ci: Define a clean workflow for argilla-server

.github/workflows/argilla-server.workflow.yml

@@ -0,0 +1,168 @@
+name: Build `argilla-server` package
+
+concurrency:
+  group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
+  cancel-in-progress: true
+
+on:
+  workflow_dispatch:
+
+  pull_request:
+    paths:
+      - 'argilla-server/**'
+    types:
+      - opened
+      - edited
+      - reopened
+      - synchronize
+      - ready_for_review
+
+jobs:
+
+  build:
+    name: Build `argilla-server` package
+    runs-on: ubuntu-latest
+
+    defaults:
+      run:
+        shell: bash -l {0}
+        working-directory: ./argilla-server
+
+    services:
+      search_engine:
+        image: docker.elastic.co/elasticsearch/elasticsearch:8.8.2
+        ports:
+          - 9200:9200
+        env:
+          discovery.type: single-node
+          xpack.security.enabled: false
+      # NOTE: postgres service will not be executed by default.
+      # It will only be executed if the postgresDockerImage input is provided.
+      postgres:
+        image: postgres:14
+        env:
+          POSTGRES_HOST: localhost
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_DB: argilla
+        options: >-
+          --health-cmd pg_isready
+          --health-interval 10s
+          --health-timeout 5s
+          --health-retries 5
+        ports:
+          - 5432:5432
+
+    env:
+      ARGILLA_ENABLE_TELEMETRY: 0
+
+    steps:
+      - name: Checkout Code 🛎
+        uses: actions/checkout@v4
+
+      - name: Setup PDM
+        uses: pdm-project/setup-pdm@v4
+        with:
+          python-version-file: argilla-server/pyproject.toml
+          cache-dependency-path: argilla-server/pdm.lock
+          cache: true
+
+      - name: Install dependencies
+        run: pdm install
+
+      - name: Run tests 📈
+        run: |
+          ARGILLA_DATABASE_URL=postgresql://postgres:postgres@localhost:5432/argilla
+          pdm test tests/unit -vs --cov=argilla_server --cov-report=xml:coverage.xml
+
+      - name: Upload test coverage
+        uses: codecov/codecov-action@v4
+        with:
+          file: coverage.xml
+
+      - name: Build package
+        run: |
+          pdm build
+
+      - name: Upload artifact
+        uses: actions/upload-artifact@v4
+        with:
+          name: argilla-server
+          path: argilla-server/dist
+
+  build_server_docker_image:
+    name: Build Argilla server docker image
+    uses: ./.github/workflows/argilla-server.build-push-docker.yml
+    if: github.event_name == 'pull_request' && github.event.pull_request.draft == false
+    needs:
+      - build
+    with:
+      download-python-package: true
+      image-name: argilla/argilla-server
+      dockerfile: argilla-server/docker/server/Dockerfile
+      context: argilla-server/docker/server
+      readme: argilla-server/README.md
+      platforms: linux/amd64,linux/arm64
+    secrets: inherit
+
+  build_quickstart_docker_image:
+    name: Build Argilla quickstart docker image
+    uses: ./.github/workflows/argilla-server.build-push-docker.yml
+    needs: build_server_docker_image
+    if: github.event_name == 'pull_request' && github.event.pull_request.draft == false
+    with:
+      download-python-package: false
+      image-name: argilla/argilla-quickstart
+      dockerfile: argilla-server/docker/quickstart/Dockerfile
+      context: argilla-server/docker/quickstart
+      readme: argilla-server/docker/quickstart/README.md
+      platforms: linux/amd64,linux/arm64
+      build-args: |
+        ARGILLA_VERSION=${{ needs.build_server_docker_image.outputs.version }}
+    secrets: inherit
+
+  # This job will publish argilla-server python package into PyPI repository
+  publish_release:
+    name: Publish Release
+    runs-on: ubuntu-latest
+    if: ${{ github.event_name == 'release' }}
+
+    needs:
+      - build
+      - build_server_docker_image
+      - build_quickstart_docker_image
+
+    permissions:
+      # This permission is needed for private repositories.
+      # contents: read
+      # IMPORTANT: this permission is mandatory for trusted publishing on PyPI
+      id-token: write
+
+    defaults:
+      run:
+        shell: bash -l {0}
+        working-directory: ./argilla-server
+
+    steps:
+      - name: Checkout Code 🛎
+        uses: actions/checkout@v4
+
+      - name: Download python package
+        uses: actions/download-artifact@v4
+        with:
+          name: argilla-server
+          path: argilla-server/dist
+
+      - name: Setup PDM
+        uses: pdm-project/setup-pdm@v4
+        with:
+          cache: true
+
+      - name: Publish Package to PyPI test environment 🥪
+        run: pdm publish --no-build --repository testpypi
+
+      - name: Test Installing 🍿
+        run: pip install --index-url https://test.pypi.org/simple --no-deps argilla-server==${GITHUB_REF#refs/*/v}
+
+      - name: Publish Package to PyPI 🥩
+        run: pdm publish --no-build

argilla-server/src/argilla_server/apis/v1/controllers/search.py

@@ -134,5 +134,4 @@ async def search(
     dataset_id: UUID,
     search_query: RecordsSearchQuery,
 ):
-
     return RecordsSearchResponse(query=search_query)

argilla-server/src/argilla_server/bulk/records_bulk.py

@@ -43,7 +43,6 @@
 
 
 class CreateRecordsBulk:
-
     def __init__(self, db: AsyncSession, search_engine: SearchEngine):
         self._db = db
         self._search_engine = search_engine
@@ -73,7 +72,6 @@ async def create_records_bulk(self, dataset: Dataset, bulk_create: RecordsBulkCr
         return RecordsBulk(items=records)
 
     async def _upsert_records_relationships(self, records: List[Record], records_create: List[RecordCreate]) -> None:
-
         records_and_suggestions = list(zip(records, [r.suggestions for r in records_create]))
         records_and_responses = list(zip(records, [r.responses for r in records_create]))
         records_and_vectors = list(zip(records, [r.vectors for r in records_create]))
@@ -87,7 +85,6 @@ async def _upsert_records_relationships(self, records: List[Record], records_cre
     async def _upsert_records_suggestions(
         self, records_and_suggestions: List[Tuple[Record, List[SuggestionCreate]]]
     ) -> List[Suggestion]:
-
         upsert_many_suggestions = []
         for idx, (record, suggestions) in enumerate(records_and_suggestions):
             try:
@@ -118,7 +115,6 @@ async def _upsert_records_suggestions(
     async def _upsert_records_responses(
         self, records_and_responses: List[Tuple[Record, List[UserResponseCreate]]]
     ) -> List[Response]:
-
         user_ids = [response.user_id for _, responses in records_and_responses for response in responses or []]
         users_by_id = await fetch_users_by_ids_as_dict(self._db, user_ids)
 
@@ -147,7 +143,6 @@ async def _upsert_records_responses(
     async def _upsert_records_vectors(
         self, records_and_vectors: List[Tuple[Record, Dict[str, List[float]]]]
     ) -> List[Vector]:
-
         upsert_many_vectors = []
         for idx, (record, vectors) in enumerate(records_and_vectors):
             try:
@@ -176,9 +171,7 @@ def _metadata_is_set(self, record_create: RecordCreate) -> bool:
 
 
 class UpsertRecordsBulk(CreateRecordsBulk):
-
     async def upsert_records_bulk(self, dataset: Dataset, bulk_upsert: RecordsBulkUpsert) -> RecordsBulkWithUpdateInfo:
-
         found_records = await self._fetch_existing_dataset_records(dataset, bulk_upsert.items)
         # found_records is passed to the validator to avoid querying the database again, but ideally, it should be
         # computed inside the validator
@@ -220,7 +213,6 @@ async def _fetch_existing_dataset_records(
         dataset: Dataset,
         records_upsert: List[RecordUpsert],
     ) -> Dict[Union[str, UUID], Record]:
-
         records_by_external_id = await fetch_records_by_external_ids_as_dict(
             self._db, dataset, [r.external_id for r in records_upsert]
         )

argilla-server/src/argilla_server/contexts/records.py

@@ -25,15 +25,13 @@
 async def list_dataset_records_by_ids(
     db: AsyncSession, dataset_id: UUID, record_ids: Sequence[UUID]
 ) -> Sequence[Record]:
-
     query = select(Record).filter(Record.id.in_(record_ids), Record.dataset_id == dataset_id)
     return (await db.execute(query)).unique().scalars().all()
 
 
 async def list_dataset_records_by_external_ids(
     db: AsyncSession, dataset_id: UUID, external_ids: Sequence[str]
 ) -> Sequence[Record]:
-
     query = (
         select(Record)
         .filter(Record.external_id.in_(external_ids), Record.dataset_id == dataset_id)

argilla-server/src/argilla_server/search_engine/commons.py

@@ -33,6 +33,7 @@
 )
 from argilla_server.search_engine.base import (
     AndFilter,
+    FieldFilterScope,
     Filter,
     FilterScope,
     FloatMetadataFilter,
@@ -43,6 +44,7 @@
     MetadataFilterScope,
     MetadataMetrics,
     Order,
+    QueryStringFilter,
     RangeFilter,
     RecordFilterScope,
     ResponseFilterScope,
@@ -56,8 +58,6 @@
     TermsMetadataMetrics,
     TextQuery,
     UserResponseStatusFilter,
-    FieldFilterScope,
-    QueryStringFilter,
 )
 
 ALL_RESPONSES_STATUSES_FIELD = "all_responses_statuses"

argilla-server/tests/profiling/__init__.py

@@ -0,0 +1,13 @@
+#  Copyright 2021-present, the Recognai S.L. team.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.

argilla-server/tests/profiling/conftest.py

@@ -1,5 +1,18 @@
-import pytest
+#  Copyright 2021-present, the Recognai S.L. team.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
 
+import pytest
 from argilla_server.settings import settings
 
 

argilla-server/tests/profiling/test_records_bulk.py

@@ -1,17 +1,31 @@
-from typing import Union
+#  Copyright 2021-present, the Recognai S.L. team.
+#
+#  Licensed under the Apache License, Version 2.0 (the "License");
+#  you may not use this file except in compliance with the License.
+#  You may obtain a copy of the License at
+#
+#      http://www.apache.org/licenses/LICENSE-2.0
+#
+#  Unless required by applicable law or agreed to in writing, software
+#  distributed under the License is distributed on an "AS IS" BASIS,
+#  WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+#  See the License for the specific language governing permissions and
+#  limitations under the License.
 
-import pytest
-from pyinstrument import Profiler
-from sqlalchemy.ext.asyncio import AsyncSession
+from typing import Union
 
 import argilla_server.apis.v1.handlers.datasets.records_bulk
+import pytest
 from argilla_server.contexts import datasets
 from argilla_server.models import Dataset, Question, VectorSettings
-from argilla_server.schemas.v1.records import RecordsCreate, RecordCreate
+from argilla_server.schemas.v1.records import RecordCreate, RecordsCreate
 from argilla_server.schemas.v1.records_bulk import RecordsBulkCreate
 from argilla_server.schemas.v1.responses import UserDraftResponseCreate
 from argilla_server.schemas.v1.suggestions import SuggestionCreate
 from argilla_server.search_engine import ElasticSearchEngine
+from pyinstrument import Profiler
+from sqlalchemy.ext.asyncio import AsyncSession
+
 from tests.factories import UserFactory
 from tests.unit.api.v1.datasets.records.records_bulk.test_dataset_records_bulk import TestDatasetRecordsBulk
 
@@ -30,9 +44,7 @@ def _get_dataset_vector_settings_by_name(dataset: Dataset, name: str) -> Union["
 
 @pytest.mark.asyncio
 class TestUpsertRecordsBulk:
-
     async def test_profiling_for_create_records(self, db: AsyncSession, elasticsearch_config: dict):
-
         engine = ElasticSearchEngine(config=elasticsearch_config, number_of_replicas=0, number_of_shards=1)
         dataset = await TestDatasetRecordsBulk().test_dataset()
 

argilla-server/tests/unit/api/v1/datasets/records/records_bulk/test_dataset_records_bulk.py

@@ -31,7 +31,6 @@
 
 @pytest.mark.asyncio
 class TestDatasetRecordsBulk:
-
     def url(self, dataset_id: UUID) -> str:
         return f"/api/v1/datasets/{dataset_id}/records/bulk"
 

argilla-server/tests/unit/api/v1/datasets/records/records_bulk/test_dataset_records_bulk_with_responses.py

@@ -34,7 +34,6 @@
 
 @pytest.mark.asyncio
 class TestDatasetRecordsBulkWithResponses:
-
     def url(self, dataset_id: UUID) -> str:
         return f"/api/v1/datasets/{dataset_id}/records/bulk"
 
@@ -361,7 +360,6 @@ async def _configure_dataset_fields(self, dataset: Dataset):
         await dataset.awaitable_attrs.fields
 
     async def _configure_dataset_questions(self, dataset: Dataset):
-
         await LabelSelectionQuestionFactory.create(
             dataset=dataset,
             name="label",

argilla-server/tests/unit/api/v1/datasets/records/records_bulk/test_dataset_records_bulk_with_suggestions.py

@@ -33,7 +33,6 @@
 
 @pytest.mark.asyncio
 class TestDatasetRecordsBulkWithSuggestions:
-
     def url(self, dataset_id: UUID) -> str:
         return f"/api/v1/datasets/{dataset_id}/records/bulk"
 

argilla-server/tests/unit/api/v1/datasets/records/records_bulk/test_dataset_records_bulk_with_vectors.py

@@ -31,7 +31,6 @@
 
 @pytest.mark.asyncio
 class TestDatasetRecordsBulkWithVectors:
-
     def url(self, dataset_id: UUID) -> str:
         return f"/api/v1/datasets/{dataset_id}/records/bulk"