EksiSozlukSiteParser.scala

package parsers

import java.net.URL
import java.time.LocalDate

import core.Rezalet
import org.jsoup.Jsoup
import org.jsoup.nodes.{Element, Document}

import scala.collection.JavaConversions._

trait EksiSozlukSiteParser extends SiteParser {

  private val months = Seq("ocak", "şubat", "mart", "nisan", "mayıs", "haziran", "temmuz", "ağustos", "eylül", "ekim", "kasım", "aralık")

  override def source: String = "eksisozluk"

  def searchPageUrl(page: Integer) = new URL(s"https://eksisozluk.com/basliklar/ara?searchForm.Keywords=rezaleti&searchForm.SortOrder=Date&p=$page")

  def parseTitle(title: String): Option[TitleInfo]  = {
    val pattern = "(\\d{1,2})\\s(\\w+)\\s(\\d{4})\\s(.*)rezaleti".r
    title match {
      case pattern(day, month, year, company) if months.contains(month) =>
        Option((company.trim, LocalDate.of(year.toInt, months.indexOf(month) + 1, day.toInt)))
      case _ => None
    }
  }

  override def parsePage(page: Int)(processor: RezaletProcessor): Boolean = {
    val document: Document = Jsoup.parse(searchPageUrl(page), 1000)
    val links = document.select("section#content-body ul.topic-list li a")

    def getEntryCount(link:Element): Option[Int] = link.childNodeSize() match {
      case c: Int if c > 1 => Some(link.child(0).ownText().toInt)
      case _ => None
    }

    for {
      link <- links
      titleInfo <- parseTitle(link.ownText())
    } processor(Rezalet(company = titleInfo._1, date = titleInfo._2, source = source, link = link.attr("href"), popularity = getEntryCount(link).getOrElse(1), isActive = true))
    links.nonEmpty
  }

}