add html parsing & file sending

lib/ics.js

@@ -0,0 +1,21 @@
+const ics = require('ics')
+
+const generateICS = async(data) => {
+  return new Promise((resolve, reject) => {
+    ics.createEvent({
+      title: data.title || '',
+      start: data.start || [],
+      duration: data.duration || { hours: 2 },
+      location: data.location || '',
+    }, (err, value) => {
+      if (err) {
+        reject(err)
+        return
+      }
+
+      resolve(value)
+    })
+  })
+}
+
+module.exports = generateICS

lib/index.js

@@ -4,6 +4,7 @@ const path = require('path')
 
 const crawl = require('./crawler')
 const parseHTML = require('./parser')
+const generateICS = require('./ics')
 
 const port = process.env.PORT || 3000
 const app = express()
@@ -17,19 +18,36 @@ app.get('/', (req, res) => {
 })
 
 app.get('*', (req, res) => {
-  res.render('404')
+  res.status(400).render('404')
 })
 
 app.post('/download', async (req, res) => {
   const { url } = req.body
 
+  if (!/facebook/.test(url)) {
+    return res
+      .status(500)
+      .render(
+        'error',
+        { error: 'Not Facebook URL!' }
+      )
+  }
+
   try {
     const html = await crawl(url)
     const data = parseHTML(html)
-    console.log(data)
+    const ics = await generateICS(data)
+
+    if (ics) {
+      return res
+        .contentType('text/calendar')
+        .send(200, new Buffer(ics, 'utf8'))
+    }
   } catch (err) {
     console.error(err)
-    return res.render('error', { error: err.toString() })
+    return res
+      .status(500)
+      .render('error', { error: err.toString() })
   }
 
   return res.render('download', { url })

lib/parser.js

@@ -1,22 +1,67 @@
-const { parse } = require('node-html-parser')
+const cheerio = require('cheerio')
+const dayjs = require('dayjs')
 
 const parseHTML = (html) => {
   let data = {}
 
-  const root = parse(html)
+  const root = cheerio.load(html)
-  const eventSummary = root.querySelector('#event_summary')
 
-  const timeInfo = eventSummary.querySelector('#event_time_info')
+  const $title = root('#event_header_primary h1')
-  const timeContent = timeInfo.querySelector('[content]')
+  const titleContent = $title ? $title.text() : null
 
-  if (
+  if (titleContent) {
-    timeContent &&
-    timeContent.attributes &&
-    timeContent.attributes.content
-  ) {
     data = {
       ...data,
-      time: timeContent.attributes.content.value,
+      title: titleContent,
+    }
+  }
+
+  const $timeInfo = root('#event_time_info')
+  const $time = $timeInfo.find('[content]')
+
+  const timeContent = $time ? $time.attr('content') : null
+
+  if (timeContent) {
+    const parsedTimeContent = timeContent
+      .split('to')
+      .map(part => part.trim())
+    const startDate = dayjs(parsedTimeContent[0]) || dayjs()
+    const endDate = dayjs(parsedTimeContent[1]) || dayjs()
+
+    const start = [
+      startDate.year(),
+      startDate.month(),
+      startDate.date(),
+      startDate.hour(),
+      startDate.minute(),
+    ]
+
+    const end = [
+      endDate.year(),
+      endDate.month(),
+      endDate.date(),
+      endDate.hour(),
+      endDate.minute()
+    ]
+
+    data = {
+      ...data,
+      start,
+      end,
+      // TODO: Create duration from end date
+      duration: { hours: 1 },
+    }
+  }
+
+  const $locations = $timeInfo.next().find('table tr td').get(1)
+  const $location = $locations.children[0].children[0]
+
+  const locationContent = $location ? root($location).text() : null
+
+  if (locationContent) {
+    data = {
+      ...data,
+      location: locationContent,
     }
   }
 

package-lock.json

@@ -4,6 +4,45 @@
   "lockfileVersion": 1,
   "requires": true,
   "dependencies": {
+    "@hapi/address": {
+      "version": "2.1.2",
+      "resolved": "https://registry.npmjs.org/@hapi/address/-/address-2.1.2.tgz",
+      "integrity": "sha512-O4QDrx+JoGKZc6aN64L04vqa7e41tIiLU+OvKdcYaEMP97UttL0f9GIi9/0A4WAMx0uBd6SidDIhktZhgOcN8Q=="
+    },
+    "@hapi/bourne": {
+      "version": "1.3.2",
+      "resolved": "https://registry.npmjs.org/@hapi/bourne/-/bourne-1.3.2.tgz",
+      "integrity": "sha512-1dVNHT76Uu5N3eJNTYcvxee+jzX4Z9lfciqRRHCU27ihbUcYi+iSc2iml5Ke1LXe1SyJCLA0+14Jh4tXJgOppA=="
+    },
+    "@hapi/hoek": {
+      "version": "8.3.1",
+      "resolved": "https://registry.npmjs.org/@hapi/hoek/-/hoek-8.3.1.tgz",
+      "integrity": "sha512-75ocgnI7HG/I01iGA3/rs0y6PXydUA/kxhFZM0HoT8NLSTnt/J8Gq03iKl4a4B/2A3iMG0ctXtxr5Hg9SGr1gw=="
+    },
+    "@hapi/joi": {
+      "version": "15.1.1",
+      "resolved": "https://registry.npmjs.org/@hapi/joi/-/joi-15.1.1.tgz",
+      "integrity": "sha512-entf8ZMOK8sc+8YfeOlM8pCfg3b5+WZIKBfUaaJT8UsjAAPjartzxIYm3TIbjvA4u+u++KbcXD38k682nVHDAQ==",
+      "requires": {
+        "@hapi/address": "2.x.x",
+        "@hapi/bourne": "1.x.x",
+        "@hapi/hoek": "8.x.x",
+        "@hapi/topo": "3.x.x"
+      }
+    },
+    "@hapi/topo": {
+      "version": "3.1.6",
+      "resolved": "https://registry.npmjs.org/@hapi/topo/-/topo-3.1.6.tgz",
+      "integrity": "sha512-tAag0jEcjwH+P2quUfipd7liWCNX2F8NvYjQp2wtInsZxnMlypdw0FtAOLxtvvkO+GSRRbmNi8m/5y42PQJYCQ==",
+      "requires": {
+        "@hapi/hoek": "^8.3.0"
+      }
+    },
+    "@types/node": {
+      "version": "12.7.12",
+      "resolved": "https://registry.npmjs.org/@types/node/-/node-12.7.12.tgz",
+      "integrity": "sha512-KPYGmfD0/b1eXurQ59fXD1GBzhSQfz6/lKBxkaHX9dKTzjXbK68Zt7yGUxUsCS1jeTy/8aL+d9JEr+S54mpkWQ=="
+    },
     "abbrev": {
       "version": "1.1.1",
       "resolved": "https://registry.npmjs.org/abbrev/-/abbrev-1.1.1.tgz",
@@ -207,6 +246,11 @@
         "type-is": "~1.6.17"
       }
     },
+    "boolbase": {
+      "version": "1.0.0",
+      "resolved": "https://registry.npmjs.org/boolbase/-/boolbase-1.0.0.tgz",
+      "integrity": "sha1-aN/1++YMUes3cl6p4+0xDcwed24="
+    },
     "boxen": {
       "version": "1.3.0",
       "resolved": "https://registry.npmjs.org/boxen/-/boxen-1.3.0.tgz",
@@ -310,6 +354,19 @@
         "supports-color": "^5.3.0"
       }
     },
+    "cheerio": {
+      "version": "1.0.0-rc.3",
+      "resolved": "https://registry.npmjs.org/cheerio/-/cheerio-1.0.0-rc.3.tgz",
+      "integrity": "sha512-0td5ijfUPuubwLUu0OBoe98gZj8C/AA+RW3v67GPlGOrvxWjZmBXiBCRU+I8VEiNyJzjth40POfHiz2RB3gImA==",
+      "requires": {
+        "css-select": "~1.2.0",
+        "dom-serializer": "~0.1.1",
+        "entities": "~1.1.1",
+        "htmlparser2": "^3.9.1",
+        "lodash": "^4.15.0",
+        "parse5": "^3.0.1"
+      }
+    },
     "chokidar": {
       "version": "2.1.8",
       "resolved": "https://registry.npmjs.org/chokidar/-/chokidar-2.1.8.tgz",
@@ -486,6 +543,27 @@
       "integrity": "sha1-ojD2T1aDEOFJgAmUB5DsmVRbyn4=",
       "dev": true
     },
+    "css-select": {
+      "version": "1.2.0",
+      "resolved": "https://registry.npmjs.org/css-select/-/css-select-1.2.0.tgz",
+      "integrity": "sha1-KzoRBTnFNV8c2NMUYj6HCxIeyFg=",
+      "requires": {
+        "boolbase": "~1.0.0",
+        "css-what": "2.1",
+        "domutils": "1.5.1",
+        "nth-check": "~1.0.1"
+      }
+    },
+    "css-what": {
+      "version": "2.1.3",
+      "resolved": "https://registry.npmjs.org/css-what/-/css-what-2.1.3.tgz",
+      "integrity": "sha512-a+EPoD+uZiNfh+5fxw2nO9QwFa6nJe2Or35fGY6Ipw1R3R4AGz1d1TEZrCegvw2YTmZ0jXirGYlzxxpYSHwpEg=="
+    },
+    "dayjs": {
+      "version": "1.8.16",
+      "resolved": "https://registry.npmjs.org/dayjs/-/dayjs-1.8.16.tgz",
+      "integrity": "sha512-XPmqzWz/EJiaRHjBqSJ2s6hE/BUoCIHKgdS2QPtTQtKcS9E4/Qn0WomoH1lXanWCzri+g7zPcuNV4aTZ8PMORQ=="
+    },
     "debug": {
       "version": "2.6.9",
       "resolved": "https://registry.npmjs.org/debug/-/debug-2.6.9.tgz",
@@ -557,6 +635,37 @@
       "resolved": "https://registry.npmjs.org/destroy/-/destroy-1.0.4.tgz",
       "integrity": "sha1-l4hXRCxEdJ5CBmE+N5RiBYJqvYA="
     },
+    "dom-serializer": {
+      "version": "0.1.1",
+      "resolved": "https://registry.npmjs.org/dom-serializer/-/dom-serializer-0.1.1.tgz",
+      "integrity": "sha512-l0IU0pPzLWSHBcieZbpOKgkIn3ts3vAh7ZuFyXNwJxJXk/c4Gwj9xaTJwIDVQCXawWD0qb3IzMGH5rglQaO0XA==",
+      "requires": {
+        "domelementtype": "^1.3.0",
+        "entities": "^1.1.1"
+      }
+    },
+    "domelementtype": {
+      "version": "1.3.1",
+      "resolved": "https://registry.npmjs.org/domelementtype/-/domelementtype-1.3.1.tgz",
+      "integrity": "sha512-BSKB+TSpMpFI/HOxCNr1O8aMOTZ8hT3pM3GQ0w/mWRmkhEDSFJkkyzz4XQsBV44BChwGkrDfMyjVD0eA2aFV3w=="
+    },
+    "domhandler": {
+      "version": "2.4.2",
+      "resolved": "https://registry.npmjs.org/domhandler/-/domhandler-2.4.2.tgz",
+      "integrity": "sha512-JiK04h0Ht5u/80fdLMCEmV4zkNh2BcoMFBmZ/91WtYZ8qVXSKjiw7fXMgFPnHcSZgOo3XdinHvmnDUeMf5R4wA==",
+      "requires": {
+        "domelementtype": "1"
+      }
+    },
+    "domutils": {
+      "version": "1.5.1",
+      "resolved": "https://registry.npmjs.org/domutils/-/domutils-1.5.1.tgz",
+      "integrity": "sha1-3NhIiib1Y9YQeeSMn3t+Mjc2gs8=",
+      "requires": {
+        "dom-serializer": "0",
+        "domelementtype": "1"
+      }
+    },
     "dot-prop": {
       "version": "4.2.0",
       "resolved": "https://registry.npmjs.org/dot-prop/-/dot-prop-4.2.0.tgz",
@@ -587,6 +696,11 @@
       "resolved": "https://registry.npmjs.org/encodeurl/-/encodeurl-1.0.2.tgz",
       "integrity": "sha1-rT/0yG7C0CkyL1oCw6mmBslbP1k="
     },
+    "entities": {
+      "version": "1.1.2",
+      "resolved": "https://registry.npmjs.org/entities/-/entities-1.1.2.tgz",
+      "integrity": "sha512-f2LZMYl1Fzu7YSBKg+RoROelpOaNrcGmE9AZubeDfrCEia483oW4MI4VyFd5VNHIgQ/7qm1I0wUHK1eJnn2y2w=="
+    },
     "es6-promise": {
       "version": "4.2.8",
       "resolved": "https://registry.npmjs.org/es6-promise/-/es6-promise-4.2.8.tgz",
@@ -1541,10 +1655,30 @@
         }
       }
     },
-    "he": {
+    "htmlparser2": {
-      "version": "1.1.1",
+      "version": "3.10.1",
-      "resolved": "https://registry.npmjs.org/he/-/he-1.1.1.tgz",
+      "resolved": "https://registry.npmjs.org/htmlparser2/-/htmlparser2-3.10.1.tgz",
-      "integrity": "sha1-k0EP0hsAlzUVH4howvJx80J+I/0="
+      "integrity": "sha512-IgieNijUMbkDovyoKObU1DUhm1iwNYE/fuifEoEHfd1oZKZDaONBSkal7Y01shxsM49R4XaMdGez3WnF9UfiCQ==",
+      "requires": {
+        "domelementtype": "^1.3.1",
+        "domhandler": "^2.3.0",
+        "domutils": "^1.5.1",
+        "entities": "^1.1.1",
+        "inherits": "^2.0.1",
+        "readable-stream": "^3.1.1"
+      },
+      "dependencies": {
+        "readable-stream": {
+          "version": "3.4.0",
+          "resolved": "https://registry.npmjs.org/readable-stream/-/readable-stream-3.4.0.tgz",
+          "integrity": "sha512-jItXPLmrSR8jmTRmRWJXCnGJsfy85mB3Wd/uINMXA65yrnFo0cPClFIUWzo2najVNSl+mx7/4W8ttlLWJe99pQ==",
+          "requires": {
+            "inherits": "^2.0.3",
+            "string_decoder": "^1.1.1",
+            "util-deprecate": "^1.0.1"
+          }
+        }
+      }
     },
     "http-errors": {
       "version": "1.7.2",
@@ -1590,6 +1724,17 @@
         "safer-buffer": ">= 2.1.2 < 3"
       }
     },
+    "ics": {
+      "version": "2.18.0",
+      "resolved": "https://registry.npmjs.org/ics/-/ics-2.18.0.tgz",
+      "integrity": "sha512-QEhHNWk8Bh/ukjnhXIDGoaphnPA6yiCUiym6k9slqv60SBfMVM7uM2jzLOpqQzgq+YArde3Qyf0+fz7XxLmPDQ==",
+      "requires": {
+        "@hapi/joi": "^15.1.1",
+        "lodash": "^4.17.15",
+        "moment": "^2.24.0",
+        "uuid": "^3.3.3"
+      }
+    },
     "ignore-by-default": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/ignore-by-default/-/ignore-by-default-1.0.1.tgz",
@@ -1859,6 +2004,11 @@
         "package-json": "^4.0.0"
       }
     },
+    "lodash": {
+      "version": "4.17.15",
+      "resolved": "https://registry.npmjs.org/lodash/-/lodash-4.17.15.tgz",
+      "integrity": "sha512-8xOcRHvCjnocdS5cpwXQXVzmmh5e5+saE2QGoeQmbKmRS6J3VQppPOIt0MnmE+4xlZoumy0GPG0D0MVIQbNA1A=="
+    },
     "lowercase-keys": {
       "version": "1.0.1",
       "resolved": "https://registry.npmjs.org/lowercase-keys/-/lowercase-keys-1.0.1.tgz",
@@ -1995,6 +2145,11 @@
         "minimist": "0.0.8"
       }
     },
+    "moment": {
+      "version": "2.24.0",
+      "resolved": "https://registry.npmjs.org/moment/-/moment-2.24.0.tgz",
+      "integrity": "sha512-bV7f+6l2QigeBBZSM/6yTNq4P2fNpSWj/0e7jQcy87A8e7o2nAfP/34/2ky5Vw4B9S446EtIhodAzkFCcR4dQg=="
+    },
     "ms": {
       "version": "2.0.0",
       "resolved": "https://registry.npmjs.org/ms/-/ms-2.0.0.tgz",
@@ -2031,14 +2186,6 @@
       "resolved": "https://registry.npmjs.org/negotiator/-/negotiator-0.6.2.tgz",
       "integrity": "sha512-hZXc7K2e+PgeI1eDBe/10Ard4ekbfrrqG8Ep+8Jmf4JID2bNg7NvCPOZN+kfF574pFQI7mum2AUqDidoKqcTOw=="
     },
-    "node-html-parser": {
-      "version": "1.1.16",
-      "resolved": "https://registry.npmjs.org/node-html-parser/-/node-html-parser-1.1.16.tgz",
-      "integrity": "sha512-cfqTZIYDdp5cGh3NvCD5dcEDP7hfyni7WgyFacmDynLlIZaF3GVlRk8yMARhWp/PobWt1KaCV8VKdP5LKWiVbg==",
-      "requires": {
-        "he": "1.1.1"
-      }
-    },
     "nodemon": {
       "version": "1.19.3",
       "resolved": "https://registry.npmjs.org/nodemon/-/nodemon-1.19.3.tgz",
@@ -2098,6 +2245,14 @@
         "path-key": "^2.0.0"
       }
     },
+    "nth-check": {
+      "version": "1.0.2",
+      "resolved": "https://registry.npmjs.org/nth-check/-/nth-check-1.0.2.tgz",
+      "integrity": "sha512-WeBOdju8SnzPN5vTUJYxYUxLeXpCaVP5i5e0LF8fg7WORF2Wd7wFX/pk0tYZk7s8T+J7VLy0Da6J1+wCT0AtHg==",
+      "requires": {
+        "boolbase": "~1.0.0"
+      }
+    },
     "object-copy": {
       "version": "0.1.0",
       "resolved": "https://registry.npmjs.org/object-copy/-/object-copy-0.1.0.tgz",
@@ -2181,6 +2336,14 @@
         "semver": "^5.1.0"
       }
     },
+    "parse5": {
+      "version": "3.0.3",
+      "resolved": "https://registry.npmjs.org/parse5/-/parse5-3.0.3.tgz",
+      "integrity": "sha512-rgO9Zg5LLLkfJF9E6CCmXlSE4UVceloys8JrFqCcHloC3usd/kJCyPDwH2SOlzix2j3xaP9sUX3e8+kvkuleAA==",
+      "requires": {
+        "@types/node": "*"
+      }
+    },
     "parseurl": {
       "version": "1.3.3",
       "resolved": "https://registry.npmjs.org/parseurl/-/parseurl-1.3.3.tgz",
@@ -3004,6 +3167,11 @@
       "resolved": "https://registry.npmjs.org/utils-merge/-/utils-merge-1.0.1.tgz",
       "integrity": "sha1-n5VxD1CiZ5R7LMwSR0HBAoQn5xM="
     },
+    "uuid": {
+      "version": "3.3.3",
+      "resolved": "https://registry.npmjs.org/uuid/-/uuid-3.3.3.tgz",
+      "integrity": "sha512-pW0No1RGHgzlpHJO1nsVrHKpOEIxkGg1xB+v0ZmdNH5OAeAwzAVrCnI2/6Mtx+Uys6iaylxa+D3g4j63IKKjSQ=="
+    },
     "vary": {
       "version": "1.1.2",
       "resolved": "https://registry.npmjs.org/vary/-/vary-1.1.2.tgz",

package.json

@@ -6,6 +6,7 @@
   "scripts": {
     "start": "PORT=80 node lib/index.js",
     "start:dev": "nodemon lib/index.js",
+    "start:dev:inspect": "nodemon --inspect lib/index.js",
     "test": "echo \"Error: no test specified\" && exit 1"
   },
   "keywords": [
@@ -19,9 +20,11 @@
   "license": "ISC",
   "dependencies": {
     "body-parser": "^1.19.0",
+    "cheerio": "^1.0.0-rc.3",
+    "dayjs": "^1.8.16",
     "ejs": "^2.7.1",
     "express": "^4.17.1",
-    "node-html-parser": "^1.1.16",
+    "ics": "^2.18.0",
     "puppeteer": "^1.20.0"
   },
   "devDependencies": {