bookwyrm import models
-from bookwyrm.management.merge import merge_objects
 
 
-def dedupe_model(model):
+def dedupe_model(model, dry_run=False):
     """combine duplicate editions and update related models"""
+    print(f"deduplicating {model.__name__}:")
     fields = model._meta.get_fields()
     dedupe_fields = [
         f for f in fields if hasattr(f, "deduplication_field") and f.deduplication_field
@@ -16,30 +17,42 @@ def dedupe_model(model):
         dupes = (
             model.objects.values(field.name)
             .annotate(Count(field.name))
-            .filter(**{"%s__count__gt" % field.name: 1})
+            .filter(**{f"{field.name}__count__gt": 1})
+            .exclude(**{field.name: ""})
+            .exclude(**{f"{field.name}__isnull": True})
         )
 
         for dupe in dupes:
             value = dupe[field.name]
-            if not value or value == "":
-                continue
             print("----------")
-            print(dupe)
             objs = model.objects.filter(**{field.name: value}).order_by("id")
             canonical = objs.first()
-            print("keeping", canonical.remote_id)
+            action = "would merge" if dry_run else "merging"
+            print(
+                f"{action} into {model.__name__} {canonical.remote_id} based on {field.name} {value}:"
+            )
             for obj in objs[1:]:
-                print(obj.remote_id)
-                merge_objects(canonical, obj)
+                print(f"- {obj.remote_id}")
+                absorbed_fields = obj.merge_into(canonical, dry_run=dry_run)
+                print(f"  absorbed fields: {absorbed_fields}")
 
 
 class Command(BaseCommand):
     """deduplicate allllll the book data models"""
 
     help = "merges duplicate book data"
+
+    def add_arguments(self, parser):
+        """add the arguments for this command"""
+        parser.add_argument(
+            "--dry_run",
+            action="store_true",
+            help="don't actually merge, only print what would happen",
+        )
+
     # pylint: disable=no-self-use,unused-argument
     def handle(self, *args, **options):
         """run deduplications"""
-        dedupe_model(models.Edition)
-        dedupe_model(models.Work)
-        dedupe_model(models.Author)
+        dedupe_model(models.Edition, dry_run=options["dry_run"])
+        dedupe_model(models.Work, dry_run=options["dry_run"])
+        dedupe_model(models.Author, dry_run=options["dry_run"])

bookwyrm/management/merge_command.py

@@ -1,4 +1,3 @@
-from bookwyrm.management.merge import merge_objects
 from django.core.management.base import BaseCommand
 
 
@@ -9,6 +8,11 @@ def add_arguments(self, parser):
         """add the arguments for this command"""
         parser.add_argument("--canonical", type=int, required=True)
         parser.add_argument("--other", type=int, required=True)
+        parser.add_argument(
+            "--dry_run",
+            action="store_true",
+            help="don't actually merge, only print what would happen",
+        )
 
     # pylint: disable=no-self-use,unused-argument
     def handle(self, *args, **options):
@@ -26,4 +30,8 @@ def handle(self, *args, **options):
             print("other book doesn’t exist!")
             return
 
-        merge_objects(canonical, other)
+        absorbed_fields = other.merge_into(canonical, dry_run=options["dry_run"])
+
+        action = "would be" if options["dry_run"] else "has been"
+        print(f"{other.remote_id} {action} merged into {canonical.remote_id}")
+        print(f"absorbed fields: {absorbed_fields}")

bookwyrm/migrations/0197_mergedauthor_mergedbook.py

@@ -0,0 +1,48 @@
+# Generated by Django 3.2.24 on 2024-02-28 21:30
+
+from django.db import migrations, models
+import django.db.models.deletion
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0196_merge_pr3134_into_main"),
+    ]
+
+    operations = [
+        migrations.CreateModel(
+            name="MergedBook",
+            fields=[
+                ("deleted_id", models.IntegerField(primary_key=True, serialize=False)),
+                (
+                    "merged_into",
+                    models.ForeignKey(
+                        on_delete=django.db.models.deletion.PROTECT,
+                        related_name="absorbed",
+                        to="bookwyrm.book",
+                    ),
+                ),
+            ],
+            options={
+                "abstract": False,
+            },
+        ),
+        migrations.CreateModel(
+            name="MergedAuthor",
+            fields=[
+                ("deleted_id", models.IntegerField(primary_key=True, serialize=False)),
+                (
+                    "merged_into",
+                    models.ForeignKey(
+                        on_delete=django.db.models.deletion.PROTECT,
+                        related_name="absorbed",
+                        to="bookwyrm.author",
+                    ),
+                ),
+            ],
+            options={
+                "abstract": False,
+            },
+        ),
+    ]

bookwyrm/migrations/0199_status_bookwyrm_st_remote__06aeba_idx.py

@@ -0,0 +1,19 @@
+# Generated by Django 3.2.25 on 2024-04-02 19:53
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0198_book_search_vector_author_aliases"),
+    ]
+
+    operations = [
+        migrations.AddIndex(
+            model_name="status",
+            index=models.Index(
+                fields=["remote_id"], name="bookwyrm_st_remote__06aeba_idx"
+            ),
+        ),
+    ]

bookwyrm/migrations/0200_status_bookwyrm_st_thread__cf064f_idx.py

@@ -0,0 +1,19 @@
+# Generated by Django 3.2.25 on 2024-04-03 19:05
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0199_status_bookwyrm_st_remote__06aeba_idx"),
+    ]
+
+    operations = [
+        migrations.AddIndex(
+            model_name="status",
+            index=models.Index(
+                fields=["thread_id"], name="bookwyrm_st_thread__cf064f_idx"
+            ),
+        ),
+    ]

bookwyrm/migrations/0201_keypair_bookwyrm_ke_remote__472927_idx.py

@@ -0,0 +1,19 @@
+# Generated by Django 3.2.25 on 2024-04-03 19:10
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0200_status_bookwyrm_st_thread__cf064f_idx"),
+    ]
+
+    operations = [
+        migrations.AddIndex(
+            model_name="keypair",
+            index=models.Index(
+                fields=["remote_id"], name="bookwyrm_ke_remote__472927_idx"
+            ),
+        ),
+    ]

bookwyrm/migrations/0202_user_bookwyrm_us_usernam_b2546d_idx.py

@@ -0,0 +1,19 @@
+# Generated by Django 3.2.25 on 2024-04-03 19:14
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0201_keypair_bookwyrm_ke_remote__472927_idx"),
+    ]
+
+    operations = [
+        migrations.AddIndex(
+            model_name="user",
+            index=models.Index(
+                fields=["username"], name="bookwyrm_us_usernam_b2546d_idx"
+            ),
+        ),
+    ]

bookwyrm/migrations/0203_user_bookwyrm_us_is_acti_972dc4_idx.py

@@ -0,0 +1,19 @@
+# Generated by Django 3.2.25 on 2024-04-03 19:22
+
+from django.db import migrations, models
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0202_user_bookwyrm_us_usernam_b2546d_idx"),
+    ]
+
+    operations = [
+        migrations.AddIndex(
+            model_name="user",
+            index=models.Index(
+                fields=["is_active", "local"], name="bookwyrm_us_is_acti_972dc4_idx"
+            ),
+        ),
+    ]

bookwyrm/migrations/0204_merge_20240409_1042.py

@@ -0,0 +1,13 @@
+# Generated by Django 3.2.25 on 2024-04-09 10:42
+
+from django.db import migrations
+
+
+class Migration(migrations.Migration):
+
+    dependencies = [
+        ("bookwyrm", "0197_mergedauthor_mergedbook"),
+        ("bookwyrm", "0203_user_bookwyrm_us_is_acti_972dc4_idx"),
+    ]
+
+    operations = []