Merge pull request #379 from extractus/dev

v8.0.5

.github/workflows/ci-test.yml

@@ -15,7 +15,7 @@ jobs:
         node_version: [18.x, 20.x, 21.x]
 
     steps:
-    - uses: actions/checkout@v3
+    - uses: actions/checkout@v4
 
     - name: setup Node.js v${{ matrix.node_version }}
       uses: actions/setup-node@v4
@@ -31,15 +31,28 @@ jobs:
         npm run build --if-present
         npm run test
 
-    - name: Report Coveralls
+    - name: Coveralls Parallel
       uses: coverallsapp/github-action@v2
       with:
+        flag-name: run-${{ join(matrix.*, '-') }}
+        parallel: true
         github-token: ${{ secrets.GITHUB_TOKEN }}
 
     - name: cache node modules
-      uses: actions/cache@v3
+      uses: actions/cache@v4
       with:
         path: ~/.npm
         key: ${{ runner.os }}-node-${{ hashFiles('**/package-lock.json') }}
         restore-keys: |
           ${{ runner.os }}-node-
+
+  finish:
+    needs: test
+    if: ${{ always() }}
+    runs-on: ubuntu-latest
+    steps:
+    - name: Coveralls Finished
+      uses: coverallsapp/github-action@v2
+      with:
+        parallel-finished: true
+        carryforward: "run-18.x,run-20.x,run-21.x"

package.json

@@ -1,5 +1,5 @@
 {
-  "version": "8.0.4",
+  "version": "8.0.5",
   "name": "@extractus/article-extractor",
   "description": "To extract main article from given URL",
   "homepage": "https://github.com/extractus/article-extractor",
@@ -30,18 +30,18 @@
     "reset": "node reset"
   },
   "dependencies": {
-    "@mozilla/readability": "^0.4.4",
+    "@mozilla/readability": "^0.5.0",
     "bellajs": "^11.1.2",
     "cross-fetch": "^4.0.0",
-    "linkedom": "^0.16.4",
+    "linkedom": "^0.16.6",
     "sanitize-html": "2.11.0"
   },
   "devDependencies": {
     "@types/sanitize-html": "^2.9.5",
-    "eslint": "^8.55.0",
+    "eslint": "^8.56.0",
     "https-proxy-agent": "^7.0.2",
     "jest": "^29.7.0",
-    "nock": "^13.4.0"
+    "nock": "^13.5.0"
   },
   "keywords": [
     "article",

src/utils/extractLdSchema.js

@@ -35,6 +35,14 @@ const attributeLists = {
   type: '@type',
 }
 
+const parseJson = (text) => {
+  try {
+    return JSON.parse(text)
+  } catch {
+    return null
+  }
+}
+
 /**
  * Parses JSON-LD data from a document and populates an entry object.
  * Only populates if the original entry object is empty or undefined.
@@ -46,26 +54,25 @@ const attributeLists = {
 export default (document, entry) => {
   const ldSchema = document.querySelector('script[type="application/ld+json"]')?.textContent
 
-  if (!ldSchema) {
-    return entry
-  }
+  const ldJson = ldSchema ? parseJson(ldSchema) : null
 
-  const ldJson = JSON.parse(ldSchema)
-  Object.entries(attributeLists).forEach(([key, attr]) => {
-    if ((typeof entry[key] === 'undefined' || entry[key] === '') && ldJson[attr]) {
-      if (key === 'type' && typeof ldJson[attr] === 'string') {
-        return entry[key] = typeSchemas.includes(ldJson[attr].toLowerCase()) ? ldJson[attr].toLowerCase() : ''
-      }
+  if (ldJson) {
+    Object.entries(attributeLists).forEach(([key, attr]) => {
+      if ((typeof entry[key] === 'undefined' || entry[key] === '') && ldJson[attr]) {
+        if (key === 'type' && typeof ldJson[attr] === 'string') {
+          return entry[key] = typeSchemas.includes(ldJson[attr].toLowerCase()) ? ldJson[attr].toLowerCase() : ''
+        }
 
-      if (typeof ldJson[attr] === 'string') {
-        return entry[key] = ldJson[attr].toLowerCase()
-      }
+        if (typeof ldJson[attr] === 'string') {
+          return entry[key] = ldJson[attr].toLowerCase()
+        }
 
-      if (Array.isArray(ldJson[attr]) && typeof ldJson[attr][0] === 'string') {
-        return entry[key] = ldJson[attr][0].toLowerCase()
+        if (Array.isArray(ldJson[attr]) && typeof ldJson[attr][0] === 'string') {
+          return entry[key] = ldJson[attr][0].toLowerCase()
+        }
       }
-    }
-  })
+    })
+  }
 
   return entry
 }