Snippet to test nxmlreader on a subset of PubMed OA

nxmlReaderTests.scala

import java.io.File

import scala.util.Random
import ai.lum.common.FileUtils._
import ai.lum.common.RandomUtils._
import ai.lum.nxmlreader.NxmlReader

val rand = new Random(42)
val nxmlDir = new File("/net/kate/storage/data/nlp/corpora/bmgf/OA-100K-sample/data/nlp/corpora/pmc_openaccess/pmc_aug2016/")
val nxmlFiles = nxmlDir.listFilesByWildcard("*.nxml", recursive = true)
val nxmlFilesSample = rand.sample(nxmlFiles, 1000).toVector
val reader = new NxmlReader()

nxmlFilesSample.foreach{ f =>
  val contents = f.readString()
  try {
    reader.parse(contents)  
  } catch {
    case e: Exception =>
      println(s"problem with ${f.getCanonicalPath}")
      println(e.getMessage)
  }
}