Scrap timezone offsets

scrap-timezones.sc

import $ivy.`net.ruippeixotog::scala-scraper:2.2.0`
import net.ruippeixotog.scalascraper.browser.JsoupBrowser
import net.ruippeixotog.scalascraper.dsl.DSL._
import net.ruippeixotog.scalascraper.dsl.DSL.Extract._
import net.ruippeixotog.scalascraper.dsl.DSL.Parse._
import net.ruippeixotog.scalascraper.model._

def extractMinutesOffset(s: String): Long = {
    val offset = raw"([+−]{1})(\d{2}):(\d{2})".r

    s match {
        case offset(sign, hh, mm) =>
            (if (sign == "+") 1 else - 1) * (hh.toInt * 60 + mm.toInt)
    }
}

val browser = JsoupBrowser()
val doc = browser.get("https://en.wikipedia.org/wiki/List_of_tz_database_time_zones")

val content = doc >> elements("#mw-content-text")
val tables = content.extract(".wikitable").toArray
val table = tables(0)
val rows = table.extract("td").toArray.grouped(8).toArray

rows.foreach { row =>
    val name       = row(2).text
    val deprecated = row(4).text == "Deprecated"
    val stdOffset  = row(5).text.trim
    val dstOffset  = row(6).text.trim

    val offset = extractMinutesOffset(dstOffset) - extractMinutesOffset(stdOffset)
    if (offset != 60L && offset != 0 && !deprecated) {
        println(s"$name has an offset of $offset")
    }
}