Merge pull request community#166243 from community/ebndev-2025-07-15

Update actions_labeller.yml to improve discussion handling...

Author: ebndev

.github/workflows/actions_labeller.yml

@@ -9,11 +9,11 @@ jobs:
     if: ${{ contains(github.event.discussion.category.name, 'Actions') }}
     runs-on: ubuntu-latest
     env:
-      GH_TOKEN: ${{ secrets.GITHUB_TOKEN }} # Global authentication for gh CLI
+      GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
     steps:
       - name: Get discussion body html
         id: get_discussion_body_html
-        env: 
+        env:
           OWNER: ${{ github.repository_owner }}
           REPO: ${{ github.event.repository.name }}
           DISCUSSION_NUMBER: ${{ github.event.discussion.number }}
@@ -30,9 +30,8 @@ jobs:
 
           echo 'DISCUSSION_BODY_HTML='$(jq -r '.data.repository.discussion.bodyHTML' discussion_data.json) >> $GITHUB_ENV
           echo 'DISCUSSION_ID='$(jq -r '.data.repository.discussion.id' discussion_data.json) >> $GITHUB_ENV
-      
-      - run: npm install jsdom
 
+      - run: npm install jsdom dompurify
 
       - name: Extract Title and Body Text
         id: extract_text
@@ -48,7 +47,6 @@ jobs:
             const { DISCUSSION_BODY_HTML } = process.env;
             const fragment = JSDOM.fragment(DISCUSSION_BODY_HTML);
             let body = '';
-            // Find all <h3> and <p> pairs
             const h3s = Array.from(fragment.querySelectorAll('h3'));
             h3s.forEach(h3 => {
               const heading = h3.textContent.trim();
@@ -59,7 +57,6 @@ jobs:
                 body = p.textContent.trim();
               }
             });
-            // Remove leading/trailing quotes from body
             body = body.replace(/^['\"]+|['\"]+$/g, '');
             const title = process.env.DISCUSSION_TITLE || '';
             core.info(`Extracted title: ${title}`);
@@ -69,6 +66,8 @@ jobs:
       - name: Extract Primary and Secondary Topic Areas
         id: extract_topics
         uses: actions/github-script@v6
+        env:
+          DISCUSSION_BODY_HTML: ${{ env.DISCUSSION_BODY_HTML }}
         with:
           result-encoding: string
           script: |
@@ -78,11 +77,9 @@ jobs:
             const fragment = JSDOM.fragment(DISCUSSION_BODY_HTML);
             let primary = '';
             let secondary = '';
-            // Find all <h3> and <p> pairs (form headings as h3, answers as p)
             const h3s = Array.from(fragment.querySelectorAll('h3'));
             h3s.forEach(h3 => {
               const heading = h3.textContent.trim();
-              // Look for the next <p> sibling after each heading
               let p = h3.nextElementSibling;
               while (p && p.tagName !== 'P') p = p.nextElementSibling;
               if (!p) return;
@@ -100,78 +97,87 @@ jobs:
       - name: Auto-label by keyword search
         id: auto_label_keywords
         uses: actions/github-script@v6
+        env:
+          EXTRACT_TEXT_RESULT: ${{ steps.extract_text.outputs.result }}
         with:
           result-encoding: string
           script: |
-            // Keyphrase to label mapping
+            const jsdom = require('jsdom');
+            const { JSDOM } = jsdom;
+            const createDOMPurify = require('dompurify');
+            const window = (new JSDOM('')).window;
+            const DOMPurify = createDOMPurify(window);
+
             const labelMap = [
               {
                 label: 'Workflow Deployment',
                 keywords: [
-                    "deployment error",
-                    "publish artifact",
-                    "release failure",
-                    "deployment target",
-                    "github pages",
-                    "deployment issue",
-                    "release workflow",
-                    "target environment"
-                  ]
+                  "deployment error",
+                  "publish artifact",
+                  "release failure",
+                  "deployment target",
+                  "github pages",
+                  "deployment issue",
+                  "release workflow",
+                  "target environment"
+                ]
               },
               {
                 label: 'Workflow Configuration',
                 keywords: [
-                    "yaml syntax",
-                    "job dependency",
-                    "setup error",
-                    "workflow file",
-                    "configuration issue",
-                    "matrix strategy",
-                    "define env",
-                    "secret management",
-                    "environment setup",
-                    "config job"
-                  ]
+                  "yaml syntax",
+                  "job dependency",
+                  "setup error",
+                  "workflow file",
+                  "configuration issue",
+                  "matrix strategy",
+                  "define env",
+                  "secret management",
+                  "environment setup",
+                  "config job"
+                ]
               },
               {
                 label: 'Schedule & Cron Jobs',
                 keywords: [
-                    "cron job",
-                    "scheduled workflow",
-                    "timing issue",
-                    "delay trigger",
-                    "timezone error",
-                    "periodic run",
-                    "recurring schedule",
-                    "interval workflow",
-                    "scheduled trigger",
-                    "cron expression"
-                  ]
+                  "cron job",
+                  "scheduled workflow",
+                  "timing issue",
+                  "delay trigger",
+                  "timezone error",
+                  "periodic run",
+                  "recurring schedule",
+                  "interval workflow",
+                  "scheduled trigger",
+                  "cron expression"
+                ]
               },
               {
                 label: 'Metrics & Insights',
                 keywords: [
-                    "usage metrics",
-                    "performance trend",
-                    "analytics graph",
-                    "stats dashboard",
-                    "timeseries graph",
-                    "insight report",
-                    "metric tracking",
-                    "workflow analytics",
-                    "performance metric",
-                    "statistics report"
-                  ]
+                  "usage metrics",
+                  "performance trend",
+                  "analytics graph",
+                  "stats dashboard",
+                  "timeseries graph",
+                  "insight report",
+                  "metric tracking",
+                  "workflow analytics",
+                  "performance metric",
+                  "statistics report"
+                ]
               }
             ];
             const miscLabel = 'Misc';
             let title = '';
             let body = '';
             try {
-              const parsed = JSON.parse(`${{ steps.extract_text.outputs.result }}`);
-              title = parsed.title || '';
-              body = parsed.body || '';
-            } catch (e) {}
+              const parsed = JSON.parse(process.env.EXTRACT_TEXT_RESULT);
+              title = DOMPurify.sanitize(parsed.title || '', { ALLOWED_TAGS: [], ALLOWED_ATTR: [] }).trim();
+              body  = DOMPurify.sanitize(parsed.body  || '', { ALLOWED_TAGS: [], ALLOWED_ATTR: [] }).trim();
+            } catch (e) {
+              core.error('Failed to parse or sanitize discussion text: ' + e.message);
+            }
             const text = (title + ' ' + body).toLowerCase();
             let foundLabel = miscLabel;
             core.info(`Auto-label debug: text to match: '${text}'`);
@@ -188,6 +194,7 @@ jobs:
             }
             core.info(`Auto-label debug: selected label: '${foundLabel}'`);
             return foundLabel;
+
       - name: Fetch label ID for primary topic
         id: fetch_primary_label_id
         env:
@@ -239,6 +246,7 @@ jobs:
 
           SECONDARY_LABEL_ID=$(jq -r '.data.repository.labels.edges[0]?.node?.id // empty' secondary_label_data.json)
           echo "SECONDARY_LABEL_ID=$SECONDARY_LABEL_ID" >> $GITHUB_ENV
+
       - name: Fetch label ID for auto-label
         id: fetch_auto_label_id
         env: