refactor: some cleanup on etl's code structure

etl/src/helpers/__init__.py

@@ -0,0 +1,5 @@
+"""Helper classes for the pipeline"""
+
+from helpers.upsert_products_to_pg import UpsertProductsToPg
+from helpers.read_from_csv import ReadFromCsv
+from helpers.process_rows import ProcessRows

etl/src/helpers/parsers/__init__.py

@@ -0,0 +1,4 @@
+"""Helper parser functions to extract and clean data from the input CSV file
+Only `parse_row` needs to be exported."""
+
+from helpers.parsers.parse_row import parse_row

etl/src/helpers/parsers/dimensions.py

@@ -6,7 +6,7 @@ import logging
 from typing import Dict, Optional
 import re
 
-from helpers.misc import convert_units
+from helpers.parsers.misc import convert_units
 
 
 UNIT_CONVERSIONS = {"inches": 2.54, "feet": 30.48, "cm": 1}

etl/src/helpers/parsers/parse_row.py

@@ -4,11 +4,11 @@ the destination database"""
 import logging
 from typing import TypedDict, Dict, Optional, List
 
-from helpers.parse_xml import parse_raw_specs
-from helpers.materials import parse_materials
-from helpers.origin import clean_origin_name
-from helpers.dimensions import parse_dimensions
-from helpers.weight import parse_weight, dimensional_weight
+from helpers.parsers.parse_xml import parse_raw_specs
+from helpers.parsers.materials import parse_materials
+from helpers.parsers.origin import clean_origin_name
+from helpers.parsers.dimensions import parse_dimensions
+from helpers.parsers.weight import parse_weight, dimensional_weight
 
 
 class CleanRow(TypedDict):

etl/src/helpers/parsers/weight.py

@@ -6,7 +6,7 @@ import logging
 from typing import Optional
 import re
 
-from helpers.misc import convert_units
+from helpers.parsers.misc import convert_units
 
 
 UNIT_CONVERSIONS = {"pounds": 453.592, "ounces": 28.3495, "g": 1, "kg": 1000}

etl/src/helpers/process_rows.py

@@ -0,0 +1,19 @@
+"""Module containing necessary functionality to write to the PostgreSQL sink"""
+
+import logging
+
+import apache_beam as beam
+
+from helpers.parsers import parse_row
+
+
+class ProcessRows(beam.DoFn):
+    """DoFn to process and parse rows from the input file into structured
+    dictionaries"""
+
+    # pylint: disable=abstract-method,arguments-differ
+    def process(self, element):
+        if (row := parse_row(element)) is not None:
+            yield row
+        else:
+            logging.warning("could not successfully parse this row: %s", element)

etl/src/helpers/read_from_csv.py

@@ -0,0 +1,28 @@
+"""Module containing ReadFromCsv DoFn to create a PTransform to read from a CSV
+input file"""
+
+import io
+import logging
+import csv
+
+import apache_beam as beam
+from apache_beam.io.filesystems import FileSystems
+
+
+class ReadFromCsv(beam.DoFn):
+    """This custom DoFn will read from a CSV file and yield each row as a
+    dictionary where the row names are the keys and the cells are the values
+    """
+
+    # pylint: disable=abstract-method,arguments-differ
+    def process(self, element):
+        logging.info("reading from input file: %s", element)
+        with FileSystems.open(element) as file:
+            text_wrapper = io.TextIOWrapper(file)
+            reader = csv.reader(text_wrapper)
+            try:
+                header = next(reader)
+            except StopIteration:
+                return
+            for row in reader:
+                yield dict(zip(header, row))

etl/src/helpers/upsert_products_to_pg.py

@@ -1,36 +1,11 @@
-"""Module containing the IO parts of the pipeline"""
+"""Module containing necessary functionality to write to the PostgreSQL sink"""
 
-#!/usr/bin/env python
-
-import io
 import logging
-import csv
 from typing import Dict
 
 import apache_beam as beam
 import psycopg2
 
-from apache_beam.io.filesystems import FileSystems
-
-
-class ReadFromCsv(beam.DoFn):
-    """This custom DoFn will read from a CSV file and yield each row as a
-    dictionary where the row names are the keys and the cells are the values
-    """
-
-    # pylint: disable=abstract-method,arguments-differ
-    def process(self, element):
-        logging.info("reading from input file: %s", element)
-        with FileSystems.open(element) as file:
-            text_wrapper = io.TextIOWrapper(file)
-            reader = csv.reader(text_wrapper)
-            try:
-                header = next(reader)
-            except StopIteration:
-                return
-            for row in reader:
-                yield dict(zip(header, row))
-
 
 class WriteToPostgreSQL(beam.DoFn):
     """DoFn to write elements to a PostgreSQL database"""
@@ -68,8 +43,13 @@ class WriteToPostgreSQL(beam.DoFn):
     def process(self, element):
         if self.connection is not None:
             cursor = self.connection.cursor()
+            logging.info(
+                "inserting the following element into the database: %s", element
+            )
             self.execute_insert(element, cursor)
             cursor.close()
+        else:
+            logging.error("something went wrong with the connection to postresql")
 
     def teardown(self):
         if self.connection is not None:

etl/src/main.py

@@ -6,11 +6,9 @@ database
 import logging
 
 import apache_beam as beam
-
 from apache_beam.options.pipeline_options import PipelineOptions
 
-from helpers.data_io import ReadFromCsv, UpsertProductsToPg
-from helpers.parse_row import parse_row
+from helpers import UpsertProductsToPg, ReadFromCsv, ProcessRows
 
 
 class SustainabilityScoreOptions(PipelineOptions):
@@ -47,7 +45,7 @@ def main():
         pipeline \
             | beam.Create([opts.input]) \
             | beam.ParDo(ReadFromCsv()) \
-            | beam.Map(parse_row) \
+            | beam.ParDo(ProcessRows()) \
             | beam.ParDo(UpsertProductsToPg(
                 connection_details=pg_connection_details,
                 table=opts.pg_table,

etl/src/tests/parsers/test_convert_units.py

@@ -1,8 +1,8 @@
 """Test the `convert_units`"""
 
-from helpers.dimensions import UNIT_CONVERSIONS as dimension_unit_conversions
-from helpers.weight import UNIT_CONVERSIONS as weight_unit_conversions
-from helpers.misc import convert_units
+from helpers.parsers.dimensions import UNIT_CONVERSIONS as dimension_unit_conversions
+from helpers.parsers.weight import UNIT_CONVERSIONS as weight_unit_conversions
+from helpers.parsers.misc import convert_units
 
 
 def test_units_to_cm_inches():

etl/src/tests/parsers/test_dimensions.py

@@ -1,6 +1,6 @@
 """Test the `parse_dimensions` function and its helpers"""
 
-from helpers.dimensions import parse_dimensions, parse_dimensions_measure
+from helpers.parsers.dimensions import parse_dimensions, parse_dimensions_measure
 
 
 def test_none():

etl/src/tests/parsers/test_materials.py

@@ -1,6 +1,10 @@
 """Test the `parse_materials` function and its helpers"""
 
-from helpers.materials import parse_materials, clean_material_name, material_classifier
+from helpers.parsers.materials import (
+    parse_materials,
+    clean_material_name,
+    material_classifier,
+)
 
 
 def test_none():

etl/src/tests/parsers/test_origin.py

@@ -1,6 +1,6 @@
 """Test the `clean_material_name`"""
 
-from helpers.origin import clean_origin_name
+from helpers.parsers.origin import clean_origin_name
 
 
 def test_none():
@@ -46,4 +46,3 @@ def test_clean_origin_name5():
 def test_clean_origin_name6():
     """Test a sample input for clean_origin_name"""
     assert clean_origin_name(" made in the USA or imported") == "mixed"
-

etl/src/tests/parsers/test_parse_xml.py

@@ -2,7 +2,7 @@
 
 
 import xml.etree.ElementTree as ET
-from helpers.parse_xml import parse_raw_specs, iter_parse
+from helpers.parsers.parse_xml import parse_raw_specs, iter_parse
 
 
 def test_parse_raw_specs0():

etl/src/tests/parsers/test_weight.py

@@ -1,6 +1,6 @@
 """Test the `parse_weight`  and `dimensional_weight`"""
 
-from helpers.weight import parse_weight, dimensional_weight
+from helpers.parsers.weight import parse_weight, dimensional_weight
 
 
 def test_parse_weight_none():