Merge v0.7.2 (production) into spinoza

Conflicts:
	requirements.txt

.env.example

@@ -137,3 +137,6 @@ TWO_FACTOR_LOGIN_MAX_SECONDS=60
 # and AWS_S3_CUSTOM_DOMAIN (if used) are added by default.
 # Value should be a comma-separated list of host names.
 CSP_ADDITIONAL_HOSTS=
+# The last number here means "megabytes"
+# Increase if users are having trouble uploading BookWyrm export files.
+DATA_UPLOAD_MAX_MEMORY_SIZE = (1024**2 * 100)

VERSION

@@ -1 +1 @@
-0.7.0
+0.7.1

bookwyrm/activitypub/base_activity.py

@@ -236,7 +236,7 @@ def serialize(self, **kwargs):
         omit = kwargs.get("omit", ())
         data = self.__dict__.copy()
         # recursively serialize
-        for (k, v) in data.items():
+        for k, v in data.items():
             try:
                 if issubclass(type(v), ActivityObject):
                     data[k] = v.serialize()
@@ -396,19 +396,15 @@ def resolve_remote_id(
 
 def get_representative():
     """Get or create an actor representing the instance
-    to sign requests to 'secure mastodon' servers"""
-    username = f"{INSTANCE_ACTOR_USERNAME}@{DOMAIN}"
-    email = "bookwyrm@localhost"
-    try:
-        user = models.User.objects.get(username=username)
-    except models.User.DoesNotExist:
-        user = models.User.objects.create_user(
-            username=username,
-            email=email,
+    to sign outgoing HTTP GET requests"""
+    return models.User.objects.get_or_create(
+        username=f"{INSTANCE_ACTOR_USERNAME}@{DOMAIN}",
+        defaults=dict(
+            email="bookwyrm@localhost",
             local=True,
             localname=INSTANCE_ACTOR_USERNAME,
-        )
-    return user
+        ),
+    )[0]
 
 
 def get_activitypub_data(url):

bookwyrm/activitypub/verbs.py

@@ -171,9 +171,19 @@ class Reject(Verb):
     type: str = "Reject"
 
     def action(self, allow_external_connections=True):
-        """reject a follow request"""
-        obj = self.object.to_model(save=False, allow_create=False)
-        obj.reject()
+        """reject a follow or follow request"""
+
+        for model_name in ["UserFollowRequest", "UserFollows", None]:
+            model = apps.get_model(f"bookwyrm.{model_name}") if model_name else None
+            if obj := self.object.to_model(
+                model=model,
+                save=False,
+                allow_create=False,
+                allow_external_connections=allow_external_connections,
+            ):
+                # Reject the first model that can be built.
+                obj.reject()
+                break
 
 
 @dataclass(init=False)

bookwyrm/book_search.py

@@ -43,6 +43,7 @@ def search(
     min_confidence: float = 0,
     filters: Optional[list[Any]] = None,
     return_first: bool = False,
+    books: Optional[QuerySet[models.Edition]] = None,
 ) -> Union[Optional[models.Edition], QuerySet[models.Edition]]:
     """search your local database"""
     filters = filters or []
@@ -54,13 +55,15 @@ def search(
     # first, try searching unique identifiers
     # unique identifiers never have spaces, title/author usually do
     if not " " in query:
-        results = search_identifiers(query, *filters, return_first=return_first)
+        results = search_identifiers(
+            query, *filters, return_first=return_first, books=books
+        )
 
     # if there were no identifier results...
     if not results:
         # then try searching title/author
         results = search_title_author(
-            query, min_confidence, *filters, return_first=return_first
+            query, min_confidence, *filters, return_first=return_first, books=books
         )
     return results
 
@@ -98,9 +101,17 @@ def format_search_result(search_result):
 
 
 def search_identifiers(
-    query, *filters, return_first=False
+    query,
+    *filters,
+    return_first=False,
+    books=None,
 ) -> Union[Optional[models.Edition], QuerySet[models.Edition]]:
-    """tries remote_id, isbn; defined as dedupe fields on the model"""
+    """search Editions by deduplication fields
+
+    Best for cases when we can assume someone is searching for an exact match on
+    commonly unique data identifiers like isbn or specific library ids.
+    """
+    books = books or models.Edition.objects
     if connectors.maybe_isbn(query):
         # Oh did you think the 'S' in ISBN stood for 'standard'?
         normalized_isbn = query.strip().upper().rjust(10, "0")
@@ -111,7 +122,7 @@ def search_identifiers(
         for f in models.Edition._meta.get_fields()
         if hasattr(f, "deduplication_field") and f.deduplication_field
     ]
-    results = models.Edition.objects.filter(
+    results = books.filter(
         *filters, reduce(operator.or_, (Q(**f) for f in or_filters))
     ).distinct()
 
@@ -121,12 +132,17 @@ def search_identifiers(
 
 
 def search_title_author(
-    query, min_confidence, *filters, return_first=False
+    query,
+    min_confidence,
+    *filters,
+    return_first=False,
+    books=None,
 ) -> QuerySet[models.Edition]:
     """searches for title and author"""
+    books = books or models.Edition.objects
     query = SearchQuery(query, config="simple") | SearchQuery(query, config="english")
     results = (
-        models.Edition.objects.filter(*filters, search_vector=query)
+        books.filter(*filters, search_vector=query)
         .annotate(rank=SearchRank(F("search_vector"), query))
         .filter(rank__gt=min_confidence)
         .order_by("-rank")
@@ -137,7 +153,7 @@ def search_title_author(
 
     # filter out multiple editions of the same work
     list_results = []
-    for work_id in set(editions_of_work[:30]):
+    for work_id in editions_of_work[:30]:
         result = (
             results.filter(parent_work=work_id)
             .order_by("-rank", "-edition_rank")

bookwyrm/forms/books.py

@@ -1,8 +1,9 @@
 """ using django model forms """
 from django import forms
 
+from file_resubmit.widgets import ResubmitImageWidget
+
 from bookwyrm import models
-from bookwyrm.models.fields import ClearableFileInputWithWarning
 from .custom_form import CustomForm
 from .widgets import ArrayWidget, SelectDateWidget, Select
 
@@ -70,9 +71,7 @@ class Meta:
             "published_date": SelectDateWidget(
                 attrs={"aria-describedby": "desc_published_date"}
             ),
-            "cover": ClearableFileInputWithWarning(
-                attrs={"aria-describedby": "desc_cover"}
-            ),
+            "cover": ResubmitImageWidget(attrs={"aria-describedby": "desc_cover"}),
             "physical_format": Select(
                 attrs={"aria-describedby": "desc_physical_format"}
             ),

bookwyrm/forms/forms.py

@@ -25,6 +25,10 @@ class ImportForm(forms.Form):
     csv_file = forms.FileField()
 
 
+class ImportUserForm(forms.Form):
+    archive_file = forms.FileField()
+
+
 class ShelfForm(CustomForm):
     class Meta:
         model = models.Shelf

bookwyrm/importers/__init__.py

@@ -1,6 +1,7 @@
 """ import classes """
 
 from .importer import Importer
+from .bookwyrm_import import BookwyrmImporter
 from .calibre_import import CalibreImporter
 from .goodreads_import import GoodreadsImporter
 from .librarything_import import LibrarythingImporter

bookwyrm/importers/bookwyrm_import.py

@@ -0,0 +1,24 @@
+"""Import data from Bookwyrm export files"""
+from django.http import QueryDict
+
+from bookwyrm.models import User
+from bookwyrm.models.bookwyrm_import_job import BookwyrmImportJob
+
+
+class BookwyrmImporter:
+    """Import a Bookwyrm User export file.
+    This is kind of a combination of an importer and a connector.
+    """
+
+    # pylint: disable=no-self-use
+    def process_import(
+        self, user: User, archive_file: bytes, settings: QueryDict
+    ) -> BookwyrmImportJob:
+        """import user data from a Bookwyrm export file"""
+
+        required = [k for k in settings if settings.get(k) == "on"]
+
+        job = BookwyrmImportJob.objects.create(
+            user=user, archive_file=archive_file, required=required
+        )
+        return job

bookwyrm/management/commands/erase_deleted_user_data.py

@@ -0,0 +1,43 @@
+""" Erase any data stored about deleted users """
+import sys
+from django.core.management.base import BaseCommand, CommandError
+from bookwyrm import models
+from bookwyrm.models.user import erase_user_data
+
+# pylint: disable=missing-function-docstring
+class Command(BaseCommand):
+    """command-line options"""
+
+    help = "Remove Two Factor Authorisation from user"
+
+    def add_arguments(self, parser):  # pylint: disable=no-self-use
+        parser.add_argument(
+            "--dryrun",
+            action="store_true",
+            help="Preview users to be cleared without altering the database",
+        )
+
+    def handle(self, *args, **options):  # pylint: disable=unused-argument
+
+        # Check for anything fishy
+        bad_state = models.User.objects.filter(is_deleted=True, is_active=True)
+        if bad_state.exists():
+            raise CommandError(
+                f"{bad_state.count()} user(s) marked as both active and deleted"
+            )
+
+        deleted_users = models.User.objects.filter(is_deleted=True)
+        self.stdout.write(f"Found {deleted_users.count()} deleted users")
+        if options["dryrun"]:
+            self.stdout.write("\n".join(u.username for u in deleted_users[:5]))
+            if deleted_users.count() > 5:
+                self.stdout.write("... and more")
+            sys.exit()
+
+        self.stdout.write("Erasing user data:")
+        for user_id in deleted_users.values_list("id", flat=True):
+            erase_user_data.delay(user_id)
+            self.stdout.write(".", ending="")
+
+        self.stdout.write("")
+        self.stdout.write("Tasks created successfully")

bookwyrm/middleware/__init__.py

@@ -1,3 +1,4 @@
 """ look at all this nice middleware! """
 from .timezone_middleware import TimezoneMiddleware
 from .ip_middleware import IPBlocklistMiddleware
+from .file_too_big import FileTooBig

bookwyrm/middleware/file_too_big.py

@@ -0,0 +1,30 @@
+"""Middleware to display a custom 413 error page"""
+
+from django.http import HttpResponse
+from django.shortcuts import render
+from django.core.exceptions import RequestDataTooBig
+
+
+class FileTooBig:
+    """Middleware to display a custom page when a
+    RequestDataTooBig exception is thrown"""
+
+    def __init__(self, get_response):
+        """boilerplate __init__ from Django docs"""
+
+        self.get_response = get_response
+
+    def __call__(self, request):
+        """If RequestDataTooBig is thrown, render the 413 error page"""
+
+        try:
+            body = request.body  # pylint: disable=unused-variable
+
+        except RequestDataTooBig:
+
+            rendered = render(request, "413.html")
+            response = HttpResponse(rendered)
+            return response
+
+        response = self.get_response(request)
+        return response

bookwyrm/migrations/0184_auto_20231106_0421.py

@@ -22,17 +22,6 @@ def update_deleted_users(apps, schema_editor):
     ).update(is_deleted=True)
 
 
-def erase_deleted_user_data(apps, schema_editor):
-    """Retroactively clear user data"""
-    for user in User.objects.filter(is_deleted=True):
-        user.erase_user_data()
-        user.save(
-            broadcast=False,
-            update_fields=["email", "avatar", "preview_image", "summary", "name"],
-        )
-        user.erase_user_statuses(broadcast=False)
-
-
 class Migration(migrations.Migration):
 
     dependencies = [
@@ -43,7 +32,4 @@ class Migration(migrations.Migration):
         migrations.RunPython(
             update_deleted_users, reverse_code=migrations.RunPython.noop
         ),
-        migrations.RunPython(
-            erase_deleted_user_data, reverse_code=migrations.RunPython.noop
-        ),
     ]