格式化markdown表格

NormaMdTable.scala

#!/bin/sh
exec scala "$0" "$@"
!#
/**
 * 目标：格式化md table 表格语法 http://www.tablesgenerator.com/markdown_tables
 * 注意点:文件的编码 涉及到字符串长度的计算
 */
if (args.length < 1 || args.length > 2) {
  println(
    """Usage: scala NormaMdTable.scala inputMdFile [outputMdFile]
      |       * inputMdFile: the path of the inputFile, eg. a.md
      |       * inputMdFile: the path of the outputFile, if not specified will add a ".md" postfix
    """.stripMargin)
} else {
  process(args(0), if (args.length == 2) args(1) else args(0) + ".md")
}

def process(in: String, out: String) = {
  val source = recurProcess(
    scala.io.Source.fromFile(args(0), "utf-8").getLines(), List[String]()).reverse

  import java.io._
  val pw = new PrintWriter(new OutputStreamWriter(new FileOutputStream(out), "UTF-8"))
  pw.write(source.mkString("\n"))
  pw.close()

  println(s"generate new file in $out")
}

def recurProcess(source: Iterator[String], content: List[String]): List[String] = {
  if (!source.hasNext) content
  else {
    val line = source.next()
    if (line.startsWith("|")) {
      val (tableRest, rest) = source.span(_.startsWith("|"))
      recurProcess(rest, beautifyMdTable(line +: tableRest.toSeq) :: content)
    } else {
      recurProcess(source, line :: content)
    }
  }
}

def beautifyMdTable(table: Seq[String]): String = {
  def extractContent(table: Seq[String]): Seq[Array[String]] =
    (table.head +: table.drop(2)).foldRight(Seq[Array[String]]()) { (line, contents) =>
      line.substring(1).split("\\|").map(_.trim) +: contents
    }
  def calcDisplayLength(x : String) = {
    // http://stackoverflow.com/questions/28761385/single-chinese-character-determined-as-length-2-in-java-scala-string useless?
//    http://zhidao.baidu.com/question/271624295.html
    //x.length + x.count(c => Character.isIdeographic(c) || (c >= 12288 && c <= 12330) || c > 65248)
    x.length * 2 - x.count(_.toInt < 1000)
  }

  val contents = extractContent(table)
  // 要考虑中文字符
  val beforetrans = contents.map(_.length)
  if(!beforetrans.forall(_ == beforetrans.head))
    println(s"processing $beforetrans with ${contents.map(_.toList)}")
  val maxlengths: Array[Int] = contents.map(_.map(calcDisplayLength)).transpose.map(_.max + 2).toArray


  def getTableMd(line: Array[String]): String = {
    val beautified = for (i <- line.indices)
      yield " " + line(i) + " " * (maxlengths(i) - calcDisplayLength(line(i)) - 1)
    beautified.mkString("|", "|", "|")
  }

  def second = maxlengths.map("-" * _).mkString("|", "|", "|")

  (getTableMd(contents.head) +: second +: contents.tail.map(getTableMd)).mkString("\n")
}