Add function to collapse a Corpus

[neelsmith]

E.g., collapse token-level exemplar to a canonical-level exemplar.

Function should take parameter for level to collapse to, and String value for new exemplar identifier.

[neelsmith]

Here's an example of how to do two-tier sorting of nodes.Functions are poorly named for this general case, but start wtih "publishable" and you reduce the token-level analytical edition to a read/analyzed canonical-level Corpus. NB: it's lacking the required feature of identifying the new exemplar by a unique exemplar identifier

def passagesFromTokens(readingPairs: Vector[(CtsUrn, String)]) = {
  val triples = readingPairs.zipWithIndex.map( v  => TextTriple(v._1._1, v._1._2, v._2))
  val trVect = triples.groupBy(_.urn).toSeq.toVector
  trVect.map{ case (k,v) =>(k, v.sortBy(_.seq).map(_.reading).mkString(" ") ) }  
}

def idxForUrn(u: CtsUrn, urnSeq: Vector[(CtsUrn, Int)]) = {
  urnSeq.filter(_._1 == u)(0)._2
}

def publishable(scholionGroup: String, publType: String) : Vector[CitableNode] = {
  val tkns =tokensForDocument(scholionGroup, publType)
  val passages = passagesFromTokens(tkns)
  // For final sort:
  val urnSeq = tkns.map(_._1).distinct.zipWithIndex

  val sortedFinal = passages.sortBy{ case (k,v) => idxForUrn(k, urnSeq) }
  sortedFinal.map { case (k,v) => CitableNode(k,v)} 
}

[neelsmith]

There may be easier algorithms. Look at the way hmt-mom collapses a set of tokens.