wbnews 1.54.23__tar.gz → 1.59.5__tar.gz

{wbnews-1.54.23 → wbnews-1.59.5}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: wbnews
-Version: 1.54.23
+Version: 1.59.5
 Summary: A workbench module for managing news.
 Author-email: Christopher Wittlinger <c.wittlinger@stainly.com>
 Requires-Dist: feedparser==6.*

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/filters/news.py

@@ -4,7 +4,7 @@ from wbnews.models import News, NewsRelationship, NewsSource
 
 
 class NewsFilterSet(wb_filters.FilterSet):
-    datetime = wb_filters.DateTimeRangeFilter(method=wb_filters.DateRangeFilter.base_date_range_filter_method)
+    datetime = wb_filters.DateTimeRangeFilter()
 
     class Meta:
         model = News

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/import_export/parsers/emails/utils.py

@@ -22,10 +22,10 @@ class EmlContentParser:
         html = self.get_html(self.message)
         return html.decode(self.encoding) if html else None
 
-    def get_html(cls, parsed: message.Message) -> bytes | None:
+    def get_html(self, parsed: message.Message) -> bytes | None:
         if parsed.is_multipart():
             for item in parsed.get_payload():  # type:message.Message
-                if html := cls.get_html(item):
+                if html := self.get_html(item):
                     return html
         elif parsed.get_content_type() == "text/html":
             return parsed.get_payload(decode=True)

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/models/news.py

@@ -14,6 +14,7 @@ from slugify import slugify
 from wbcore.contrib.ai.llm.decorators import llm
 from wbcore.contrib.io.mixins import ImportMixin
 from wbcore.models import WBModel
+from wbcore.workers import Queue
 
 from wbnews.import_export.handlers.news import NewsImportHandler
 from wbnews.models.llm.cleaned_news import clean_news_config, summarized_news_config
@@ -23,7 +24,7 @@ from wbnews.signals import create_news_relationships
 from ..utils import detect_near_duplicates
 
 
-@shared_task
+@shared_task(queue=Queue.DEFAULT.value)
 def create_relationship(chain_results: list[list[dict[str, Any]]], news_id: int):
     objs = []
     for relationships in chain_results:
@@ -81,7 +82,8 @@ class News(ImportMixin, WBModel):
         """
         tasks = []
         for sender, task_signature in create_news_relationships.send(sender=News, instance=self):
-            assert isinstance(task_signature, Signature), self.errors["relationship_signal"].format(sender)
+            if not isinstance(task_signature, Signature):
+                raise AssertionError(self.errors["relationship_signal"].format(sender))
             tasks.append(task_signature)
         if tasks:
             res = chord(tasks, create_relationship.s(self.id))

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/tasks.py

@@ -1,11 +1,12 @@
 from datetime import date, timedelta
 
 from celery import shared_task
+from wbcore.workers import Queue
 
 from wbnews.models import News
 
 
-@shared_task()
+@shared_task(queue=Queue.BACKGROUND.value)
 def handle_daily_news_duplicates(
     task_date: date | None = None,
     day_interval: int = 7,

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/tests/parsers/test_emails.py

@@ -1,5 +1,4 @@
-from unittest import mock
-from unittest.mock import patch, PropertyMock
+from unittest.mock import PropertyMock, patch
 
 import pytest
 
@@ -7,19 +6,20 @@ from wbnews.import_export.parsers.emails.utils import EmlContentParser
 
 
 class TestEmlContentParser:
-
     @pytest.fixture
     def content_parser(self):
-        parser = EmlContentParser(b'')
+        parser = EmlContentParser(b"")
         parser.message = {"From": "main@acme.com"}
         return parser
 
     @patch.object(EmlContentParser, "text", new_callable=PropertyMock)
     def test_source_from_in_text(self, mock_text, content_parser):
-        mock_text.return_value = "some random email content with a From field From: source name <email@test.com> and the rest of the email"
+        mock_text.return_value = (
+            "some random email content with a From field From: source name <email@test.com> and the rest of the email"
+        )
         assert content_parser.source == {"title": "Source Name", "endpoint": "email@test.com", "type": "EMAIL"}
 
     @patch.object(EmlContentParser, "text", new_callable=PropertyMock)
-    def test_source_from_in_text(self, mock_text, content_parser):
+    def test_source_from_in_text_alt(self, mock_text, content_parser):
         mock_text.return_value = "some random email content without a From field"
         assert content_parser.source == {"title": "Acme.Com", "endpoint": "main@acme.com", "type": "EMAIL"}

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/tests/test_models.py

@@ -1,14 +1,15 @@
-from datetime import timedelta, datetime, timezone
-from django.utils import timezone as django_timezone
+from datetime import timedelta, timezone
 from unittest.mock import patch
 
 import pytest
+from django.utils import timezone as django_timezone
+from faker import Faker
 
 from wbnews.models import News, NewsSource
-from faker import Faker
 
 fake = Faker()
 
+
 @pytest.mark.django_db
 class TestSource:
     @pytest.mark.parametrize("news_source__title", ["source1"])
@@ -19,17 +20,19 @@ class TestSource:
         ns1 = news_source_factory.create()
         ns2 = news_source_factory.create()
 
-        assert NewsSource.source_dict_to_model({"id": ns1.id, "identifier": ns2.identifier}) == ns1 # priority to "id"
-        assert NewsSource.source_dict_to_model({"endpoint": ns1.endpoint, "identifier": ns2.identifier}) == ns2 # priority to "identifier"
-        assert NewsSource.source_dict_to_model({"endpoint": ns2.endpoint}) == ns2 # exact match on endpoint
+        assert NewsSource.source_dict_to_model({"id": ns1.id, "identifier": ns2.identifier}) == ns1  # priority to "id"
+        assert (
+            NewsSource.source_dict_to_model({"endpoint": ns1.endpoint, "identifier": ns2.identifier}) == ns2
+        )  # priority to "identifier"
+        assert NewsSource.source_dict_to_model({"endpoint": ns2.endpoint}) == ns2  # exact match on endpoint
 
         ns1.endpoint = ".*@test.com"
         ns1.save()
-        assert NewsSource.source_dict_to_model({"endpoint": "abc@test.com"}) == ns1 # regex match on endpoint
+        assert NewsSource.source_dict_to_model({"endpoint": "abc@test.com"}) == ns1  # regex match on endpoint
 
         new_source = NewsSource.source_dict_to_model({"endpoint": "abc@main_source.com", "title": "New Source"})
         assert new_source not in [ns1, ns2]
-        assert new_source.endpoint == ".*@main_source\.com"
+        assert new_source.endpoint == r".*@main_source\.com"
         assert new_source.title == "New Source"
         assert new_source.author == "Main Source"
 
@@ -45,24 +48,31 @@ class TestNews:
 
     def test_get_default_guid(self):
         assert News.get_default_guid("This is a title", None) == "this-is-a-title"
-        assert News.get_default_guid("This is a title", "http://mylink.com") == "http://mylink.com" # link takes precendence
+        assert (
+            News.get_default_guid("This is a title", "http://mylink.com") == "http://mylink.com"
+        )  # link takes precendence
         assert News.get_default_guid("a" * 24, None, max_length=20) == "a" * 20
 
     def test_future_news(self, news_factory):
         # ensure a future datetime always default to now
         now = django_timezone.now()
         future_news = news_factory.create(datetime=now + timedelta(days=1))
-        assert (future_news.datetime - now).seconds < 1 # we do that to account for clock difference
+        assert (future_news.datetime - now).seconds < 1  # we do that to account for clock difference
 
     @patch("wbnews.models.news.detect_near_duplicates")
     def test_handle_duplicates(self, mock_fct, news_factory):
         val_date = fake.date_time(tzinfo=timezone.utc)
-        n0 = news_factory.create(datetime=val_date - timedelta(days=1)) # we exclude this news from the duplicate search
+        n0 = news_factory.create(
+            datetime=val_date - timedelta(days=1)
+        )  # we exclude this news from the duplicate search
         n1 = news_factory.create(datetime=val_date)
         n2 = news_factory.create(datetime=val_date)
         n3 = news_factory.create(datetime=val_date)
 
-        mock_fct.return_value = [n0.id, n3.id] # n0 is considered as duplicate but does not fall within the specified daterange so it will not be marked
+        mock_fct.return_value = [
+            n0.id,
+            n3.id,
+        ]  # n0 is considered as duplicate but does not fall within the specified daterange so it will not be marked
         News.handle_duplicates(val_date, val_date)
 
         n3.refresh_from_db()

{wbnews-1.54.23 → wbnews-1.59.5}/wbnews/utils.py

@@ -11,7 +11,7 @@ logger = logging.getLogger("news")
 
 def _get_similarity_matrix_df(data: dict[int, str]) -> pd.DataFrame:
     # Convert texts to TF-IDF vectors
-    ids, texts = zip(*data.items())
+    ids, texts = zip(*data.items(), strict=False)
     vectorizer = TfidfVectorizer()
     tfidf_matrix = vectorizer.fit_transform(texts)
     # Compute pairwise cosine similarity...