Merge pull request #283 from ndaidong/v7.x.x

v7.0.0

dist/cjs/package.json

@@ -1,5 +1,5 @@
 {
   "name": "article-parser-cjs",
-  "version": "7.0.0rc4",
+  "version": "7.0.0",
   "main": "./article-parser.js"
 }

package.json

@@ -1,5 +1,5 @@
 {
-  "version": "7.0.0rc4",
+  "version": "7.0.0",
   "name": "article-parser",
   "description": "To extract main article from given URL",
   "homepage": "https://ndaidong.github.io/article-parser-demo/",
@@ -8,8 +8,7 @@
     "url": "[email protected]:ndaidong/article-parser.git"
   },
   "author": "@ndaidong",
-  "main": "./dist/cjs/article-parser.js",
-  "module": "./src/main.js",
+  "main": "./src/main.js",
   "browser": {
     "linkedom": "./src/browser/linkedom.js",
     "./main.js": "./dist/article-parser.browser.js"
@@ -24,18 +23,17 @@
     "pretest": "npm run lint",
     "test": "cross-env NODE_ENV=test NODE_OPTIONS=--experimental-vm-modules jest --unhandled-rejections=strict",
     "build": "node build",
-    "eval": "cross-env DEBUG=*:* node eval",
-    "eval:cjs": "cross-env DEBUG=*:* node eval.cjs",
+    "eval": "cross-env node eval",
+    "eval:cjs": "cross-env node eval.cjs",
     "reset": "node reset"
   },
   "dependencies": {
     "@mozilla/readability": "^0.4.2",
     "axios": "^0.27.2",
     "bellajs": "^11.0.3",
-    "debug": "^4.3.4",
-    "html-crush": "^5.0.19",
+    "html-crush": "^5.0.20",
     "linkedom": "^0.14.12",
-    "sanitize-html": "^2.7.0",
+    "sanitize-html": "^2.7.1",
     "string-comparison": "^1.1.0",
     "tldts": "^5.7.84",
     "urlpattern-polyfill": "^5.0.5"
@@ -48,9 +46,9 @@
   "devDependencies": {
     "@types/sanitize-html": "^2.6.2",
     "cross-env": "^7.0.3",
-    "esbuild": "^0.14.49",
+    "esbuild": "^0.14.50",
     "jest": "^28.1.3",
-    "nock": "^13.2.8",
+    "nock": "^13.2.9",
     "standard": "^17.0.0"
   },
   "keywords": [

src/utils/linker.js

@@ -7,8 +7,6 @@ import * as stringComparison from 'string-comparison'
 
 import { getParserOptions } from '../config.js'
 
-import logger from './logger.js'
-
 export const isValid = (url = '') => {
   try {
     const ourl = new URL(url)
@@ -23,7 +21,6 @@ export const absolutify = (fullUrl = '', relativeUrl = '') => {
     const result = new URL(relativeUrl, fullUrl)
     return result.toString()
   } catch (err) {
-    logger.error(err)
     return ''
   }
 }
@@ -98,7 +95,6 @@ export const purify = (url) => {
 
     return pureUrl.toString().replace(pureUrl.hash, '')
   } catch (err) {
-    logger.error(err)
     return null
   }
 }

src/utils/parseFromHtml.js

@@ -23,8 +23,6 @@ import { execPreParser, execPostParser } from './transformation.js'
 
 import getTimeToRead from './getTimeToRead.js'
 
-import logger from './logger.js'
-
 import { getParserOptions } from '../config.js'
 
 const summarize = (desc, txt, threshold, maxlen) => {
@@ -68,11 +66,9 @@ export default async (inputHtml, inputUrl = '') => {
 
   // gather title
   if (!title) {
-    logger.info('Could not detect article title from meta!')
     title = extractTitleWithReadability(html, inputUrl)
   }
   if (!title) {
-    logger.info('Could not detect article title!')
     return null
   }
 
@@ -84,7 +80,6 @@ export default async (inputHtml, inputUrl = '') => {
   )
 
   if (!links.length) {
-    logger.info('Could not detect article link!')
     return null
   }
 
@@ -112,13 +107,11 @@ export default async (inputHtml, inputUrl = '') => {
   const content = fns(html)
 
   if (!content) {
-    logger.info('Could not detect article content!')
     return null
   }
 
   const textContent = stripTags(content)
   if (textContent.length < contentLengthThreshold) {
-    logger.info('Main article is too short!')
     return null
   }
 

src/utils/retrieve.js

@@ -2,8 +2,6 @@
 
 import axios from 'axios'
 
-import logger from './logger.js'
-
 import { getRequestOptions } from '../config.js'
 
 export default async (url) => {
@@ -12,13 +10,10 @@ export default async (url) => {
 
     const contentType = res.headers['content-type'] || ''
     if (!contentType || !contentType.includes('text/html')) {
-      logger.error(`Content type must be "text/html", not "${contentType}"`)
-      return null
+      throw new Error(`Content type must be "text/html", not "${contentType}"`)
     }
-
     return res.data
   } catch (err) {
-    logger.error(err.message || err)
-    return null
+    throw new Error(`${err.name}: ${err.message}`)
   }
 }

src/utils/retrieve.test.js

@@ -31,8 +31,11 @@ test('test retrieve with unsupported content type', async () => {
   scope.get(path).reply(200, '', {
     'Content-Type': 'something/strange'
   })
-  const result = await retrieve(url)
-  expect(result).toBe(null)
+  try {
+    await retrieve(url)
+  } catch (err) {
+    expect(err).toBeTruthy()
+  }
 })
 
 test('test retrieve from bad source', async () => {
@@ -42,6 +45,9 @@ test('test retrieve from bad source', async () => {
   scope.get(path).reply(500, '<div>this is content</div>', {
     'Content-Type': 'text/html'
   })
-  const result = await retrieve(url)
-  expect(result).toBe(null)
+  try {
+    await retrieve(url)
+  } catch (err) {
+    expect(err).toBeTruthy()
+  }
 })