Allow for filtering axe-selenium-python accessibility violations by t…

…ag. Closes [#1752.

components/collector/src/source_collectors/file_source_collectors/axe_selenium_python.py

@@ -1,11 +1,12 @@
 """axe-selenium-python accessibility analysis metric source."""
 
 from datetime import datetime
+from typing import Collection
 
 from dateutil.parser import parse
 
 from base_collectors import JSONFileSourceCollector, SourceUpToDatenessCollector
-from collector_utilities.functions import md5_hash
+from collector_utilities.functions import md5_hash, match_string_or_regular_expression
 from collector_utilities.type import Response
 from source_model import Entity, SourceMeasurement, SourceResponses
 
@@ -14,35 +15,53 @@ class AxeSeleniumPythonAccessibility(JSONFileSourceCollector):
     """Collector class to get accessibility violations."""
 
     async def _parse_source_responses(self, responses: SourceResponses) -> SourceMeasurement:
-        impact_levels = self._parameter("impact")
         entity_attributes = []
         for response in responses:
             json = await response.json(content_type=None)
             url = json["url"]
             for violation in json.get("violations", []):
                 for node in violation.get("nodes", []):
-                    if node.get("impact") not in impact_levels:
-                        continue
-                    entity_attributes.append(
-                        dict(
-                            description=violation.get("description"),
-                            element=node.get("html"),
-                            help=violation.get("helpUrl"),
-                            impact=node.get("impact"),
-                            page=url,
-                            url=url,
-                            tags=", ".join(sorted(violation.get("tags", []))),
-                            violation_type=violation.get("id"),
+                    tags = violation.get("tags", [])
+                    impact = node.get("impact")
+                    if self.__include_violation(impact, tags):
+                        entity_attributes.append(
+                            dict(
+                                description=violation.get("description"),
+                                element=node.get("html"),
+                                help=violation.get("helpUrl"),
+                                impact=impact,
+                                page=url,
+                                url=url,
+                                tags=", ".join(sorted(tags)),
+                                violation_type=violation.get("id"),
+                            )
                         )
-                    )
-        entities = [
-            Entity(
-                key=md5_hash(",".join(str(value) for key, value in attributes.items() if key != "tags")), **attributes
-            )
-            for attributes in entity_attributes
-        ]
+        entities = [Entity(key=self.__create_key(attributes), **attributes) for attributes in entity_attributes]
         return SourceMeasurement(entities=entities)
 
+    def __include_violation(self, impact: str, tags: Collection[str]) -> bool:
+        """Return whether to include the violation."""
+        if impact not in self._parameter("impact"):
+            return False
+        if tags_to_include := self._parameter("tags_to_include"):
+            for tag in tags:
+                if match_string_or_regular_expression(tag, tags_to_include):
+                    break
+            else:
+                return False
+        if tags_to_ignore := self._parameter("tags_to_ignore"):
+            for tag in tags:
+                if match_string_or_regular_expression(tag, tags_to_ignore):
+                    return False
+        return True
+
+    @staticmethod
+    def __create_key(attributes) -> str:
+        """Create a key for the entity based on the attributes."""
+        # We ignore tags for two reasons: 1) If the violation is the same, so should the tags be. 2) Tags were added to
+        # the entities later and including them in the key would change the key for existing entities.
+        return md5_hash(",".join(str(value) for key, value in attributes.items() if key != "tags"))
+
 
 class AxeSeleniumPythonSourceUpToDateness(JSONFileSourceCollector, SourceUpToDatenessCollector):
     """Collector to get the source up-to-dateness of axe-selenium-python JSON reports."""

components/collector/tests/source_collectors/file_source_collectors/test_axe_selenium_python.py

@@ -86,6 +86,18 @@ async def test_filter_by_impact(self):
         response = await self.collect(self.metric, get_request_json_return_value=self.json)
         self.assert_measurement(response, value="1")
 
+    async def test_filter_by_tag_include(self):
+        """Test that violations can be filtered by tag."""
+        self.sources["source_id"]["parameters"]["tags_to_include"] = ["wcag2aa"]
+        response = await self.collect(self.metric, get_request_json_return_value=self.json)
+        self.assert_measurement(response, value="1", entities=[self.expected_entities[0]])
+
+    async def test_filter_by_tag_ignore(self):
+        """Test that violations can be filtered by tag."""
+        self.sources["source_id"]["parameters"]["tags_to_ignore"] = ["wcag2aa"]
+        response = await self.collect(self.metric, get_request_json_return_value=self.json)
+        self.assert_measurement(response, value="1", entities=[self.expected_entities[1]])
+
     async def test_zipped_json(self):
         """Test that a zip archive with JSON files is processed correctly."""
         self.sources["source_id"]["parameters"]["url"] = f"{self.axe_json_url}.zip"

components/server/src/data/datamodel.json

@@ -921,6 +921,30 @@
                         "source_up_to_dateness"
                     ]
                 },
+                "tags_to_include": {
+                    "name": "Tags to include (regular expressions or tags)",
+                    "short_name": "tags to include",
+                    "help": "Tags to include can be specified by tag or by regular expression.",
+                    "type": "multiple_choice_with_addition",
+                    "placeholder": "all",
+                    "mandatory": false,
+                    "default_value": [],
+                    "metrics": [
+                        "accessibility"
+                    ]
+                },
+                "tags_to_ignore": {
+                    "name": "Tags to ignore (regular expressions or tags)",
+                    "short_name": "tags to ignore",
+                    "help": "Tags to ignore can be specified by tag or by regular expression.",
+                    "type": "multiple_choice_with_addition",
+                    "placeholder": "none",
+                    "mandatory": false,
+                    "default_value": [],
+                    "metrics": [
+                        "accessibility"
+                    ]
+                },
                 "impact": {
                     "name": "Impact levels",
                     "short_name": "impact levels",
@@ -1373,7 +1397,7 @@
                     "help": "Pipelines to ignore can be specified by pipeline name or by regular expression. Use {folder name}/{pipeline name} for the names of pipelines in folders.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "none",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "failed_jobs",
@@ -2314,7 +2338,7 @@
                     "help": "Jobs to ignore can be specified by job name or by regular expression.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "none",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "failed_jobs",
@@ -2546,7 +2570,7 @@
                     "help": "Jobs to include can be specified by job name or by regular expression. Use {parent job name}/{child job name} for the names of nested jobs.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "all",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "failed_jobs",
@@ -2560,7 +2584,7 @@
                     "help": "Jobs to ignore can be specified by job name or by regular expression. Use {parent job name}/{child job name} for the names of nested jobs.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "none",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "failed_jobs",
@@ -4061,7 +4085,7 @@
                     "help": "Transactions to ignore can be specified by transaction name or by regular expression.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "none",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "slow_transactions",
@@ -4074,7 +4098,7 @@
                     "help": "Transactions to include can be specified by transaction name or by regular expression.",
                     "type": "multiple_choice_with_addition",
                     "placeholder": "all",
-                    "mandatory":  false,
+                    "mandatory": false,
                     "default_value": [],
                     "metrics": [
                         "slow_transactions",
@@ -5143,7 +5167,7 @@
                 "component": {
                     "name": "Project key",
                     "short_name": "project key",
-                    "help":  "The project key can be found by opening the project in SonarQube and looking at the bottom of the grey column on the right.",
+                    "help": "The project key can be found by opening the project in SonarQube and looking at the bottom of the grey column on the right.",
                     "type": "string",
                     "mandatory": true,
                     "default_value": "",

docs/CHANGELOG.md

@@ -14,6 +14,7 @@ and this project adheres to [Semantic Versioning](http://semver.org/spec/v2.0.0.
 ### Added
 
 - Add the tags of accessibility rules to the detail information of axe-selenium-python sources. Closes [#1751](https://github.com/ICTU/quality-time/issues/1751).
+- Allow for filtering axe-selenium-python accessibility violations by tag. Closes [#1752](https://github.com/ICTU/quality-time/issues/1752).
 
 ## [3.16.0] - [2020-12-12]
 

docs/METRICS_AND_SOURCES.md

@@ -101,6 +101,8 @@ This document lists all [metrics](#metrics) that *Quality-time* can measure and
 | Impact levels | Multiple choice | No | If provided, only count accessibility violations with the impact levels. |
 | Password for basic authentication | Password | No |  |
 | Private token | Password | No |  |
+| Tags to ignore (regular expressions or tags) | Multiple choice with addition | No | Tags to ignore can be specified by tag or by regular expression. |
+| Tags to include (regular expressions or tags) | Multiple choice with addition | No | Tags to include can be specified by tag or by regular expression. |
 | URL to an axe-selenium-python report in JSON format or to a zip with axe-selenium-python reports in JSON format | URL | Yes |  |
 | URL to an axe-selenium-python report in a human readable format | String | No | If provided, users clicking the source URL will visit this URL instead of the axe-selenium-report report in JSON format. |
 | Username for basic authentication | String | No |  |