Feature: added Result.run_id (#141)

* added Result.run_id

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* added run_id to _Metric._add_to_database

* added migration

* got run_id from envvar

* excluded migrations

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

.gitignore

@@ -133,3 +133,4 @@ dmypy.json
 
 # OSX hidden files
 .DS_Store
+.idea

Dockerfile

@@ -16,8 +16,6 @@ RUN apt-get install libpq-dev postgresql-client -y
 COPY requirements.txt requirements.txt
 RUN pip install -U pip wheel
 RUN pip install -r requirements.txt
-COPY alembic.ini /code/
-COPY migrations /code/migrations
 COPY scripts /code/scripts
 COPY --from=builder /dist/*.whl /dist/
 RUN pip install /dist/*.whl

README.md

@@ -43,6 +43,14 @@ or multiple dataframes.
 >>> p.plot_dataset([df1, df2])
 ```
 
+### Migrations
+
+`insight` populates the results to the Postgres database configured by environment variables. To run migrations against it, simply:
+
+```bash
+insight-migrations
+```
+
 <div align="center">
 <img alt="distribution plots" src="https://user-images.githubusercontent.com/13236749/181318624-95ad8702-5113-4a9e-8dd4-32f4a3dc8c9f.png" width="85%">
 </div>

docker-compose.yaml

@@ -7,24 +7,27 @@ services:
           - postgres
         tty: true
         environment:
-          - POSTGRES_USER
-          - POSTGRES_PASSWORD
-          - POSTGRES_HOST
-          - POSTGRES_PORT
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_HOST: postgres
+          POSTGRES_PORT: 5432
+        volumes:
+          - ./src:/code/src
     postgres:
         image: postgres:latest
         restart: always
         environment:
-          - POSTGRES_USER
-          - POSTGRES_PASSWORD
-          - POSTGRES_HOST
-          - POSTGRES_PORT
+          POSTGRES_USER: postgres
+          POSTGRES_PASSWORD: postgres
+          POSTGRES_HOST: 0.0.0.0
+          POSTGRES_PORT: 5432
+        command: ["postgres", "-c", "log_statement=all"]
         logging:
           options:
             max-size: 10m
             max-file: "3"
         ports:
-          - "5432:${POSTGRES_PORT}"
+          - "5432:5432"
         volumes:
           - "./postgres-data:/var/lib/postgresql/data"
 ...

requirements.txt

@@ -1,6 +1,29 @@
-alembic == 1.7.6
-numpy == 1.21.6
-pandas == 1.3.5
-psycopg2 == 2.9.2
-SQLAlchemy[mypy] == 1.4.40
-synthesized_datasets == 0.6
+alembic==1.7.6
+contourpy==1.1.0
+cycler==0.11.0
+fonttools==4.42.1
+greenlet==2.0.2
+importlib-resources==6.0.1
+kiwisolver==1.4.5
+Mako==1.2.4
+MarkupSafe==2.1.3
+matplotlib==3.7.3
+mypy==1.5.1
+mypy-extensions==1.0.0
+numpy==1.21.6
+packaging==23.1
+pandas==1.3.5
+Pillow==10.0.0
+psycopg2==2.9.2
+pyparsing==3.1.1
+python-dateutil==2.8.2
+pytz==2023.3.post1
+scipy==1.10.1
+seaborn==0.12.2
+six==1.16.0
+SQLAlchemy==1.4.40
+sqlalchemy2-stubs==0.0.2a35
+synthesized-datasets==0.6
+tomli==2.0.1
+typing_extensions==4.7.1
+zipp==3.16.2

sonar-project.properties

@@ -4,6 +4,7 @@ sonar.projectName=synthesized-insight
 sonar.sourceEncoding=UTF-8
 
 sonar.sources=src/insight/
+sonar.exclusions=src/insight/alembic/**/*
 sonar.tests=tests/
 
 sonar.python.version=3.6, 3.7, 3.8, 3.9

src/insight/alembic/main.py

@@ -1,14 +1,16 @@
 # Python script that will apply the migrations up to head
 import os
+import sys
 
 import alembic.config
 
 here = os.path.dirname(os.path.abspath(__file__))
 
+args = sys.argv[1:]
+args = args if len(args) > 0 else ["upgrade", "head"] # default
 alembic_args = [
     '-c', os.path.join(here, 'alembic.ini'),
-    'upgrade', 'head'
-]
+] + args
 
 
 def main():

src/insight/alembic/versions/d2198fd60b0e_added_result_run_id.py

@@ -0,0 +1,33 @@
+"""Added Result.run_id
+
+Revision ID: d2198fd60b0e
+Revises: 9aca5ae68ff5
+Create Date: 2023-09-14 12:25:17.878689
+
+"""
+import sqlalchemy as sa
+from alembic import op
+
+# revision identifiers, used by Alembic.
+revision = 'd2198fd60b0e'
+down_revision = '9aca5ae68ff5'
+branch_labels = None
+depends_on = None
+
+
+def upgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.add_column('result', sa.Column('run_id', sa.VARCHAR(length=50), nullable=False))
+    op.alter_column('version', 'name',
+               existing_type=sa.VARCHAR(length=50),
+               nullable=False)
+    # ### end Alembic commands ###
+
+
+def downgrade() -> None:
+    # ### commands auto generated by Alembic - please adjust! ###
+    op.alter_column('version', 'name',
+               existing_type=sa.VARCHAR(length=50),
+               nullable=True)
+    op.drop_column('result', 'run_id')
+    # ### end Alembic commands ###

src/insight/database/schema.py

@@ -1,12 +1,9 @@
-from sqlalchemy import FLOAT, INTEGER, TIMESTAMP, VARCHAR, ForeignKey
-from sqlalchemy.orm import DeclarativeBase, Mapped, mapped_column, relationship
+from sqlalchemy import FLOAT, INTEGER, TIMESTAMP, VARCHAR, Column, ForeignKey
+from sqlalchemy.orm import Mapped, declarative_base, relationship
 from sqlalchemy.sql import func
 
-
-class Base(DeclarativeBase):
-    pass
-
-
+Base = declarative_base()
+mapped_column = Column
 
 class Dataset(Base):
     __tablename__ = "dataset"
@@ -44,6 +41,7 @@ class Result(Base):
     version_id = mapped_column(INTEGER, ForeignKey("version.id"))
     value = mapped_column(FLOAT)
     created_at = mapped_column(TIMESTAMP, default=func.now())
+    run_id = mapped_column(VARCHAR(50), nullable=True, default=None)
 
     metric: Mapped[Metric] = relationship("Metric")
     dataset: Mapped[Dataset] = relationship("Dataset")

src/insight/metrics/base.py

@@ -77,10 +77,12 @@ def _add_to_database(self,
             dataset_rows: Number of rows in the dataset.
             dataset_cols: Number of column in the dataset.
             category: The category of the metric.
-            version: The version on which the test was run.
+
+        `version` and `run_id` are taken from `VERSION` and `RUN_ID` envvars.
         """
         version = os.getenv("VERSION")
         version = version if version else "Unversioned"
+        run_id = os.getenv("RUN_ID")
 
         if model is None or utils is None:
             raise ModuleNotFoundError("The database module is not available. Please install it using the command: pip install 'insight[db]'")
@@ -98,7 +100,9 @@ def _add_to_database(self,
             metric_id = utils.get_metric_id(self.name, session, category=category)
             version_id = utils.get_version_id(version, session)
             dataset_id = utils.get_df_id(dataset_name, session, num_rows=dataset_rows, num_columns=dataset_cols)
-            result = model.Result(metric_id=metric_id, dataset_id=dataset_id, version_id=version_id, value=value)
+            result = model.Result(
+                metric_id=metric_id, dataset_id=dataset_id, version_id=version_id, value=value, run_id=run_id
+            )
             session.add(result)
             session.commit()