[incubator-nlpcraft] branch NLPCRAFT-443 updated: WIP.

[sergeykamov]

This is an automated email from the ASF dual-hosted git repository.

sergeykamov pushed a commit to branch NLPCRAFT-443
in repository https://gitbox.apache.org/repos/asf/incubator-nlpcraft.git


The following commit(s) were added to refs/heads/NLPCRAFT-443 by this push:
     new f7c3611  WIP.
f7c3611 is described below

commit f7c361193e8f599eb550c4842bfa53fe18248b41
Author: Sergey Kamov <skhdlem...@gmail.com>
AuthorDate: Sat Sep 18 12:16:45 2021 +0300

    WIP.
---
 .../mgrs/nlp/enrichers/model/NCModelEnricher.scala | 31 +++++++++++-----------
 1 file changed, 15 insertions(+), 16 deletions(-)

diff --git 
a/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/nlp/enrichers/model/NCModelEnricher.scala
 
b/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/nlp/enrichers/model/NCModelEnricher.scala
index 058d713..6c0f6f2 100644
--- 
a/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/nlp/enrichers/model/NCModelEnricher.scala
+++ 
b/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/nlp/enrichers/model/NCModelEnricher.scala
@@ -681,23 +681,22 @@ object NCModelEnricher extends NCProbeEnricher {
         usrNotes.
             filter(n => !links.contains(NoteLink(n.noteType, 
n.tokenIndexes.sorted))).
             filter(n => !parts.contains(NCTokenPartKey(n, ns))).
-            foreach(n => {
-                val hasBetter =
-                    usrNotes.exists(candidate =>
-                        candidate != n &&
-                        candidate.noteType == n.noteType &&
-                        candidate.dataOpt("parts") == n.dataOpt("parts") &&
-                        
candidate.wordIndexes.toSet.subsetOf(n.wordIndexes.toSet) &&
-                        n.wordIndexes.filter(n => 
!candidate.wordIndexes.contains(n)).
-                            forall(wordIdx => ns.tokens.exists(t => 
t.wordIndexes.contains(wordIdx) && t.isStopWord)))
-
-                if (hasBetter) {
-                    ns.removeNote(n)
-
-                    // TODO: trace.
-                    logger.info(s"Element removed: ${n}")
+            foreach(n =>
+                usrNotes.find(candidate =>
+                    candidate != n &&
+                    candidate.noteType == n.noteType &&
+                    candidate.dataOpt("parts") == n.dataOpt("parts") &&
+                    candidate.wordIndexes.toSet.subsetOf(n.wordIndexes.toSet) 
&&
+                    n.wordIndexes.filter(n => 
!candidate.wordIndexes.contains(n)).
+                        forall(wordIdx => ns.tokens.exists(t => 
t.wordIndexes.contains(wordIdx) && t.isStopWord))
+                ) match {
+                    case Some(better) =>
+                        ns.removeNote(n)
+                        // TODO: trace.
+                        logger.info(s"Element removed: $n, better: $better")
+                    case None => // No-op.
                 }
-            })
+            )
     }
 
     // TODO: simplify, add tests, check model properties (sparse etc) for 
optimization.