fix: update tests to remove dayjs where possible, update formats

package.json

@@ -132,16 +132,14 @@
   },
   "bundleDependencies": [
     "jquery",
-    "moment-timezone",
     "browser-request"
   ],
   "browser": {
     "main": "./dist/mercury.web.js",
     "cheerio": "./src/shims/cheerio-query",
     "jquery": "./node_modules/jquery/dist/jquery.min.js",
     "postman-request": "browser-request",
-    "iconv-lite": "./src/shims/iconv-lite",
-    "moment-timezone": "./node_modules/moment-timezone/builds/moment-timezone-with-data-2012-2022.min.js"
+    "iconv-lite": "./src/shims/iconv-lite"
   },
   "husky": {
     "hooks": {

src/cleaners/date-published.js

@@ -45,11 +45,17 @@ export function createDate(dateString, timezone, format) {
   }
 
   if (timezone) {
-    return format
-      ? dayjs.tz(dateString, format, timezone)
-      : dayjs.tz(new Date(dateString), timezone);
+    try {
+      return format
+        ? dayjs.tz(dateString, format, timezone)
+        : dayjs.tz(dayjs(dateString).format('YYYY-MM-DD HH:mm:ss'), timezone);
+    } catch (error) {
+      // return an intentionally invalid dayjs object,
+      // in case the input needs to be cleaned first
+      return dayjs('');
+    }
   }
-  return format ? dayjs(dateString, format) : dayjs(new Date(dateString));
+  return format ? dayjs(dateString, format) : dayjs(dateString);
 }
 
 // Take a date published string, and hopefully return a date out of
@@ -70,7 +76,7 @@ export default function cleanDatePublished(
 
   if (!date.isValid()) {
     dateString = cleanDateString(dateString);
-    date = createDate(dateString, timezone, format);
+    date = createDate(dateString, timezone);
   }
 
   return date.isValid() ? date.toISOString() : null;

src/extractors/custom/clinicaltrials.gov/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -59,7 +58,7 @@ describe('ClinicaltrialsGovExtractor', () => {
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(moment(date_published).format('YYYY-MM-DD'), '2018-11-21');
+      assert.equal(date_published, '2018-11-21T05:00:00.000Z');
     });
 
     it('returns the content', async () => {

src/extractors/custom/fortune.com/index.js

@@ -11,8 +11,6 @@ export const FortuneComExtractor = {
 
   date_published: {
     selectors: ['.MblGHNMJ'],
-
-    timezone: 'UTC',
   },
 
   lead_image_url: {

src/extractors/custom/genius.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -51,11 +50,10 @@ describe('GeniusComExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/genius.com/index.js.
       const { date_published } = await result;
-      const newDatePublished = moment(date_published).format();
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(newDatePublished.split('T')[0], '1984-06-25');
+      assert.equal(date_published, '1984-06-25T04:00:00.000Z');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/news.nationalgeographic.com/index.js

@@ -11,8 +11,7 @@ export const NewsNationalgeographicComExtractor = {
 
   date_published: {
     selectors: [['meta[name="article:published_time"]', 'value']],
-    format: 'ddd MMM DD HH:mm:ss zz YYYY',
-    timezone: 'EST',
+    timezone: 'America/New_York',
   },
 
   dek: {

src/extractors/custom/news.nationalgeographic.com/index.test.js

@@ -49,7 +49,7 @@ describe('NewsNationalgeographicComExtractor', () => {
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(date_published, '2015-08-03T17:45:00.000Z');
+      assert.equal(date_published, '2015-08-03T16:45:00.000Z');
     });
 
     it('returns the dek', async () => {

src/extractors/custom/people.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -55,13 +54,9 @@ describe('PeopleComExtractor', () => {
       // in ./src/extractors/custom/people.com/index.js.
       const { date_published } = await result;
 
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
-
       // Update these values with the expected values from
       // the article.
-      assert.equal(new_date_published, '2016-12-12');
+      assert.equal(date_published, '2016-12-12T14:22:00.000Z');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/pitchfork.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -41,11 +40,8 @@ describe('PitchforkComExtractor', () => {
 
     it('returns the date_published', async () => {
       const { date_published } = await result;
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
 
-      assert.equal(new_date_published, '2019-06-07');
+      assert.equal(date_published, '2019-06-07T04:00:00.000Z');
     });
 
     it('returns the dek', async () => {

src/extractors/custom/takagi-hiromitsu.jp/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -57,11 +56,10 @@ describe('TakagihiromitsuJpExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/takagi-hiromitsu.jp/index.js.
       const { date_published } = await result;
-      const newDatePublished = moment(date_published).format();
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(newDatePublished.split('T')[0], '2019-02-17');
+      assert.equal(date_published, '2019-02-17T14:34:06.000Z');
     });
 
     it('returns the dek', async () => {

src/extractors/custom/www.chicagotribune.com/index.js

@@ -11,6 +11,7 @@ export const WwwChicagotribuneComExtractor = {
 
   date_published: {
     selectors: ['time'],
+    timezone: 'America/Chicago',
   },
 
   lead_image_url: {

src/extractors/custom/www.chicagotribune.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -55,13 +54,10 @@ describe('WwwChicagotribuneComExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/www.chicagotribune.com/index.js.
       const { date_published } = await result;
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(new_date_published, '2016-12-13');
+      assert.equal(date_published, '2016-12-13T21:45:00.000Z');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/www.infoq.com/index.js

@@ -11,7 +11,7 @@ export const WwwInfoqComExtractor = {
 
   date_published: {
     selectors: ['.article__readTime.date'],
-    format: 'YYYY年MM月DD日',
+    format: 'YYYY[年]M[月]D[日]',
     timezone: 'Asia/Tokyo',
   },
 

src/extractors/custom/www.macrumors.com/index.js

@@ -11,8 +11,7 @@ export const WwwMacrumorsComExtractor = {
 
   date_published: {
     selectors: [['time', 'datetime']],
-
-    timezone: 'America/Los_Angeles',
+    // timezone: 'America/Los_Angeles',
   },
 
   dek: {

src/extractors/custom/www.msn.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -58,11 +57,10 @@ describe('MSNExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/www.msn.com/index.js.
       const { date_published } = await result;
-      const newDatePublished = moment(date_published).format();
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(newDatePublished.split('T')[0], '2016-09-21');
+      assert.equal(date_published.split('T')[0], '2016-09-21');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/www.nationalgeographic.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -44,13 +43,10 @@ describe('WwwNationalgeographicComExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/www.nationalgeographic.com/index.js.
       const { date_published } = await result;
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(new_date_published, '2016-12-15');
+      assert.equal(date_published.split('T')[0], '2016-12-15');
     });
 
     it('returns the dek', async () => {

src/extractors/custom/www.nbcnews.com/index.js

@@ -19,7 +19,7 @@ export const WwwNbcnewsComExtractor = {
       '.flag_article-wrapper time',
     ],
 
-    timezone: 'America/New_York',
+    // timezone: 'America/New_York',
   },
 
   lead_image_url: {

src/extractors/custom/www.nbcnews.com/index.test.js

@@ -53,7 +53,7 @@ describe('WwwNbcnewsComExtractor', () => {
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(date_published, '2016-12-13T23:06:00.000Z');
+      assert.equal(date_published, '2016-12-13T18:06:00.000Z');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/www.nydailynews.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -55,14 +54,11 @@ describe('WwwNydailynewsComExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/www.nydailynews.com/index.js.
       const { date_published } = await result;
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
 
       // Update these values with the expected values from
       // the article.
 
-      assert.equal(new_date_published, '2016-12-16');
+      assert.equal(date_published.split('T')[0], '2016-12-16');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/www.ossnews.jp/index.js

@@ -9,7 +9,7 @@ export const WwwOssnewsJpExtractor = {
 
   date_published: {
     selectors: ['p.fs12'],
-    format: 'YYYY年MM月DD日 HH:mm',
+    format: 'YYYY[年]M[月]D[日] HH:mm',
     timezone: 'Asia/Tokyo',
   },
 

src/extractors/custom/www.phoronix.com/index.js

@@ -12,7 +12,7 @@ export const WwwPhoronixComExtractor = {
   date_published: {
     selectors: ['.author'],
     // 1 June 2019 at 08:34 PM EDT
-    format: 'D MMMM YYYY at hh:mm',
+    format: 'D MMMM YYYY [at] hh:mm A',
     timezone: 'America/New_York',
   },
 

src/extractors/custom/www.phoronix.com/index.test.js

@@ -57,7 +57,7 @@ describe('WwwPhoronixComExtractor', () => {
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(date_published, '2019-06-01T12:34:00.000Z');
+      assert.equal(date_published, '2019-06-02T00:34:00.000Z');
     });
 
     it('returns the dek', async () => {

src/extractors/custom/www.politico.com/index.test.js

@@ -1,7 +1,6 @@
 import assert from 'assert';
 import URL from 'url';
 import cheerio from 'cheerio';
-import moment from 'moment-timezone';
 
 import Mercury from 'mercury';
 import getExtractor from 'extractors/get-extractor';
@@ -55,13 +54,10 @@ describe('PoliticoExtractor', () => {
       // To pass this test, fill out the date_published selector
       // in ./src/extractors/custom/www.politico.com/index.js.
       const { date_published } = await result;
-      const new_date_published = moment(date_published)
-        .format()
-        .split('T')[0];
 
       // Update these values with the expected values from
       // the article.
-      assert.equal(new_date_published, '2016-10-04');
+      assert.equal(date_published.split('T')[0], '2016-10-04');
     });
 
     it('returns the lead_image_url', async () => {

src/extractors/custom/www.prospectmagazine.co.uk/index.js

@@ -11,8 +11,6 @@ export const WwwProspectmagazineCoUkExtractor = {
 
   date_published: {
     selectors: [['meta[name="article:published_time"]', 'value'], '.post-info'],
-
-    timezone: 'Europe/London',
   },
 
   dek: {