Download from tuckercarlson.com using browserless and yt-dlp

tucker.sh

#!/usr/bin/env bash
# tucker-dl.sh - Download Tucker Carlson videos using browserless + yt-dlp

set -Eeuo pipefail
shopt -s inherit_errexit nullglob 2>/dev/null || true
trap 'echo "❌ ERROR: Script failed at line $LINENO (exit code $?)" >&2' ERR

for cmd in curl jq yt-dlp; do
    command -v "$cmd" >/dev/null || { echo "❌ ERROR: '$cmd' is required." >&2; exit 1; }
done

# ==================== CONFIG (ENV > config file > defaults) ====================
BROWSERLESS_ORIGIN="${BROWSERLESS_ORIGIN:-http://localhost:3000}"
BROWSERLESS_TOKEN="${BROWSERLESS_TOKEN:-}"

config_file="${XDG_CONFIG_HOME:-$HOME/.config}/tuckerdl.conf"
[[ -r "$config_file" ]] && while IFS= read -r line || [[ -n $line ]]; do
    [[ $line =~ ^[[:space:]]*(BROWSERLESS_(TOKEN|ORIGIN))=(.*)$ ]] && declare "${BASH_REMATCH[1]}=${BASH_REMATCH[3]}"
done < <(grep -E '^[[:space:]]*BROWSERLESS_' "$config_file" || true)

# ==================== CLI PARSING ====================
url=""
save_json=false
while [[ $# -gt 0 ]]; do
    case "$1" in
        -t|--token)    shift; BROWSERLESS_TOKEN="${1:?}";;
        -o|--origin)   shift; BROWSERLESS_ORIGIN="${1:?}";;
        -j|--json)     save_json=true;;
        -h|--help)
            cat <<EOF
Usage: $(basename "$0") [OPTIONS] <URL>

Options:
  -t, --token TOKEN     Browserless token
  -o, --origin ORIGIN   Browserless base URL (default: http://localhost:3000)
  -j, --json            Also save pretty-printed JSON-LD (+ sourceUrl) as <title>.json
  -h, --help            Show this help
  -v, --verbose         Enable debug output
EOF
            exit 0
            ;;
        -v|--verbose)  set -x;;
        -*)            echo "❌ Unknown option $1" >&2; exit 1;;
        *)             [[ -n "$url" ]] && { echo "❌ Only one URL allowed" >&2; exit 1; }; url="$1";;
    esac
    shift
done

[[ -z "$url" ]] && { echo "❌ URL is required" >&2; exit 1; }

# ==================== STRICT URL VALIDATION ====================
if [[ "$url" != https://tuckercarlson.com/* ]]; then
    echo "❌ ERROR: Only https://tuckercarlson.com/ URLs are allowed" >&2
    echo "   Received: $url" >&2
    exit 1
fi

# ==================== BROWSERLESS CALL ====================
scrape_url="${BROWSERLESS_ORIGIN}/chromium/scrape?stealth=true"
[[ -n "$BROWSERLESS_TOKEN" ]] && scrape_url+="&token=${BROWSERLESS_TOKEN}"

payload=$(jq -n --arg url "$url" '{url: $url, elements: [{selector: "script[type=\"application/ld+json\"]:first-of-type"}]}')

echo "🔄 Fetching metadata from browserless..."

tmp=$(mktemp)
http_code=$(curl -s -o "$tmp" -w "%{http_code}" -X POST "$scrape_url" \
    -H "Content-Type: application/json" -d "$payload" --max-time 30) || { rm -f "$tmp"; exit 1; }

response=$(<"$tmp")
rm -f "$tmp"

[[ "$http_code" != 200 ]] && { echo "❌ Browserless error $http_code:"; <<<"$response" jq . 2>/dev/null || echo "$response"; exit 1; }

if [[ -z "$(<<<"$response" jq -e '.data[0].results[0].text // empty' 2>/dev/null)" ]]; then
    echo "❌ No JSON-LD found"
    exit 1
fi

# ==================== EXTRACT & CLEAN ====================
video_json=$(<<<"$response" jq -r '.data[0].results[0].text')
content_url=$(<<<"$video_json" jq -r '.contentUrl // empty')
title=$(<<<"$video_json" jq -r '.name // empty')

[[ -z "$content_url" || -z "$title" ]] && { echo "❌ Missing video URL or title"; exit 1; }

clean_title() {
    local t="$1"
    t="${t//&amp;/\&}"
    t="${t//&quot;/\"}"
    t="${t//&apos;/\'}"
    t="${t//’/\'}"
    t="${t//‘/\'}"
    t="${t//“/\"}"
    t="${t//”/\"}"
    t="${t#"${t%%[![:space:]]*}" }"
    t="${t%"${t##*[![:space:]]}" }"
    printf '%s' "$t"
}

cleaned_title=$(clean_title "$title")

# ==================== SAVE JSON (if requested) ====================
if [[ "$save_json" == true ]]; then
    <<<"$video_json" jq --arg sourceUrl "$url" '
        . + {sourceUrl: $sourceUrl}
    ' > "${cleaned_title}.json"
    echo "📄 Saved JSON metadata: ${cleaned_title}.json"
fi

echo "📥 Downloading: $cleaned_title"
echo "   Stream: $content_url"

# ==================== DOWNLOAD ====================
yt-dlp \
    --ignore-config \
    --no-warnings \
    --restrict-filenames \
    --merge-output-format mp4 \
    --no-mtime \
    "$content_url" \
    -o "${cleaned_title}.%(ext)s"

echo "✅ Done: ${cleaned_title}.mp4"