This is an automated email from the ASF dual-hosted git repository.
aradzinski pushed a commit to branch master
in repository https://gitbox.apache.org/repos/asf/incubator-nlpcraft.git
The following commit(s) were added to refs/heads/master by this push:
new 835d175 Fix for NLPCRAFT-294.
835d175 is described below
commit 835d17519438850d6281db26f86b1446ed1a242a
Author: Aaron Radzinzski <[email protected]>
AuthorDate: Sat Apr 17 02:18:47 2021 +0300
Fix for NLPCRAFT-294.
---
.../scala/org/apache/nlpcraft/common/nlp/NCNlpSentenceNote.scala | 5 +----
nlpcraft/src/main/scala/org/apache/nlpcraft/model/NCToken.java | 2 +-
.../apache/nlpcraft/model/intent/compiler/NCIdlCompilerBase.scala | 1 +
.../scala/org/apache/nlpcraft/model/intent/compiler/antlr4/NCIdl.g4 | 1 +
.../nlpcraft/probe/mgrs/nlp/enrichers/model/NCModelEnricher.scala | 4 ++--
.../model/intent/idl/compiler/functions/NCIdlFunctionsToken.scala | 1 +
6 files changed, 7 insertions(+), 7 deletions(-)
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 9adbe01..3e60c61 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
@@ -42,7 +42,6 @@ class NCNlpSentenceNote(private val values: Map[String,
JSerializable]) extends
lazy val tokenIndexes: Seq[Int] =
values("tokWordIndexes").asInstanceOf[java.util.List[Int]].asScala // Includes
1st and last indices too.
lazy val wordIndexes: Seq[Int] =
values("wordIndexes").asInstanceOf[java.util.List[Int]].asScala // Includes 1st
and last indices too.
lazy val sparsity: Int = values("sparsity").asInstanceOf[Int]
- lazy val isContiguous: Boolean = values("contiguous").asInstanceOf[Boolean]
lazy val isDirect: Boolean = values("direct").asInstanceOf[Boolean]
lazy val isUser: Boolean = {
val i = noteType.indexOf(":")
@@ -198,7 +197,6 @@ object NCNlpSentenceNote {
"tokMinIndex",
"tokMaxIndex",
"tokWordIndexes",
- "contiguous",
"sparsity"
)
@@ -239,8 +237,7 @@ object NCNlpSentenceNote {
("maxIndex" → tokMaxIndex) :+
("wordIndexes" → tokWordIndexes) :+
("wordLength" → len) :+
- ("sparsity" → sparsity) :+
- ("contiguous" → (sparsity == 0))
+ ("sparsity" → sparsity)
).map(p ⇒ p._1 → p._2.asInstanceOf[JSerializable]): _*).toMap
)
}
diff --git a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/NCToken.java
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/NCToken.java
index 1e71758..bdf7fe9 100644
--- a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/NCToken.java
+++ b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/NCToken.java
@@ -432,7 +432,7 @@ public interface NCToken extends NCMetadata {
/**
* A shortcut method to get numeric value of how sparse the token is.
Sparsity zero means that all
- * individual words in the token follow each other.
+ * individual words in the token follow each other (regardless of the
order).
* <p>
* This method is equivalent to:
* <pre class="brush: java">
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/NCIdlCompilerBase.scala
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/NCIdlCompilerBase.scala
index 1846088..2389758 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/NCIdlCompilerBase.scala
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/NCIdlCompilerBase.scala
@@ -1100,6 +1100,7 @@ trait NCIdlCompilerBase {
case "tok_is_abstract" ⇒ arg1Tok() match { case x ⇒ stack.push(()
⇒ { Z(toToken(x().value).isAbstract, 1) }) }
case "tok_is_bracketed" ⇒ arg1Tok() match { case x ⇒ stack.push(()
⇒ { Z(toToken(x().value).isBracketed, 1) }) }
case "tok_is_direct" ⇒ arg1Tok() match { case x ⇒ stack.push(() ⇒
{ Z(toToken(x().value).isDirect, 1) }) }
+ case "tok_is_permutated" ⇒ arg1Tok() match { case x ⇒
stack.push(() ⇒ { Z(!toToken(x().value).isDirect, 1) }) }
case "tok_is_english" ⇒ arg1Tok() match { case x ⇒ stack.push(() ⇒
{ Z(toToken(x().value).isEnglish, 1) }) }
case "tok_is_freeWord" ⇒ arg1Tok() match { case x ⇒ stack.push(()
⇒ { Z(toToken(x().value).isFreeWord, 1) }) }
case "tok_is_quoted" ⇒ arg1Tok() match { case x ⇒ stack.push(() ⇒
{ Z(toToken(x().value).isQuoted, 1) }) }
diff --git
a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/antlr4/NCIdl.g4
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/antlr4/NCIdl.g4
index 1e62547..4d2d5dd 100644
---
a/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/antlr4/NCIdl.g4
+++
b/nlpcraft/src/main/scala/org/apache/nlpcraft/model/intent/compiler/antlr4/NCIdl.g4
@@ -157,6 +157,7 @@ FUN_NAME
| 'tok_is_abstract'
| 'tok_is_bracketed'
| 'tok_is_direct'
+ | 'tok_is_permutated'
| 'tok_is_english'
| 'tok_is_freeWord'
| 'tok_is_quoted'
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 1e1b441..85faaba 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
@@ -591,11 +591,11 @@ object NCModelEnricher extends NCProbeEnricher with
DecorateAsScala {
private def alreadyMarked(ns: Sentence, elemId: String, toks:
Seq[NlpToken], sliceToksIdxsSorted: Seq[Int]): Boolean = {
lazy val toksIdxsSorted = toks.map(_.index).sorted
- sliceToksIdxsSorted.map(ns).forall(_.exists(n ⇒ n.noteType == elemId
&& n.isContiguous)) ||
+ sliceToksIdxsSorted.map(ns).forall(_.exists(n ⇒ n.noteType == elemId
&& n.sparsity == 0)) ||
toks.exists(_.exists(n ⇒
n.noteType == elemId &&
(
- (n.isContiguous &&
+ (n.sparsity == 0 &&
(sliceToksIdxsSorted.containsSlice(n.tokenIndexes) ||
n.tokenIndexes.containsSlice(toksIdxsSorted))
)
||
diff --git
a/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctionsToken.scala
b/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctionsToken.scala
index cdf2e49..a83d448 100644
---
a/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctionsToken.scala
+++
b/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctionsToken.scala
@@ -63,6 +63,7 @@ class NCIdlFunctionsToken extends NCIdlFunctions {
mkMeta(truth = s"tok_is_abstract() == false"),
mkMeta(truth = s"tok_is_bracketed() ==
${meta("nlpcraft:nlp:bracketed")}"),
mkMeta(truth = s"tok_is_direct() ==
${meta("nlpcraft:nlp:direct")}"),
+ mkMeta(truth = s"tok_is_permutated() !=
${meta("nlpcraft:nlp:direct")}"),
mkMeta(truth = s"tok_is_english() ==
${meta("nlpcraft:nlp:english")}"),
mkMeta(truth = s"tok_is_freeWord() ==
${meta("nlpcraft:nlp:freeword")}"),
mkMeta(truth = s"tok_is_quoted() ==
${meta("nlpcraft:nlp:quoted")}"),
