Add flake8 lint checking (#253)

The CI will now fail if `flake8` lint checking fails.

A number of minor changes have been made to remove unused imports and break
excessively long lines. (Flake8's tolerance is less generous than black's.)

Several unused variable have been removed, including a `version` from the Quay
graphs `getMatchRuns` ... the file history shows this was initially used in
the query string, but it was removed and (probably inadvertently) restored
(without the associated query phrase) by REVAMP. The other main unused
variable is "data" in the `ingress-performance` case of the OCP graph API.
This code has always been unimplemented; instead of attempting to resolve that
here, I've created issue #252.

Author: dbutenhof

.github/workflows/backend-check.yaml

@@ -32,8 +32,6 @@ jobs:
           cd backend
           tox -e format,isort
       - name: Check for lint
-        # Report errors but don't fail until we achieve stability!
-        continue-on-error: true
         run: |
           cd backend
           tox -e lint

backend/.flake8

@@ -0,0 +1,6 @@
+[flake8]
+ignore = E203,E226,W503,E704
+
+# Black tries to limit lines to 88 characters: keep flake8 from complaining
+# as long as we're below 100.
+max-line-length = 100

backend/app/api/v1/commons/constants.py

@@ -168,11 +168,12 @@
 
 SPLUNK_SEMAPHORE_COUNT = 5  # Arbitrary concurrency limit in an asyncio semaphore
 
-# These products don't need to be included in the filter dict when applying filters on the Home tab,
-# as the filtering function will automatically handle them using the corresponding product-to-mapper lookup.
+# These products don't need to be included in the filter dict when applying
+# filters on the Home tab, as the filtering function will automatically handle
+# them using the corresponding product-to-mapper lookup.
 STANDOUT_PRODUCTS = ["ocp", "telco", "quay"]
 # These products aren't listed by their exact names like "hce" and "ocm",
-# but instead represent grouped categories such as "Developer", "Insights", etc.
+# but instead represent grouped categories such as "Developer" or "Insights".
 GENERAL_PRODUCTS = ["hce", "ocm"]
 TELCO_STATUS_MAP = {"success": "passed", "failed": "failed", "failure": "failure"}
 

backend/app/api/v1/commons/example_responses.py

@@ -170,8 +170,11 @@ def response_422():
             "networkType": "OVNKubernetes",
             "buildTag": "1886955825944072192",
             "jobStatus": "success",
-            "buildUrl": "https://prow.ci.openshift.org/view/gs/test-platform-results/pr-logs/pull/openshift_release/60981/rehearse-60981-periodic-ci-openshift-ols-load-generator-main-ols-load-test-100workers/1886955825944072192",
-            "upstreamJob": "rehearse-60981-periodic-ci-openshift-ols-load-generator-main-ols-load-test-100workers",
+            "buildUrl": "https://prow.ci.openshift.org/view/gs/test-platform-results/pr-logs/"
+            "pull/openshift_release/60981/rehearse-60981-periodic-ci-openshift-ols-"
+            "load-generator-main-ols-load-test-100workers/1886955825944072192",
+            "upstreamJob": "rehearse-60981-periodic-ci-openshift-ols-load-generator-main-"
+            "ols-load-test-100workers",
             "upstreamJobBuild": "778dd992-8353-4cec-9807-7343290267f8",
             "executionDate": "2025-02-05T03:19:50Z",
             "jobDuration": "3396",

backend/app/api/v1/commons/hce.py

@@ -1,4 +1,4 @@
-from datetime import date, datetime
+from datetime import date
 
 import pandas as pd
 

backend/app/api/v1/commons/telco.py

@@ -6,7 +6,6 @@
 import app.api.v1.commons.constants as constants
 import app.api.v1.commons.hasher as hasher
 import app.api.v1.commons.utils as utils
-import app.api.v1.endpoints.telco.telcoGraphs as telcoGraphs
 from app.services.splunk import SplunkService
 
 
@@ -96,12 +95,6 @@ async def getFilterData(
     start_datetime: date, end_datetime: date, filter: str, configpath: str
 ):
 
-    cfg = config.get_config()
-    try:
-        jenkins_url = cfg.get("telco.config.job_url")
-    except Exception as e:
-        print(f"Error reading telco configuration: {e}")
-
     query = {
         "earliest_time": "{}T00:00:00".format(start_datetime.strftime("%Y-%m-%d")),
         "latest_time": "{}T23:59:59".format(end_datetime.strftime("%Y-%m-%d")),

backend/app/api/v1/commons/utils.py

@@ -246,12 +246,15 @@ def update_filter_product(filter):
     filter_dict = get_dict_from_qs(filter) if filter else {}
     filter_product = filter_dict.pop("product", None)
 
-    # The product filter dropdown includes options like "ocp", "telco", "quay", "Developer", "Insights", etc.
+    # The product filter dropdown includes options like "ocp", "telco", "quay",
+    # "Developer", "Insights", etc
+    #
     # Among these, "ocp", "quay", and "telco" are considered STANDOUT_PRODUCTS.
     # These standout products are handled through a product-to-mapper lookup,
-    # so they do NOT need to be included in the filter_dict explicitly.
-    # The remaining products (like "Developer", "Insights", etc.) must be filtered using "hce" and "ocm" mappers,
-    # so they are retained in the filter_dict for backend filtering.
+    # so they do NOT need to be included in the filter_dict explicitly. The
+    # remaining products (like "Developer", "Insights", etc.) must be filtered
+    # using "hce" and "ocm" mappers, so they are retained in the filter_dict
+    # for backend filtering.
 
     if filter_product:
         matched = [p for p in filter_product if p in constants.STANDOUT_PRODUCTS]

backend/app/api/v1/endpoints/cpt/cptJobs.py

@@ -1,7 +1,6 @@
 import asyncio
 from datetime import date, datetime, timedelta
 import json
-from multiprocessing import cpu_count
 import traceback
 from urllib.parse import urlencode
 

backend/app/api/v1/endpoints/cpt/maps/hce.py

@@ -1,5 +1,4 @@
 from datetime import date
-import traceback
 from urllib.parse import urlencode
 
 import pandas as pd
@@ -35,7 +34,7 @@ async def hceMapper(
     updated_filter = await get_updated_filter(filter)
 
     response = await getData(
-        start_datetime, end_datetime, size, offset, updated_filter, f"hce.elasticsearch"
+        start_datetime, end_datetime, size, offset, updated_filter, "hce.elasticsearch"
     )
 
     if isinstance(response, pd.DataFrame) or not response:
@@ -79,7 +78,7 @@ async def hceFilter(start_datetime: date, end_datetime: date, filter: str):
         updated_filter = await get_updated_filter(filter)
 
         response = await getFilterData(
-            start_datetime, end_datetime, updated_filter, f"hce.elasticsearch"
+            start_datetime, end_datetime, updated_filter, "hce.elasticsearch"
         )
 
         if isinstance(response, pd.DataFrame) or not response:

backend/app/api/v1/endpoints/cpt/maps/ocm.py

@@ -24,7 +24,7 @@ async def ocmMapper(
     updated_filter = await get_updated_filter(filter)
 
     response = await getData(
-        start_datetime, end_datetime, size, offset, filter, f"ocm.elasticsearch"
+        start_datetime, end_datetime, size, offset, updated_filter, "ocm.elasticsearch"
     )
     if isinstance(response, pd.DataFrame) or not response:
         df = response["data"]
@@ -53,7 +53,7 @@ async def ocmFilter(start_datetime: date, end_datetime: date, filter: str):
     updated_filter = await get_updated_filter(filter)
 
     response = await getFilterData(
-        start_datetime, end_datetime, updated_filter, f"ocm.elasticsearch"
+        start_datetime, end_datetime, updated_filter, "ocm.elasticsearch"
     )
     if isinstance(response, pd.DataFrame) or not response:
         return {"total": 0, "filterData": [], "summary": {}}