tas_councils/lib/scraper_helpers.rb

# lib/scraper_helpers.rb
# Shared top-level helpers required by individual DA scrapers.
#
# Usage — at the top of a scraper, after other requires:
#
#   require_relative "../lib/scraper_helpers"
#
# This file requires db and enrich so scrapers don't need separate
# require lines for those two libs.

require "uri"
require_relative "./db"
require_relative "./enrich"

# Resolve a possibly-relative href against a base URL.
# Returns the href string unchanged if URI.join raises.
def abs_url(base, href)
  return "" if href.to_s.strip.empty?
  URI.join(base, href).to_s
rescue URI::InvalidURIError
  href.to_s
end

# Return node.text.strip, or default when node is nil.
def text_or(node, default = "")
  node ? node.text.strip : default
end

# Upsert a DA row, run enrichment, and optionally UPDATE extra columns.
#
# table  - validated DA table name (e.g. "da_glamorgan")
# row    - hash passed to DB.upsert; must include :council_reference and :address
# extras - optional hash of { column_name => value } pairs to UPDATE after upsert
#          e.g. { document_url: "https://...", on_notice_to: Date.new(2025,6,1) }
#
# Prints "Upserted <ref> -> <address>" on success.
def upsert_and_enrich!(table:, row:, extras: {})
  DB.upsert(table, row)
  enrich_after_upsert!(
    table: table,
    council_reference: row[:council_reference],
    address: row[:address]
  )
  unless extras.empty?
    begin
      esc = DB.client.escape(table)
      set_clause = extras.keys.map { |k| "`#{k}` = ?" }.join(", ")
      vals = extras.values + [row[:council_reference], row[:address]]
      upd = DB.client.prepare(
        "UPDATE `#{esc}` SET #{set_clause} WHERE council_reference = ? AND address = ?"
      )
      upd.execute(*vals)
    rescue Mysql2::Error => e
      warn "[scraper_helpers] extras update skipped for #{row[:council_reference]}: #{e.message}"
    end
  end
  puts "Upserted #{row[:council_reference]} -> #{row[:address]}"
end