This is an automated email from the ASF dual-hosted git repository.
aradzinski pushed a commit to branch scala-2.13
in repository https://gitbox.apache.org/repos/asf/incubator-nlpcraft.git
The following commit(s) were added to refs/heads/scala-2.13 by this push:
new dc4936f WIP.
dc4936f is described below
commit dc4936f1b0b75efcbcd02d6f1f704e2d9db52b45
Author: Aaron Radzinzski <[email protected]>
AuthorDate: Thu May 20 11:11:45 2021 -0700
WIP.
---
.../apache/nlpcraft/common/debug/NCLogHolder.scala | 2 +-
.../nlpcraft/common/nlp/NCNlpSentenceNote.scala | 10 +++---
.../org/apache/nlpcraft/common/util/NCUtils.scala | 2 +-
.../apache/nlpcraft/model/impl/NCTokenLogger.scala | 1 -
.../nlpcraft/probe/mgrs/NCProbeVariants.scala | 38 +++++++++++-----------
.../nlp/core/opennlp/NCOpenNlpNerEnricher.scala | 26 +++++++--------
.../nlp/enrichers/NCServerEnrichmentManager.scala | 2 +-
7 files changed, 39 insertions(+), 42 deletions(-)
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/debug/NCLogHolder.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/debug/NCLogHolder.scala
index 6fb5f55..76f9b79 100644
--- a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/debug/NCLogHolder.scala
+++ b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/debug/NCLogHolder.scala
@@ -23,7 +23,7 @@ import java.util
import com.google.gson.Gson
import org.apache.nlpcraft.model._
-import scala.jdk.CollectionConverters.{CollectionHasAsScala,
IterableHasAsJava, MapHasAsJava, SeqHasAsJava}
+import scala.jdk.CollectionConverters._
//
// NOTE: these classes are specifically designed for JSON marshalling.
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/nlp/NCNlpSentenceNote.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/nlp/NCNlpSentenceNote.scala
index a29d3ed..7d6b860 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/nlp/NCNlpSentenceNote.scala
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/nlp/NCNlpSentenceNote.scala
@@ -216,12 +216,12 @@ object NCNlpSentenceNote {
* @param params Parameters.
*/
def apply(
- indexes: Seq[Int],
- wordIndexesOpt: Option[Seq[Int]],
+ indexes: collection.Seq[Int],
+ wordIndexesOpt: Option[collection.Seq[Int]],
typ: String,
params: (String, Any)*
): NCNlpSentenceNote = {
- def calc(seq: Seq[Int]): (Int, Int, Int, java.util.List[Int], Int) =
+ def calc(seq: collection.Seq[Int]): (Int, Int, Int,
java.util.List[Int], Int) =
(U.calcSparsity(seq), seq.min, seq.max, seq.asJava, seq.length)
val (sparsity, tokMinIndex, tokMaxIndex, tokWordIndexes, len) =
calc(wordIndexesOpt.getOrElse(indexes))
@@ -249,7 +249,7 @@ object NCNlpSentenceNote {
* @param typ Type of the node.
* @param params Parameters.
*/
- def apply(indexes: Seq[Int], typ: String, params: (String, Any)*):
NCNlpSentenceNote =
+ def apply(indexes: collection.Seq[Int], typ: String, params: (String,
Any)*): NCNlpSentenceNote =
apply(indexes, None, typ, params: _*)
/**
@@ -260,6 +260,6 @@ object NCNlpSentenceNote {
* @param typ Type of the node.
* @param params Parameters.
*/
- def apply(indexes: Seq[Int], wordIndexes: Seq[Int], typ: String, params:
(String, Any)*): NCNlpSentenceNote =
+ def apply(indexes: collection.Seq[Int], wordIndexes: collection.Seq[Int],
typ: String, params: (String, Any)*): NCNlpSentenceNote =
apply(indexes, Some(wordIndexes), typ, params: _*)
}
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/util/NCUtils.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/util/NCUtils.scala
index 6b3b25c..588af4f 100644
--- a/nlpcraft/src/main/scala/org/apache/nlpcraft/common/util/NCUtils.scala
+++ b/nlpcraft/src/main/scala/org/apache/nlpcraft/common/util/NCUtils.scala
@@ -2095,7 +2095,7 @@ object NCUtils extends LazyLogging {
* @param idx Sequence of indexes.
* @return
*/
- def calcSparsity(idx: Seq[Int]): Int =
+ def calcSparsity(idx: collection.Seq[Int]): Int =
idx.zipWithIndex.tail.map { case (v, i) => Math.abs(v - idx(i - 1))
}.sum - idx.length + 1
/**
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/impl/NCTokenLogger.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/impl/NCTokenLogger.scala
index 72f8a35..c3a83a6 100644
--- a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/impl/NCTokenLogger.scala
+++ b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/impl/NCTokenLogger.scala
@@ -27,7 +27,6 @@ import org.apache.nlpcraft.model.NCToken
import org.apache.nlpcraft.model.impl.NCTokenPimp._
import org.apache.nlpcraft.common.ansi.NCAnsi._
-import scala.collection.mutable.ArrayBuffer
import scala.jdk.CollectionConverters._
/**
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/NCProbeVariants.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/NCProbeVariants.scala
index cd349ba..dcba4da 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/NCProbeVariants.scala
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/probe/mgrs/NCProbeVariants.scala
@@ -17,8 +17,8 @@
package org.apache.nlpcraft.probe.mgrs
+import org.apache.nlpcraft.common.nlp.{NCNlpSentence, NCNlpSentenceNote,
NCNlpSentenceToken}
import org.apache.nlpcraft.common.nlp.pos.NCPennTreebank
-import org.apache.nlpcraft.common.nlp.{NCNlpSentence => NlpSentence,
NCNlpSentenceNote => NlpNote, NCNlpSentenceToken => NlpToken}
import org.apache.nlpcraft.common.{NCE, TOK_META_ALIASES_KEY}
import org.apache.nlpcraft.model.impl.{NCTokenImpl, NCTokenLogger,
NCVariantImpl}
import org.apache.nlpcraft.model.{NCToken, NCVariant}
@@ -26,8 +26,8 @@ import org.apache.nlpcraft.model.{NCToken, NCVariant}
import java.io.{Serializable => JSerializable}
import java.util
import java.util.Collections.singletonList
-import scala.collection.JavaConverters._
import scala.collection.{Seq, mutable}
+import scala.jdk.CollectionConverters._
/**
* Sentence to variants converter.
@@ -82,10 +82,10 @@ object NCProbeVariants {
*/
private def findDeletedToken(
key: Key,
- delNotes: Map[NlpNote, Seq[NlpToken]],
+ delNotes: Map[NCNlpSentenceNote, Seq[NCNlpSentenceToken]],
noteTypePred: String => Boolean
- ): Option[NlpToken] =
- delNotes.toStream.
+ ): Option[NCNlpSentenceNote] =
+ delNotes.to(LazyList).
flatMap { case (delNote, delNoteToks) =>
if (noteTypePred(delNote.noteType)) {
val toks =
@@ -98,12 +98,12 @@ object NCProbeVariants {
toks.size match {
case 0 => None
case _ =>
- val artTok = NlpToken(IDX)
+ val artTok = NCNlpSentenceToken(IDX)
artTok.add(mkNote(toks))
if (key.id != "nlpcraft:nlp") {
- val ps = mkNlpNoteParams()
+ var ps = mkNlpNoteParams()
delNote.noteType match {
case "nlpcraft:relation" |
"nlpcraft:limit" => ps += "indexes" -> IDXS
@@ -125,17 +125,17 @@ object NCProbeVariants {
*
* @return
*/
- private def mkNlpNoteParams(): mutable.ArrayBuffer[(String,
JSerializable)] =
- mutable.ArrayBuffer.empty[(String, JSerializable)] ++
Seq("tokMinIndex" -> IDX, "tokMaxIndex" -> IDX)
+ private def mkNlpNoteParams(): Seq[(String, JSerializable)] =
+ Seq("tokMinIndex" -> IDX, "tokMaxIndex" -> IDX)
/**
*
* @param srcToks
* @return
*/
- private def mkNote(srcToks: Seq[NlpToken]): NlpNote = {
+ private def mkNote(srcToks: Seq[NCNlpSentenceToken]): NCNlpSentenceNote = {
// Note, it adds stop-words too.
- def mkValue(get: NlpToken => String): String = {
+ def mkValue(get: NCNlpSentenceToken => String): String = {
val buf = mutable.Buffer.empty[String]
val n = srcToks.size - 1
@@ -150,10 +150,10 @@ object NCProbeVariants {
buf.mkString
}
- def all(is: NlpToken => Boolean): Boolean = srcToks.forall(is)
- def exists(is: NlpToken => Boolean): Boolean = srcToks.exists(is)
+ def all(is: NCNlpSentenceToken => Boolean): Boolean =
srcToks.forall(is)
+ def exists(is: NCNlpSentenceToken => Boolean): Boolean =
srcToks.exists(is)
- val origText = mkValue((t: NlpToken) => t.origText)
+ val origText = mkValue((t: NCNlpSentenceToken) => t.origText)
val params = Seq(
"index" -> IDX,
@@ -175,7 +175,7 @@ object NCProbeVariants {
"swear" -> exists(_.isSwearWord)
)
- NlpNote(Seq(IDX.intValue()),
srcToks.flatMap(_.wordIndexes).distinct.sorted, "nlpcraft:nlp", params: _*)
+ NCNlpSentenceNote(Seq(IDX.intValue()),
srcToks.flatMap(_.wordIndexes).distinct.sorted, "nlpcraft:nlp", params: _*)
}
/**
@@ -186,12 +186,12 @@ object NCProbeVariants {
* @param nlpSens Sentences.
* @param lastPhase Flag.
*/
- def convert(srvReqId: String, mdl: NCProbeModel, nlpSens:
Seq[NlpSentence], lastPhase: Boolean = false): Seq[NCVariant] = {
+ def convert(srvReqId: String, mdl: NCProbeModel, nlpSens:
Seq[NCNlpSentence], lastPhase: Boolean = false): Seq[NCVariant] = {
var vars =
nlpSens.flatMap(nlpSen => {
var ok = true
- def mkToken(nlpTok: NlpToken): NCTokenImpl = {
+ def mkToken(nlpTok: NCNlpSentenceToken): NCTokenImpl = {
val ncTok = NCTokenImpl(mdl, srvReqId, nlpTok)
nlpSen.addNlpToken(nlpTok)
@@ -202,7 +202,7 @@ object NCProbeVariants {
val toks = nlpSen.map(mkToken)
val keys2Toks = toks.map(t => Key(t) -> t).toMap
- def process(tok: NCTokenImpl, tokNlp: NlpToken): Unit = {
+ def process(tok: NCTokenImpl, tokNlp: NCNlpSentenceToken):
Unit = {
val optList: Option[util.List[util.HashMap[String,
JSerializable]]] =
tokNlp.find(_.isUser) match {
case Some(u) => u.dataOpt("parts")
@@ -232,7 +232,7 @@ object NCProbeVariants {
case None =>
nlpSen.getInitialNlpNote(key.from, key.to) match {
case Some(nlpNote) =>
- val artTok =
NlpToken(IDX)
+ val artTok =
NCNlpSentenceToken(IDX)
artTok.add(nlpNote.clone(mkNlpNoteParams(): _*))
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/core/opennlp/NCOpenNlpNerEnricher.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/core/opennlp/NCOpenNlpNerEnricher.scala
index 1e95e6d..586641b 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/core/opennlp/NCOpenNlpNerEnricher.scala
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/core/opennlp/NCOpenNlpNerEnricher.scala
@@ -124,20 +124,18 @@ object NCOpenNlpNerEnricher extends NCService with
NCNlpNerEnricher with NCIgnit
case class Holder(start: Int, end: Int, name: String, probability:
Double)
- val hs =
- this.
- synchronized {
- val res = nerFinders.
- filter { case (_, tokName) =>
ebiTokens.contains(tokName)}.
- flatMap {
- case (finder, name) =>
- finder.find(words).map(p =>
Holder(p.getStart, p.getEnd - 1, name, p.getProb))
- }
-
- nerFinders.keys.foreach(_.clearAdaptiveData())
-
- res
- }.toSeq
+ val hs = this. synchronized {
+ val res = nerFinders.
+ filter { case (_, tokName) =>
ebiTokens.contains(tokName)}.toSeq.
+ flatMap {
+ case (finder, name) =>
+ finder.find(words).map(p => Holder(p.getStart,
p.getEnd - 1, name, p.getProb))
+ }
+
+ nerFinders.keys.foreach(_.clearAdaptiveData())
+
+ res
+ }
hs.
filter(h => ebiTokens.contains(h.name)).
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/enrichers/NCServerEnrichmentManager.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/enrichers/NCServerEnrichmentManager.scala
index dee1e6b..9fc7697 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/enrichers/NCServerEnrichmentManager.scala
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/server/nlp/enrichers/NCServerEnrichmentManager.scala
@@ -205,7 +205,7 @@ object NCServerEnrichmentManager extends NCService with
NCIgniteInstance {
getOrElse(throw new NCE(s"Header not found for:
${hdr.noteType}"))._2
(x._1 * 100) + x._2.indexOf(hdr.noteName)
- })
+ }).toSeq
val tbl = NCAsciiTable(headers.map(_.header): _*)
