Synchronize data with AWS

This pushes up records to AWS. It pivots the JSON on the way, using the case ID as each object's key.

sync.sh

@@ -0,0 +1,18 @@
+#!/usr/bin/env bash
+
+# Delete temporary files on exit
+trap 'rm -f "$tmp" "$tmp2"' EXIT
+
+# Rearrange the JSON to use case IDs as keys
+tmp=$(mktemp)
+jq 'reduce .[] as $item ({}; .[$item.case_id] = $item)' arguments.json > "$tmp"
+tmp2=$(mktemp)
+jq 'with_entries(.value |= del(.case_id))' "$tmp" > "$tmp2"
+
+# Copy the file
+aws s3 cp --no-progress "$tmp2" s3://courtaudio/arguments.json
+aws s3 sync --no-progress transcripts/ s3://courtaudio/transcripts/
+aws s3 sync --no-progress audio/ s3://courtaudio/audio/ --exclude "*" --include "*.mp3"
+
+# Invalidate the cached JSON
+aws cloudfront create-invalidation --distribution-id EM9XCQXYT1OYQ  --paths "/arguments.json"