Add Azure AI Search trigger for ops-runbook index (#678)

* Add Azure AI Search trigger for ops-runbook index

Add .ci scripts (install-htmlq, install-azcopy, upload, trigger-index)
and wire them into the Pages deploy workflow, mirroring the pattern
used in hmcts.github.io for the-hmcts-way index. Uploads built HTML
to the ops-runbook blob container and triggers the ops-runbook
indexer after each deploy.

* adding spelling to exceptions file

Author: niallk

.ci/install-azcopy.sh

@@ -0,0 +1,11 @@
+#!/usr/bin/env bash
+set -ex
+
+sudo apt-get update
+sudo apt-get install -y wget curl
+
+wget https://packages.microsoft.com/config/ubuntu/22.04/packages-microsoft-prod.deb
+sudo dpkg -i packages-microsoft-prod.deb
+sudo apt-get update
+sudo apt-get install -y azcopy
+rm -f packages-microsoft-prod.deb

.ci/install-htmlq.sh

@@ -0,0 +1,8 @@
+#!/usr/bin/env bash
+
+set -ex
+
+wget https://github.com/mgdm/htmlq/releases/download/v0.4.0/htmlq-x86_64-linux.tar.gz
+tar -xzf htmlq-x86_64-linux.tar.gz
+sudo install htmlq /usr/bin/htmlq
+rm -f htmlq-x86_64-linux.tar.gz

.ci/trigger-index.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+set -e
+
+API_KEY=$(az search admin-key show \
+  --service-name platops-slack-help-bot-ptl \
+  --resource-group slack-help-bot-cftptl-intsvc-rg \
+  --subscription DTS-CFTPTL-INTSVC \
+  --query primaryKey \
+  -o tsv)
+
+curl --fail-with-body \
+  -X POST \
+  -H "api-key: ${API_KEY}" \
+  "https://platops-slack-help-bot-ptl.search.windows.net/indexers('ops-runbook')/search.run?api-version=2024-07-01" \
+  -d ''
+
+echo "Index triggered"

.ci/upload.sh

@@ -0,0 +1,48 @@
+#!/usr/bin/env bash
+
+set -e
+
+rm -rf dist
+mkdir dist/
+
+function extractMain () {
+    local path=$1
+
+    DIRECTORY=$(dirname "$path")
+    mkdir -p dist/$DIRECTORY
+
+    cat $path | htmlq --text 'main' --remove-nodes '[data-module="page-expiry"]' > dist/$path
+}
+
+export -f extractMain
+
+find build -name "*.html" -not -path "build/search/*" -exec bash -c "extractMain \"{}\"" \;
+
+export STORAGE_ACCOUNT_NAME=platopslackhelpbotai
+export CONTAINER_NAME=ops-runbook
+export AZCOPY_AUTO_LOGIN_TYPE=AZCLI
+export AZCOPY_TENANT_ID=531ff96d-0ae9-462a-8d2d-bec7c0b42082
+
+# we use sync here so that deleted files get removed
+azcopy sync --compare-hash=md5 \
+  --delete-destination=true \
+  --include-pattern="*.html"  \
+  "dist/" "https://$STORAGE_ACCOUNT_NAME.blob.core.windows.net/$CONTAINER_NAME/"
+
+function setTitle() {
+    local path=$1
+
+    TITLE=$(cat ../$path | htmlq 'h1' --text)
+
+    echo "Setting title to $TITLE for $path"
+
+    az storage blob metadata update --auth-mode login --account-name $STORAGE_ACCOUNT_NAME --container-name $CONTAINER_NAME --name $path --metadata title="$TITLE"
+}
+
+export -f setTitle
+
+pushd dist
+# another pass over to set the file title as a metadata attribute
+find . -name "*.html" -exec bash -c "setTitle \"{}\"" \;
+
+popd

.github/actions/spelling/allow.txt

@@ -793,6 +793,11 @@ sas
 FLUSHALL
 PDB
 PDBs
+AZCLI
+azcopy
+platopslackhelpbotai
+popd
+pushd
 XPOST
 sitamp
 sitamp01

.github/workflows/deploy.yaml

@@ -47,3 +47,22 @@ jobs:
       - name: Deploy to GitHub Pages
         id: deployment
         uses: actions/deploy-pages@v4
+
+      # For Azure AI Search used in hmcts/slack-help-bot
+      - name: Azure login
+        uses: azure/login@v3
+        with:
+          client-id: ${{ secrets.AZURE_CLIENT_ID }}
+          tenant-id: ${{ secrets.AZURE_TENANT_ID }}
+          subscription-id: ${{ secrets.AZURE_SUBSCRIPTION_ID }}
+      - name: Install dependencies
+        run: |
+          ./.ci/install-htmlq.sh
+          ./.ci/install-azcopy.sh
+      - name: Process content
+        run: |
+          ./.ci/upload.sh
+      - name: Trigger index
+        run: |
+          ./.ci/trigger-index.sh
+      # end of Azure AI Search