Update to use poetry package manager (#165)

.github/workflows/pull_request.yml

@@ -15,17 +15,17 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: ${{ env.PYTHON_VERSION }}
-      - name: Install Pipenv
-        run: pip install pipenv==2022.9.24
-      - name: Cache virtualenv
-        id: cache-virtualenv
-        uses: actions/cache@v4
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
         with:
-          path: ~/.local/share/virtualenvs/
-          key: ${{ runner.os }}-${{ env.PYTHON_VERSION }}-virtualenvs-${{ hashFiles('Pipfile.lock') }}
+          version: 1.8.3
+          virtualenvs-create: true
+      - uses: actions/setup-python@v5
+        with:
+          python-version: ${{ env.PYTHON_VERSION }}
+          cache: 'poetry'
       - name: Install virtual environment
-        if: steps.cache-virtualenv.outputs.cache-hit != 'true'
-        run: pipenv install --dev
+        run: poetry install
       - name: Lint scripts
         run: make lint
   test:
@@ -37,10 +37,13 @@ jobs:
       - uses: actions/setup-python@v5
         with:
           python-version: ${{ env.PYTHON_VERSION }}
-      - name: Install pipenv
-        run: pip install pipenv==2023.8.22
+      - name: Install Poetry
+        uses: snok/install-poetry@v1
+        with:
+          version: 1.8.3
+          virtualenvs-create: true
       - name: Install virtual environment
-        run: pipenv install --dev
+        run: poetry install
       - name: Running unit tests
         run: make test
   docker-push:

Dockerfile

@@ -7,8 +7,11 @@ COPY . /benchmark
 RUN apt-get update && apt-get install -y gcc python3-dev
 
 # Install the required dependencies via pip
-RUN pip install pipenv==2022.9.24
-RUN pipenv install --deploy --system
+COPY pyproject.toml pyproject.toml
+COPY poetry.lock poetry.lock
+RUN pip install "poetry==1.8.3"
+RUN poetry config virtualenvs.create false
+RUN poetry install
 
 # Start Locust using LOCUS_OPTS environment variable
 ENTRYPOINT ["bash", "./docker_entrypoint.sh"]

Makefile

@@ -1,15 +1,15 @@
 lint: flake8
-	pipenv run black --check .
+	poetry run black --check .
 
 flake8:
-	pipenv run flake8 --max-complexity 10 --count
+	poetry run flake8 --max-complexity 10 --count
 
 format:
-	pipenv run isort .
-	pipenv run black .
+	poetry run isort .
+	poetry run black .
 
 run:
-	pipenv run ./run.sh requests/test_checkbox.json
+	poetry run ./run.sh requests/test_checkbox.json
 
 test:
-	pipenv run ./scripts/run_tests.sh
+	poetry run ./scripts/run_tests.sh

README.md

@@ -21,19 +21,19 @@ The benchmark consumes a requests JSON file that contains a list of HTTP request
 To run a benchmark, use:
 
 ```bash
-pipenv run ./run.sh <REQUESTS_JSON> <INCLUDE_SCHEMA_URL_IN_TOKEN: Optional> <HOST: Optional>
+poetry run ./run.sh <REQUESTS_JSON> <INCLUDE_SCHEMA_URL_IN_TOKEN: Optional> <HOST: Optional>
 ```
 e.g.
 ```bash
-pipenv run ./run.sh requests/test_checkbox.json
+poetry run ./run.sh requests/test_checkbox.json
 ```
 
 This will run 1 minute of locust requests with 1 user and no wait time between requests. The output files are `output_stats.csv`, `output_stats_history.csv` and `output_failures.csv`.
 
 For the web interface:
 
 ```bash
-REQUESTS_JSON=requests/test_checkbox.json HOST=http://localhost:5000 pipenv run locust
+REQUESTS_JSON=requests/test_checkbox.json HOST=http://localhost:5000 poetry run locust
 ```
 
 ## Configuration
@@ -57,11 +57,11 @@ Open the network inspector in Chrome or Firefox and ensure 'preserve log' is tic
 After the test is complete, right-click on one of the requests in the network inspector and save the log as a HAR file. To generate a requests file from the HAR file run:
 
 ```bash
-pipenv run python generate_requests.py <HAR_FILEPATH> <REQUESTS_FILEPATH> <SCHEMA_NAME>
+poetry run python generate_requests.py <HAR_FILEPATH> <REQUESTS_FILEPATH> <SCHEMA_NAME>
 ```
 e.g.
 ```bash
-pipenv run python generate_requests.py requests.har requests/test_checkbox.json test_checkbox
+poetry run python generate_requests.py requests.har requests/test_checkbox.json test_checkbox
 ```
 
 ## Dealing with repeating sections
@@ -202,21 +202,21 @@ export GOOGLE_APPLICATION_CREDENTIALS="<path_to_json_credentials_file>"
 
 To run the script and download results:
 ```bash
-OUTPUT_BUCKET="<bucket_name>" pipenv run python -m scripts.get_benchmark_results
+OUTPUT_BUCKET="<bucket_name>" poetry run python -m scripts.get_benchmark_results
 ```
 
 This script also accepts optional `NUMBER_OF_DAYS` and `OUTPUT_DIR` environment variables which allows the user to download a subset of results and set
 a specific output directory e.g.
 ```bash
-OUTPUT_BUCKET="<bucket_name>" NUMBER_OF_DAYS=<number_of_days> OUTPUT_DIR="<output_directory>" pipenv run python -m scripts.get_benchmark_results
+OUTPUT_BUCKET="<bucket_name>" NUMBER_OF_DAYS=<number_of_days> OUTPUT_DIR="<output_directory>" poetry run python -m scripts.get_benchmark_results
 ```
 
 ### Summarise the Daily Benchmark results
 You can get a breakdown of the average response times for a result set by doing:
 ```bash
 OUTPUT_DIR="outputs/daily-test" \
 OUTPUT_DATE="2020-01-01" \
-pipenv run python -m scripts.get_summary
+poetry run python -m scripts.get_summary
 ```
 
 This will output something like:
@@ -245,7 +245,7 @@ If `OUTPUT_DATE` is not provided, then it will output a summary for all results
 To get a breakdown of results for a stress test use the `get_aggregated_summary` script. This accepts a folder containing
 results as a parameter, and will provide aggregate totals at the folder level:
 ```bash
-OUTPUT_DIR="outputs/stress-test" pipenv run python -m scripts.get_aggregated_summary
+OUTPUT_DIR="outputs/stress-test" poetry run python -m scripts.get_aggregated_summary
 ```
 
 This will output something like:
@@ -275,7 +275,7 @@ For example, to visualise results for the last 7 days:
 ```bash
 OUTPUT_DIR="outputs/daily-test" \
 NUMBER_OF_DAYS="7" \
-pipenv run python -m scripts.visualise_results
+poetry run python -m scripts.visualise_results
 ```
 
 A line chart will be generated and saved as `performance_graph.png`.

ci/output-results-to-github.yaml

@@ -26,21 +26,21 @@ run:
 
       cd eq-survey-runner-benchmark
 
-      pip3 install pipenv
-      pipenv install --deploy
+      pip3 install poetry
+      poetry install
 
       # Get benchmark outputs
       OUTPUT_BUCKET="$OUTPUT_BUCKET" \
       OUTPUT_DIR="outputs" \
       NUMBER_OF_DAYS="$NUMBER_OF_DAYS" \
-      pipenv run python -m scripts.get_benchmark_results
+      poetry run python -m scripts.get_benchmark_results
 
       # Date to get summary for
       RUNTIME_DATE_STRING="$(date +'%Y-%m-%d')"
       SUMMARY=$(OUTPUT_DIR="outputs/${OUTPUT_DIR}" \
       OUTPUT_DATE="$RUNTIME_DATE_STRING" \
       OUTPUT_TO_GITHUB="True" \
-      pipenv run python -m scripts.get_summary)
+      poetry run python -m scripts.get_summary)
 
       # Post summary to Github
       cd ../

ci/output-results-to-slack.yaml

@@ -26,33 +26,33 @@ run:
 
       cd eq-survey-runner-benchmark
 
-      pip3 install pipenv
-      pipenv install --deploy
+      pip3 install poetry
+      poetry install
 
       # Get benchmark outputs
       OUTPUT_BUCKET="$OUTPUT_BUCKET" \
       OUTPUT_DIR="outputs" \
       NUMBER_OF_DAYS="$NUMBER_OF_DAYS" \
-      pipenv run python -m scripts.get_benchmark_results
+      poetry run python -m scripts.get_benchmark_results
 
       # Create performance graph
       OUTPUT_DIR="outputs/${OUTPUT_DIR}" \
       NUMBER_OF_DAYS="$NUMBER_OF_DAYS" \
-      pipenv run python -m scripts.visualise_results
+      poetry run python -m scripts.visualise_results
 
       # Date to get summary for
       RUNTIME_DATE_STRING="$(date +'%Y-%m-%d')"
       SUMMARY=$(OUTPUT_DIR="outputs/${OUTPUT_DIR}" \
       OUTPUT_DATE="$RUNTIME_DATE_STRING" \
-      pipenv run python -m scripts.get_summary)
+      poetry run python -m scripts.get_summary)
 
       # Post summary to Slack
       TITLE="Results For Latest Benchmark" \
       INITIAL_COMMENT="Latest Daily Performance Metrics" \
       CONTENT="$SUMMARY" \
-      pipenv run python -m scripts.slack_notification
+      poetry run python -m scripts.slack_notification
 
       # Post performance graph to Slack
       TITLE="Performance Graph" \
       ATTACHMENT_FILENAME="performance_graph.png" \
-      pipenv run python -m scripts.slack_notification
+      poetry run python -m scripts.slack_notification

doc/performance-investigations/README.md

@@ -51,12 +51,12 @@ outputs/
 
 To get a summary for each dated folder, run the following:
 ```
-OUTPUT_DIR=outputs/baseline pipenv run python -m scripts.get_summary 
+OUTPUT_DIR=outputs/baseline poetry run python -m scripts.get_summary 
 ```
 
 To get an aggregated summary for all dated folders, run the following:
 ```
-OUTPUT_DIR=outputs/baseline pipenv run python -m scripts.get_aggregated_summary
+OUTPUT_DIR=outputs/baseline poetry run python -m scripts.get_aggregated_summary
 ```
 
 

generate_requests.py

@@ -17,49 +17,49 @@ def parse_har_file(har_file):
     requests = []
 
     for page in har_parser.pages:
-        entries = page.filter_entries(content_type=r'(text/html|application/pdf)')
+        entries = page.filter_entries(content_type=r"(text/html|application/pdf)")
         for entry in entries:
-            entry_request = entry['request']
+            entry_request = entry["request"]
 
             request_base_url = "{0.scheme}://{0.netloc}".format(
-                urlsplit(entry_request['url'])
+                urlsplit(entry_request["url"])
             )
 
             request = {
-                'method': entry_request['method'],
-                'url': entry_request['url'].replace(request_base_url, ""),
-                'datetime': dateutil.parser.parse(entry['startedDateTime']),
+                "method": entry_request["method"],
+                "url": entry_request["url"].replace(request_base_url, ""),
+                "datetime": dateutil.parser.parse(entry["startedDateTime"]),
             }
 
-            if entry_request['method'] == 'POST':
-                request['data'] = {
-                    unquote_plus(item['name']): unquote_plus(item['value'])
-                    for item in entry_request['postData']['params']
+            if entry_request["method"] == "POST":
+                request["data"] = {
+                    unquote_plus(item["name"]): unquote_plus(item["value"])
+                    for item in entry_request["postData"]["params"]
                 }
-                request['data'].pop('csrf_token', None)
+                request["data"].pop("csrf_token", None)
 
             requests.append(request)
 
-    requests.sort(key=itemgetter('datetime'))
+    requests.sort(key=itemgetter("datetime"))
 
     for request in requests:
-        request.pop('datetime', None)
+        request.pop("datetime", None)
 
-    return {'requests': requests}
+    return {"requests": requests}
 
 
 @click.command()
-@click.argument('har_file', type=click.File('r'))
-@click.argument('requests_file', type=click.File('w'))
-@click.argument('schema_name')
+@click.argument("har_file", type=click.File("r"))
+@click.argument("requests_file", type=click.File("w"))
+@click.argument("schema_name")
 def generate_requests(har_file, requests_file, schema_name):
     requests = parse_har_file(har_file)
-    requests['schema_name'] = schema_name
-    requests['schema_url'] = (
+    requests["schema_name"] = schema_name
+    requests["schema_url"] = (
         f"https://storage.googleapis.com/eq-questionnaire-schemas/{schema_name}.json"
     )
     json.dump(requests, requests_file, indent=4)
 
 
-if __name__ == '__main__':
+if __name__ == "__main__":
     generate_requests()