Add mypy checks.

mypy checks have been added to the gitlab pipeline.
They can be triggered via `make mypy`.

Some (minor) fixes to error checks are inlcuded.

Makefile

@@ -26,7 +26,7 @@ ima-venv:
 	make -C ${ima_home} venv
 
 test:
-	cd  ${ima_home}; make test
+	cd  ${ima_home}; make mypy; make test
 
 archive: ima-venv
 	tar cvz --exclude=".*" -f ${gitlab_package_archive} .

image-matching/Makefile

@@ -26,6 +26,11 @@ venv:	${pip_requirements}
 							conda install conda-pack; \
 							conda-pack -n ${venv} --format ${venv_archive_format}"
 
+mypy: ${pip_requirements_test}
+	${DOCKER_CMD} bash -c "export CONDA_ALWAYS_YES=true; ${CONDA_CMD}; \
+    					pip install -r ${pip_requirements_test}; \
+    					mypy spark"
+
 test:   ${pip_requirements_test}
 	${DOCKER_CMD} bash -c "export CONDA_ALWAYS_YES=true; ${CONDA_CMD}; \
 					conda install openjdk pyspark==${pyspark_version}; \

image-matching/requirements-test.txt

 pytest==6.2.2
 pytest-spark==0.6.0
 pytest-cov==2.10.1
-flake8==3.8.4
\ No newline at end of file
+flake8==3.8.4
+mypy==0.910
+pyspark-stubs==2.4.0post10

image-matching/spark/raw2parquet.py

 from pyspark.sql import SparkSession
 from pyspark.sql import functions as F
-from schema import CsvDataset
+from .schema import CsvDataset
 
 import argparse
 
@@ -29,7 +29,7 @@ if __name__ == "__main__":
 
     csv_df = (
         (
-            spark.read.options(delimiter="\t", header=False, escape='"')
+            spark.read.options(delimiter="\t", header="false", escape='"')
             .schema(CsvDataset.schema)
             .csv(source)
         )

image-matching/spark/transform.py

@@ -2,8 +2,8 @@ from pyspark.sql import SparkSession
 from pyspark.sql import Column, DataFrame
 from pyspark.sql import functions as F
 from pyspark.sql.types import IntegerType
-from schema import RawDataset
-from instances_to_filter import InstancesToFilter
+from .schema import RawDataset
+from .instances_to_filter import InstancesToFilter
 
 import argparse
 import uuid