Reimplementation with new scraper in node.js

.github/workflows/docker-publish.yml

@@ -2,48 +2,61 @@ name: Docker
 
 on:
   push:
-    # Publish `main` as Docker `latest` image.
-    branches:
-      - main
+    branches: [ "main" ]
+  pull_request:
+    branches: [ "main" ]
+  workflow_dispatch:
+
+env:
+  # Use docker.io for Docker Hub if empty
+  REGISTRY: ghcr.io
+  # github.repository as <account>/<repo>
+  IMAGE_NAME: ${{ github.repository }}
 
 jobs:
   # Push image to GitHub Packages.
   # See also https://docs.docker.com/docker-hub/builds/
-  push:
-
-    runs-on: ubuntu-20.04
+  build:
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
 
     steps:
-      - name: Checkout
-        uses: actions/checkout@v2
+      - name: Checkout repository
+        uses: actions/checkout@v4
+
+      # Workaround: https://github.com/docker/build-push-action/issues/461
+      - name: Setup Docker buildx
+        uses: docker/setup-buildx-action@v3
 
-      - name: Login to DockerHub
-        uses: docker/login-action@v1
+      # Login against a Docker registry except on PR
+      # https://github.com/docker/login-action
+      - name: Log into registry ${{ env.REGISTRY }}
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
         with:
-          username: ${{ secrets.DOCKERHUB_USER }}
-          password: ${{ secrets.DOCKERHUB_TOKEN }}
-
-      - name: Setup Docker build
-        id: buildx
-        uses: docker/setup-buildx-action@v1
-
-      - name: Build and push backend
-        id: docker_build_backend
-        uses: docker/build-push-action@v2
+          registry: ${{ env.REGISTRY }}
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      # Extract metadata (tags, labels) for Docker
+      # https://github.com/docker/metadata-action
+      - name: Extract Docker metadata
+        id: meta
+        uses: docker/metadata-action@v5
         with:
-          context: ./lasvecka-python
-          file: ./lasvecka-python/prod.Dockerfile
-          push: true
-          tags: gudchs/lasvecka_backend:latest
-
-      - name: Build and push frontend
-        id: docker_build_frontend
-        uses: docker/build-push-action@v2
+          images: ${{ env.REGISTRY }}/${{ env.IMAGE_NAME }}
+
+      # Build and push Docker image with Buildx (don't push on PR)
+      # https://github.com/docker/build-push-action
+      - name: Build and push Docker image
+        id: build-and-push
+        uses: docker/build-push-action@v5
         with:
-          context: ./lasvecka-react
-          file: ./lasvecka-react/prod.Dockerfile
-          push: true
-          tags: gudchs/lasvecka_frontend:latest
-
-      - name: Image digest
-        run: echo ${{ steps.docker_build.outputs.digest }}
+          context: ./lasvecka-node
+          push: ${{ github.event_name != 'pull_request' }}
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max

dev.docker-compose.yaml

@@ -1,23 +1,12 @@
-version: '2'
+version: "3"
 
 services:
-  backend:
-    container_name: lasvecka-python
+  app:
+    container_name: lasvecka-node
     build:
-      context: ./lasvecka-python
-      dockerfile: dev.Dockerfile
-    ports:
-      - 5000:5000
-    volumes:
-      - ./backend:/lasvecka-python
-
-  frontend:
-    container_name: lasvecka-react
-    build:
-      context: ./lasvecka-react
-      dockerfile: dev.Dockerfile
-    stdin_open: true
+      context: ./lasvecka-node
+      dockerfile: Dockerfile
     ports:
       - 3000:3000
     volumes:
-      - ./frontend:/usr/src/lasveckor/frontend
+      - ./data:/usr/src/app/data

lasvecka-node/Dockerfile

@@ -0,0 +1,19 @@
+FROM node:lts
+
+# Create app directory
+WORKDIR /usr/src/app
+
+# Install app dependencies
+# A wildcard is used to ensure both package.json AND package-lock.json are copied
+# where available (npm@5+)
+COPY package*.json ./
+
+RUN npm install
+# If you are building your code for production
+# RUN npm ci --only=production
+
+# Bundle app source
+COPY . .
+
+EXPOSE 3000
+CMD [ "node", "index.js" ]

lasvecka-node/calc_date.js

@@ -0,0 +1,81 @@
+const moment = require('moment');
+const fs = require('fs');
+const scrape = require('./lasveckor_scraper.js');
+var dateDict = {};
+
+// check if data.json exists
+if (!fs.existsSync('./data/data.json')) {
+  scrape().then((res) => {
+    dateDict = res
+  });
+} else {
+  dateDict = JSON.parse(fs.readFileSync('./data/data.json'));
+
+  //check if it was updated after 1/7 this year and if 1/7 has occured this year
+  let updated = moment(dateDict["updated"]);
+  let firstOfJuly = moment().month(6).date(1);
+  if (updated.isBefore(firstOfJuly) && moment().isAfter(firstOfJuly)) {
+    scrape().then((res) => {
+      dateDict = res
+    });
+  }
+}
+
+function readDatePeriod(currDate) {
+  let soughtDate = "";
+  let diff = -1000;
+  for (let dat in dateDict) {
+    // if dat is updated, easter_start, easter_end or ord_cont then skip
+    if (dat === "updated" || dat === "easter_start" || dat === "easter_end" || dat === "ord_cont") {
+      continue;
+    }
+    let deltaT = moment(dat).diff(currDate, 'days');
+    if (deltaT === 0) {
+      return { date: dat, type: dateDict[dat] };
+    } else if (deltaT > diff && deltaT < 0) {
+      soughtDate = dat;
+      diff = deltaT;
+    }
+  }
+  return { date: soughtDate, type: dateDict[soughtDate] };
+}
+
+function handleEaster(easterStartDiff, easterEndDiff) {
+  if (easterStartDiff >= 0 && easterEndDiff <= 0) {
+    return "Självstudier";
+  } else if (easterEndDiff > 0) {
+    let weeks = Math.floor(easterEndDiff / 7);
+    return "Lv " + (weeks + 4);
+  }
+}
+
+function computeTime() {
+  // Find date where value is easter_start, easter_end and ord_cont in json file
+  let EASTER_START = dateDict["easter_start"]
+  let ORD_CONT = dateDict["ord_cont"]
+  let currentDate = moment();
+  let easterEndCheck = currentDate.diff(ORD_CONT, 'days');
+  let easterStartCheck = currentDate.diff(EASTER_START, 'days');
+  let { date: dat, type: typ } = readDatePeriod(currentDate);
+  if (typ === "exam_period") {
+    let deltaT = currentDate.diff(moment(dat), 'days');
+    if (deltaT > 7) {
+      return "Självstudier";
+    } else {
+      return "Tentavecka";
+    }
+  }
+  if (easterEndCheck >= 0 || easterStartCheck >= 0) {
+    return handleEaster(easterStartCheck, easterEndCheck);
+  } else {
+    let deltaT = currentDate.diff(moment(dat), 'days');
+    let weeks = Math.floor(deltaT / 7);
+    if (weeks > 7) {
+      return "Självstudier";
+    } else {
+      return "LV " + (weeks + 1);
+    }
+  }
+}
+
+module.exports = computeTime;

lasvecka-node/data/data.json

@@ -0,0 +1 @@
+{"2023-08-28":"study_period","2023-10-21":"exam_period","2023-10-30":"study_period","2024-01-08":"exam_period","2024-01-15":"study_period","2024-03-09":"exam_period","2024-03-18":"study_period","easter_start":"2024-04-03","easter_end":"2024-04-05","ord_cont":"2024-04-08","2024-05-25":"exam_period","updated":"2023-11-06"}

lasvecka-node/index.js

@@ -0,0 +1,91 @@
+const express = require('express');
+const moment = require('moment');
+const app = express();
+const computeTime = require('./calc_date.js');
+const port = process.env.PORT || 3000;
+
+app.use(express.static('public'));
+
+app.get('/', (req, res) => {
+  let studyweek = computeTime();
+  let studyweekNum = studyweek.replace('LV ', '').replace('Självstudier', 'S').replace('Tentavecka', 'T');
+  // eg. 2023-W45
+  let week = moment().format('YYYY-[W]WW');
+  let data = { studyweek, week, studyweekNum };
+  res.send(render(data));
+});
+
+app.get('/api', (req, res) => {
+  res.send(computeTime());
+});
+
+app.listen(port, () => {
+  console.log(`App listening at http://localhost:${port}`)
+})
+
+const render = (data) => {
+  const html = `
+<!DOCTYPE html>
+<html>
+  <head>
+    <meta charset="UTF-8">
+    <title>läsvecka.nu | ${data.studyweek}</title>
+    <link rel="apple-touch-icon" sizes="180x180" href="/apple-touch-icon.png">
+    <link rel="icon" type="image/png" sizes="32x32" href="/favicon-32x32.png">
+    <link rel="icon" type="image/png" sizes="16x16" href="/favicon-16x16.png" id="favicon">
+    <link rel="manifest" href="/site.webmanifest">
+    <meta name="description" content="läsvecka.nu ger dig aktuell läsvecka, utan några konstigheter!">
+    <meta property="og:type" content="website">
+    <meta property="og:url" content="https://lasvecka.nu/">
+    <meta property="og:title" content="läsvecka.nu | ${data.studyweek}">
+    <style>
+      html, body { height: 100%; background-color: #90c0de; overflow: hidden; }
+      time {
+        position: absolute;
+        top: 50%;
+        left: 0;
+        right: 0;
+        margin: -110px 0 0 0;
+        height: 220px;
+        text-align: center;
+        color: #1c7bb7;
+        font-family: Arial, serif;
+        font-size: 260px;
+        line-height: 227px;
+        font-weight: bold;
+      }
+    </style>
+  </head>
+  <body>
+    <time datetime="${data.week}">${data.studyweek}</time>
+    <script>
+      (function () {
+        document.head = document.head || document.getElementsByTagName('head')[0];
+        var canvas = document.createElement('canvas'),
+          img = document.createElement('img'),
+          oldLink = document.getElementById('favicon'),
+          link = oldLink.cloneNode(true),
+          week = '${data.studyweekNum}';
+        if (canvas.getContext) {
+          canvas.height = canvas.width = 16;
+          var ctx = canvas.getContext('2d'),
+            textWidth = ctx.measureText(week).width;
+          img.onload = function () {
+            ctx.drawImage(this, 0, 0);
+            ctx.font = 'bold 10px "helvetica", sans-serif';
+            ctx.fillStyle = '#066EB0';
+            ctx.fillText(week, (canvas.width / 2) - (textWidth / 2), 12);
+            link.href = canvas.toDataURL('image/png');
+            document.head.removeChild(oldLink);
+            document.head.appendChild(link);
+          };
+          img.src = 'faviconbkg.png';
+        }
+      })();
+
+      setTimeout(function () { location.reload(); }, 25886000);
+    </script>
+  </body>
+</html>`;
+  return html.replace(/\n/g, '').replace(/\r/g, '').replace(/\t/g, '').replace(/ {2,}/g, '');
+}

lasvecka-node/lasveckor_scraper.js

@@ -0,0 +1,51 @@
+const fs = require('fs');
+const axios = require('axios');
+
+async function scrape() {
+  console.log("Scraping data from student.chalmers.se")
+  let url = "https://www.student.chalmers.se/sp/academic_year_list"
+
+  const response = await axios.get(url, { responseType: "arraybuffer" })
+  let data = response.data.toString('latin1');
+
+  data = data.replace(/\n/g, '').replace(/\r/g, '').replace(/\t/g, '');
+  data = data.match(/<table width="100%" border="0" cellspacing="0" cellpadding="3">(.*?)<\/table>/s)[0]
+
+  // some things to make it nicer
+  data = data.replace(/<\/tr>/g, '\n')
+  data = data.replace(/<tr align="left">/g, '')
+  data = data.replace(/<tr align="left" class="fade">/g, '')
+  data = data.replace(/<td>/g, '')
+  data = data.replace(/<\/td>/g, '')
+  data = data.split('\n')
+
+  let result = {}
+  for (let i = 0; i < data.length; i++) {
+    let line = data[i]
+    if (line.startsWith('Läsperiod')) {
+      let date = line.match(/(\d{4}-\d{2}-\d{2})/)[0]
+      result[date] = "study_period"
+    } else if (line.startsWith('Tentamensperiod')) {
+      let date = line.match(/(\d{4}-\d{2}-\d{2})/)[0]
+      result[date] = "exam_period"
+    } else if (line.startsWith("Omtentamensperiod påsk")) {
+      let date1 = line.match(/(\d{4}-\d{2}-\d{2})/)[0]
+      let date2 = line.match(/(\d{4}-\d{2}-\d{2})/g)[1]
+      result["easter_start"] = date1
+      result["easter_end"] = date2
+
+      var easter_start = new Date(date1)
+      // find first monday after easter
+      while (easter_start.getDay() != 1) {
+        easter_start.setDate(easter_start.getDate() + 1)
+      }
+      result["ord_cont"] = easter_start.toISOString().slice(0, 10)
+    }
+  }
+
+  result["updated"] = new Date().toISOString().slice(0, 10)
+  await fs.promises.writeFile('./data/data.json', JSON.stringify(result))
+  return result
+}
+
+module.exports = scrape;