Update Packages and Fix minor Issues

.github/dependabot.yml

@@ -1,13 +1,5 @@
 version: 2
 updates:
-  # Enable version updates for npm
-  - package-ecosystem: "npm"
-    # Look for `package.json` and `lock` files in the `root` directory
-    directory: "/docs"
-    # Check the npm registry for updates every day (weekdays)
-    schedule:
-      interval: "weekly"
-
   # Enable version updates for Docker
   - package-ecosystem: "docker"
     # Look for a `Dockerfile` in the `root` directory

.github/workflows/pr-security.yml

@@ -88,7 +88,7 @@ jobs:
           output: 'trivy-results.sarif'
           limit-severities-for-sarif: true
           severity: 'HIGH,CRITICAL'
-          scanners: 'vuln,config,secret'
+          scanners: 'vuln,misconfig,secret'
           skip-dirs: '.vscode,docs'
           exit-code: '1'
       - name: Upload Trivy scan results to GitHub Security tab

src/base_requirements.in

@@ -1 +1 @@
-pip==24.0
+pip==24.2

src/base_requirements.txt

@@ -1,12 +1,12 @@
 #
-# This file is autogenerated by pip-compile with Python 3.11
+# This file is autogenerated by pip-compile with Python 3.10
 # by the following command:
 #
 #    pip-compile --allow-unsafe --generate-hashes base_requirements.in
 #
 
 # The following packages are considered to be unsafe in a requirements file:
-pip==24.0 \
-    --hash=sha256:ba0d021a166865d2265246961bec0152ff124de910c5cc39f1156ce3fa7c69dc \
-    --hash=sha256:ea9bd1a847e8c5774a5777bb398c19e80bcd4e2aa16a4b301b718fe6f593aba2
+pip==24.2 \
+    --hash=sha256:2cd581cf58ab7fcfca4ce8efa6dcacd0de5bf8d0a3eb9ec927e07405f4d9e2a2 \
+    --hash=sha256:5b5e490b5e9cb275c879595064adce9ebd31b854e3e803740b72f9ccf34a45b8
     # via -r base_requirements.in

src/core/operators/audio_vec_embedding_requirements.in

@@ -1,6 +1,11 @@
 numpy==1.26.4
 librosa==0.10.2.post1
 panns-inference==0.1.1
-torch==2.2.0+cpu
-memray==1.12.0
-pyinstrument==4.6.2
+# torch==2.2.0+cpu
+torch==2.4.1
+memray==1.14.0
+pyinstrument==4.7.3
+# not direct requirments, but added because of pip-audit report
+certifi==2024.7.4
+urllib3==2.2.2
+scikit-learn==1.5.0

src/core/operators/test_classify_video_zero_shot.py

@@ -16,7 +16,7 @@ def tearDownClass(cls):
         # delete config files
         pass
 
-    # @skip
+    @skip
     def test_sample_video_from_disk(self):
         video_path = VideoFactory.make_from_file_on_disk(
             r"core/operators/sample_data/sample-cat-video.mp4"

src/core/operators/test_vid_vec_rep_clip.py

@@ -16,7 +16,7 @@ def tearDownClass(cls):
         # delete config files
         pass
 
-    # @skip
+    @skip
     def test_sample_video_from_disk(self):
         video_path = VideoFactory.make_from_file_on_disk(
             r"core/operators/sample_data/sample-cat-video.mp4"

src/core/operators/vid_vec_rep_resnet_requirements.in

@@ -1,8 +1,12 @@
-torch==2.2.0+cpu
-torchvision==0.17.0+cpu
-numpy==1.26.4
-Pillow==10.3.0
-scipy==1.11.4
-opencv-python-headless==4.9.0.80
-memray==1.12.0
-pyinstrument==4.6.2
+# torch==2.2.0+cpu
+# torchvision==0.17.0+cpu
+numpy==2.1.1
+torch==2.4.1
+Pillow==10.4.0
+scipy==1.14.1
+opencv-python-headless==4.10.0.84
+memray==1.14.0
+pyinstrument==4.7.3
+# not direct requirments, but added because of pip-audit report
+certifi==2024.7.4
+urllib3==2.2.2

src/core/store/es_vec.py

@@ -84,7 +84,7 @@ def refresh(self):
             self.client.indices.refresh()
 
     def find(self, index_name, vec):
-        if type(vec) == np.ndarray:
+        if isinstance(vec, np.ndarray):
             vec = vec.tolist()
 
         calculation = ""

src/requirements.in

@@ -1,20 +1,23 @@
 flask==3.0.3
-Flask-Cors==4.0.1
-Pillow==10.3.0
-elasticsearch==8.11.1
+Flask-Cors==5.0.0
+Pillow==10.4.0
+elasticsearch==8.15.1
 wget==3.2
 pika==1.3.2
-python-dotenv==1.0.0
+python-dotenv==1.0.1
 click==8.1.7
-pytest==7.4.4
-PyYAML==6.0.1
+pytest==8.3.3
+PyYAML==6.0.2
 dacite==1.8.1
-memray==1.12.0  # dev
-pyinstrument==4.6.2
-numpy==1.26.4
+memray==1.14.0  # dev
+pyinstrument==4.7.3
+numpy==2.1.1
 requests==2.32.2
-locust==2.27.0
-nose2==0.14.1
+locust==2.31.5
+nose2==0.15.1
 psycopg2-binary==2.9.9
-boto3==1.34.64
-pydub==0.25.1
+boto3==1.35.16
+pydub==0.25.1
+# not direct requirments, but added because of pip-audit report
+certifi==2024.7.4
+urllib3==2.2.2

src/tests/core/store/test_es_vec.py

@@ -3,7 +3,7 @@
 import requests
 from requests.exceptions import ConnectTimeout
 from core.store.es_vec import ES
-from core.config import StoreConfig, StoreParameters
+from core.config import StoreConfig, StoreEntity, StoreESParameters
 from core.models.media import MediaType
 import pprint
 from datetime import datetime
@@ -33,15 +33,19 @@ def setUpClass(cls) -> None:
                 "audio_index_name": "test_audio",
             }
             cls.param = StoreConfig(
-                label="test",
-                type="es",
-                parameters=StoreParameters(
-                    host_name=param_dict["host_name"],
-                    image_index_name=param_dict["image_index_name"],
-                    text_index_name=param_dict["text_index_name"],
-                    video_index_name=param_dict["video_index_name"],
-                    audio_index_name=param_dict["audio_index_name"],
-                ),
+                entities=[
+                    StoreEntity(
+                        label="test",
+                        type="es",
+                        parameters=StoreESParameters(
+                            host_name=param_dict["host_name"],
+                            image_index_name=param_dict["image_index_name"],
+                            text_index_name=param_dict["text_index_name"],
+                            video_index_name=param_dict["video_index_name"],
+                            audio_index_name=param_dict["audio_index_name"],
+                        ),
+                    )
+                ]
             )
         except ConnectTimeout:
             print('Request has timed out')
@@ -56,8 +60,8 @@ def tearDownClass(cls) -> None:
 
     # @skip
     def test_create_indices(self):
-        print(self.param)
-        es = ES(self.param)
+        print(self.param.entities[0])
+        es = ES(self.param.entities[0])
         es.connect()
         es.optionally_create_index()
         indices = es.get_indices()
@@ -96,7 +100,7 @@ def test_store_image(self):
 
     # @skip
     def test_store_and_search_vectors(self):
-        es = ES(self.param)
+        es = ES(self.param.entities[0])
         es.connect()
         es.optionally_create_index()
         vec = np.random.randn(512).tolist()
@@ -133,7 +137,7 @@ def test_find_by_metadata_field(self):
 
     def delete_indices(self):
         print("DELETING INDICES")
-        es = ES(self.param)
+        es = ES(self.param.entities[0])
         es.connect()
         es.delete_indices()
         print("INDICES DELETED")

src/worker/clustering_media/clustering_media_worker.py

@@ -3,7 +3,8 @@
 from core.operators import audio_vec_embedding_clap
 from core.operators import vid_vec_rep_clip, classify_video_zero_shot
 from core.operators import cluster_embeddings, dimension_reduction
-import requests, json
+import requests
+import json
 from core.models.media_factory import AudioFactory, VideoFactory
 from time import sleep