# Derwent Valley Council — Development Applications being advertised
#
# Source: https://www.derwentvalley.tas.gov.au/home/latest-news?f.News+category...=Public+Notice
#
# The site is Cloudflare-protected — requires homepage warmup with browser-like
# headers (same technique as burnie.rb / kingisland.rb).
#
# The news listing links go through lgasa-search.lga.sa.gov.au → squiz.cloud.
# Rather than following that full redirect chain, we extract the `index_url`
# parameter from each lgasa href, which points at lgasa-web.squiz.cloud/?a=ID.
# A single (non-following) GET to that URL returns a Location header with the
# real derwentvalley.tas.gov.au detail page URL.
#
# Detail page structure:
#   <table><tbody>
#     <tr><td>APP No.</td><td>SITE</td><td>PROPOSAL</td></tr>
#     <tr><td>DA 2026/023</td><td>160 Wyre Forest Road, Molesworth</td>
#         <td>Dwelling and outbuilding</td></tr>
#   </tbody></table>
#   <p>...received no later than 5.00pm on 15 April 2026...</p>
#   <div class="content-container"><a href="...DA-2026-023.pdf">plans</a></div>

require "date"
require "nokogiri"
require "net/http"
require "uri"

require_relative "../lib/db"
require_relative "../lib/enrich"
require_relative "../lib/log"
require_relative "../lib/util"

TABLE    = ENV.fetch("TABLE_NAME")  # run_all.sh sets from filename: da_derwentvalley
BASE_URL = "https://www.derwentvalley.tas.gov.au"
NEWS_URL = "#{BASE_URL}/home/latest-news?f.News+category%7CnewsCategory=Public+Notice"

DB.ensure_table!(TABLE)

# ----- Browser-like headers (WAF/Cloudflare warmup) -----
BASE_HEADERS = {
    "User-Agent"                => "Mozilla/5.0 (Windows NT 10.0; Win64; x64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/127.0.0.0 Safari/537.36",
    "Accept"                    => "text/html,application/xhtml+xml,application/xml;q=0.9,*/*;q=0.8",
    "Accept-Language"           => "en-AU,en;q=0.9",
    "Accept-Encoding"           => "identity",
    "Upgrade-Insecure-Requests" => "1",
    "Sec-Fetch-Dest"            => "document",
    "Sec-Fetch-Mode"            => "navigate",
    "Sec-Fetch-Site"            => "none",
    "Sec-Fetch-User"            => "?1",
    "sec-ch-ua"                 => '"Chromium";v="127", "Not)A;Brand";v="99", "Google Chrome";v="127"',
    "sec-ch-ua-mobile"          => "?0",
    "sec-ch-ua-platform"        => '"Windows"',
    "Connection"                => "close",
}.freeze

class CookieJar
    def initialize; @h = {}; end

    def for(host)
        @h[host] || ""
    end

    def merge_from(resp, host)
        cookies = resp.get_fields("Set-Cookie") || []
        return if cookies.empty?
        existing = parse_header(@h[host])
        cookies.each do |sc|
            kv = sc.split(";", 2).first
            k, v = kv.split("=", 2)
            existing[k.to_s.strip] = v.to_s unless k.to_s.strip.empty?
        end
        @h[host] = existing.map { |k, v| "#{k}=#{v}" }.join("; ")
    end

    private

    def parse_header(s)
        s.to_s.split(";").map(&:strip).filter_map { |kv|
            k, v = kv.split("=", 2)
            [k, v] unless k.to_s.empty?
        }.to_h
    end
end

# GET url; follow redirects; return [final_url_string, body, http_code]
def http_get(url, jar:, referer: nil, fetch_site: "none", follow: true)
    uri  = URI(url)
    hdrs = BASE_HEADERS.merge("Sec-Fetch-Site" => fetch_site)
    hdrs["Referer"] = referer if referer
    cookie = jar.for(uri.host)
    hdrs["Cookie"] = cookie unless cookie.empty?

    limit = 8
    code  = 0
    body  = ""

    while limit > 0
        limit -= 1
        redirect_to = nil

        req = Net::HTTP::Get.new(uri, hdrs)
        Net::HTTP.start(uri.host, uri.port, use_ssl: uri.scheme == "https",
                        read_timeout: 30, open_timeout: 15) do |http|
            resp = http.request(req)
            jar.merge_from(resp, uri.host)
            code = resp.code.to_i

            if follow && [301, 302, 303, 307, 308].include?(code) && resp["location"]
                redirect_to = URI.join(uri, resp["location"])
            else
                body = resp.body.to_s
            end
        end

        if redirect_to
            uri = redirect_to
            # Update Referer and Sec-Fetch-Site for subsequent hops
            hdrs["Referer"]        = uri.to_s
            hdrs["Sec-Fetch-Site"] = "same-origin"
            cookie = jar.for(uri.host)
            hdrs["Cookie"] = cookie.empty? ? nil : cookie
            next
        end
        break
    end

    [uri.to_s, body, code]
rescue StandardError => e
    Log.warn "derwentvalley", "HTTP error for #{url}: #{e.class} #{e.message}"
    [url, "", 0]
end

# Resolve lgasa redirect href -> real derwentvalley.tas.gov.au URL.
# Extracts index_url from the lgasa query string, then makes a non-following
# GET to squiz.cloud/?a=ID and reads the Location header.
def resolve_detail_url(lgasa_href)
    query = URI.decode_www_form(URI(lgasa_href).query.to_s).to_h
    index_url = query["index_url"]
    return nil if index_url.to_s.empty?

    uri = URI(index_url)
    req = Net::HTTP::Get.new(uri, "User-Agent" => BASE_HEADERS["User-Agent"])
    Net::HTTP.start(uri.host, uri.port, use_ssl: uri.scheme == "https",
                    open_timeout: 10, read_timeout: 10) do |http|
        resp = http.request(req)
        loc  = resp["location"].to_s
        return loc unless loc.empty?
    end
    nil
rescue StandardError => e
    Log.warn "derwentvalley", "Could not resolve squiz redirect #{index_url}: #{e.class} #{e.message}"
    nil
end

# Parse a detail page for DA data
def parse_detail(html, page_url)
    doc = Nokogiri::HTML(html)

    # Table: header row "APP No. | SITE | PROPOSAL", then data rows
    data_rows = []
    doc.css("table").each do |tbl|
        tbl.css("tr").each do |tr|
            cells = tr.css("td").map { |td| td.text.gsub(/\u00a0|\s+/, " ").strip }
            next if cells.empty?
            next if cells.join =~ /\AAPP\s*No\.?/i   # skip header row
            next unless cells[0] =~ /\bDA\s*\d{4}\/\d+/i
            data_rows << cells
        end
    end

    return [] if data_rows.empty?

    body_text = doc.css(".content-container, main").first&.text.to_s
                   .gsub(/\u00a0/, " ").gsub(/\s+/, " ")

    # Closing date from "received no later than ... DATE"
    on_notice_to_raw = ""
    on_notice_to     = nil
    if (m = body_text.match(/no\s+later\s+than\b.{0,60}?(\d{1,2}\s+[A-Za-z]{3,}\s+\d{4})/i))
        on_notice_to_raw = m[1].strip
        on_notice_to     = Util.parse_aus_date(on_notice_to_raw)
    end

    # Commencing / start date → date_received
    date_received_raw = ""
    date_received     = nil
    if (m = body_text.match(/commencing\s+on\s+(\d{1,2}\s+[A-Za-z]{3,}\s+\d{4})/i)) ||
       (m = body_text.match(/Start\s+Date\s+(\d{1,2}\/\d{1,2}\/\d{2,4})/i))
        date_received_raw = m[1].strip
        date_received     = Util.parse_aus_date(date_received_raw)
    end

    # PDF link within the content area only
    doc_url = nil
    doc.css(".content-container a[href]").each do |a|
        href = a["href"].to_s
        if href =~ /\.pdf/i && href.include?("derwentvalley")
            doc_url = href
            break
        end
    end

    data_rows.map do |cells|
        ref         = cells[0].to_s.gsub(/\A\s+/, "").strip
        address     = cells[1].to_s.strip
        description = cells[2].to_s.strip
        description = "Development Application" if description.empty?

        next if ref.empty? || address.empty?

        {
            council_reference: ref,
            address:           address,
            description:       description,
            date_received:     date_received,
            date_received_raw: date_received_raw,
            on_notice_to:      on_notice_to,
            on_notice_to_raw:  on_notice_to_raw,
            document_url:      doc_url
        }
    end.compact
end

# ----- Warmup then fetch news listing -----
jar = CookieJar.new

Log.info "derwentvalley", "Warming up via homepage..."
_url0, _body0, code0 = http_get("#{BASE_URL}/", jar: jar)
Log.info "derwentvalley", "Homepage: #{code0}"

sleep(0.5)

Log.info "derwentvalley", "Fetching news listing..."
_url1, html1, code1 = http_get(NEWS_URL, jar: jar, referer: "#{BASE_URL}/", fetch_site: "same-origin")
Log.info "derwentvalley", "News listing: #{code1} (#{html1.bytesize} bytes)"

if code1 != 200 || html1.bytesize < 5_000
    Log.warn "derwentvalley", "Could not fetch news listing (status #{code1}). " \
        "DAs for this council are also available via planbuild.rb (council code DER)."
    puts "Done #{TABLE}. Saved 0 item(s)."
    exit 0
end

if html1.include?("Just a moment") || html1.include?("Enable JavaScript and cookies")
    Log.warn "derwentvalley", "Cloudflare challenge page returned. " \
        "DAs for this council are also available via planbuild.rb (council code DER)."
    puts "Done #{TABLE}. Saved 0 item(s)."
    exit 0
end

# ----- Extract detail page URLs from news listing -----
list_doc     = Nokogiri::HTML(html1)
detail_urls  = []

list_doc.css("li.news-listing__item a[href]").each do |a|
    href = a["href"].to_s
    next unless href.include?("lgasa-search")

    detail_url = resolve_detail_url(href)
    detail_urls << detail_url if detail_url && !detail_url.empty?
end

detail_urls.uniq!
Log.info "derwentvalley", "Found #{detail_urls.length} detail page(s)"

saved = 0

detail_urls.each do |detail_url|
    Log.info "derwentvalley", "Fetching #{detail_url}"
    sleep(0.4)

    _final_url, html2, code2 = http_get(
        detail_url, jar: jar,
        referer:    NEWS_URL,
        fetch_site: "same-origin"
    )

    if code2 != 200 || html2.bytesize < 5_000
        Log.warn "derwentvalley", "Detail page failed (#{code2}): #{detail_url}"
        next
    end

    records = parse_detail(html2, detail_url)
    if records.empty?
        Log.warn "derwentvalley", "No DA records parsed from #{detail_url}"
        next
    end

    records.each do |r|
        begin
            DB.upsert(TABLE, {
                council_reference: r[:council_reference],
                address:           r[:address][0, 255],
                description:       r[:description],
                date_received:     r[:date_received],
                date_received_raw: r[:date_received_raw],
                on_notice_to:      r[:on_notice_to],
                on_notice_to_raw:  r[:on_notice_to_raw],
                document_url:      r[:document_url],
                applicant:         "",
                owner:             ""
            })

            enrich_after_upsert!(
                table:             TABLE,
                council_reference: r[:council_reference],
                address:           r[:address]
            )

            Log.info "derwentvalley", "Upserted #{r[:council_reference]} -> #{r[:address]}"
            saved += 1
        rescue StandardError => e
            Log.warn "derwentvalley", "DB error for #{r[:council_reference]}: #{e.class} #{e.message}"
        end
    end
end

puts "Done #{TABLE}. Saved #{saved} item(s)."