From e80e1ebfc9754dd2338bf31ff4377f0b3cbe23b6 Mon Sep 17 00:00:00 2001
From: Zhanna <zklimanova@yahoo.com>
Date: Wed, 14 Jun 2023 22:50:06 -0400
Subject: [PATCH] Add link checker for testing

---
 .github/workflows/broken-link-checker.yml | 71 +++++++++++++++++++++++
 1 file changed, 71 insertions(+)
 create mode 100644 .github/workflows/broken-link-checker.yml

diff --git a/.github/workflows/broken-link-checker.yml b/.github/workflows/broken-link-checker.yml
new file mode 100644
index 000000000..f3c73e962
--- /dev/null
+++ b/.github/workflows/broken-link-checker.yml
@@ -0,0 +1,71 @@
+name: Check for Broken Links
+on: [push, pull_request]
+jobs:
+  build_and_check:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout repository
+        uses: actions/checkout@v2
+
+      - name: Install Node.js
+        uses: actions/setup-node@v2
+        with:
+          node-version: 14
+
+      - name: Install dependencies
+        run: yarn install
+
+      - name: Build website
+        run: yarn build
+
+      - name: Setup Simple HTTP Server
+        run: |
+          nohup yarn start &
+
+      - name: Check HTTP Server status
+        run: sleep 5 && curl -I http://127.0.0.1:8000
+
+      - name: Install Broken Link Checker
+        run: npm install -g broken-link-checker
+
+      - name: Execute Link Checker and Show Broken Links
+        env:
+          TEMPORARY_WEBSITE_URL: "http://127.0.0.1:8080"
+          ACTUAL_WEBSITE_URL: "https://ddmal.music.mcgill.ca/Neon/"
+        run: |
+          blc $TEMPORARY_WEBSITE_URL --filter-level=3 | \
+          grep -v "$ACTUAL_WEBSITE_URL" | \
+          awk '
+          BEGIN {
+            p=1;
+            buf=""
+          }
+          /^Getting links from:/ {
+            buf=$0;
+            next
+          }
+          /^Finished!.*0 broken\./ {
+            if (length(buf)>0) {
+              buf="";
+              next
+            }
+          }
+          {
+            if(length(buf)>0)
+              print buf;
+            if (NF > 0)
+              print;
+            buf=""
+          }
+          /^Finished!/ {
+            print ""
+          }' | sed "s|$TEMPORARY_WEBSITE_URL|$ACTUAL_WEBSITE_URL|g" | \
+          tee /dev/fd/2 | grep 'HTTP_404' > /dev/null && flag=1
+
+          if [ "$flag" -eq "1" ]; then
+              echo "Broken links were found, exiting with an error."
+              exit 1
+          else
+              echo "No broken links were found (although there may be other HTTP errors), exiting successfully."
+              exit 0
+          fi