This is an automated email from the ASF dual-hosted git repository.
sergeykamov pushed a commit to branch NLPCRAFT-278
in repository https://gitbox.apache.org/repos/asf/incubator-nlpcraft.git
The following commit(s) were added to refs/heads/NLPCRAFT-278 by this push:
new d4f5e75 WIP.
d4f5e75 is described below
commit d4f5e751430602de5939327089614c1319245373
Author: Sergey Kamov <[email protected]>
AuthorDate: Wed Mar 24 12:15:52 2021 +0300
WIP.
---
.../idl/compiler/functions/NCIdlFunctions.scala | 3 +-
.../compiler/functions/NCIdlFunctionsToken.scala | 94 +++++++++++++++++++++-
2 files changed, 93 insertions(+), 4 deletions(-)
diff --git
a/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctions.scala
b/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctions.scala
index 05ecc35..c47f569 100644
---
a/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctions.scala
+++
b/nlpcraft/src/test/scala/org/apache/nlpcraft/model/intent/idl/compiler/functions/NCIdlFunctions.scala
@@ -118,6 +118,7 @@ private[functions] trait NCIdlFunctions {
ancestors: Seq[String] = Seq.empty,
aliases: Set[String] = Set.empty,
partTokens: Seq[NCToken] = Seq.empty,
+ isAbstr: Boolean = false,
meta: Map[String, AnyRef] = Map.empty[String, AnyRef]
): NCToken = {
val map = new util.HashMap[String, AnyRef]
@@ -140,7 +141,7 @@ private[functions] trait NCIdlFunctions {
if (groups.isEmpty && id != null)
Collections.singletonList(id) else groups.asJava
override def getStartCharIndex: Int = start
override def getEndCharIndex: Int = end
- override def isAbstract: Boolean = false
+ override def isAbstract: Boolean = isAbstr
override def getMetadata: util.Map[String, AnyRef] = map
}
}
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 1d3919c..8725886 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
@@ -23,6 +23,27 @@ import org.junit.jupiter.api.Test
* Tests for 'tokens' functions.
*/
class NCIdlFunctionsToken extends NCIdlFunctions {
+ private final val meta: Map[String, AnyRef] = Map(
+ "nlpcraft:nlp:stopword" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:freeword" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:origtext" → "orig text",
+ "nlpcraft:nlp:index" → java.lang.Integer.valueOf(11),
+ "nlpcraft:nlp:normtext" → "norm text",
+ "nlpcraft:nlp:direct" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:english" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:swear" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:quoted" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:bracketed" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:dict" → java.lang.Boolean.TRUE,
+ "nlpcraft:nlp:lemma" → "lemma",
+ "nlpcraft:nlp:stem" → "stem",
+ "nlpcraft:nlp:sparsity" → java.lang.Integer.valueOf(112),
+ "nlpcraft:nlp:pos" → "pos",
+ "nlpcraft:nlp:unid" → "21421"
+ )
+
+ private def mkMeta(truth: String):TestDesc = TestDesc(truth = truth, token
= tkn(meta = meta))
+
@Test
def test(): Unit =
test(
@@ -30,9 +51,76 @@ class NCIdlFunctionsToken extends NCIdlFunctions {
truth = "tok_id() == 'a'",
token = tkn(id = "a")
),
+ mkMeta(truth = s"tok_lemma() == '${meta("nlpcraft:nlp:lemma")}'"),
+ mkMeta(truth = s"tok_stem() == '${meta("nlpcraft:nlp:stem")}'"),
+ mkMeta(truth = s"tok_pos() == '${meta("nlpcraft:nlp:pos")}'"),
+ mkMeta(truth = s"tok_sparsity() ==
${meta("nlpcraft:nlp:sparsity")}"),
+ mkMeta(truth = s"tok_unid() == '${meta("nlpcraft:nlp:unid")}'"),
+ TestDesc(
+ truth = s"tok_is_abstract() == true",
+ token = tkn(isAbstr = true)
+ ),
+ 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_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")}"),
+ mkMeta(truth = s"tok_is_stopword() ==
${meta("nlpcraft:nlp:stopword")}"),
+ mkMeta(truth = s"tok_is_swear() == ${meta("nlpcraft:nlp:swear")}"),
+ TestDesc(
+ truth = s"tok_is_user() == true",
+ token = tkn(id = "aa")
+ ),
+ TestDesc(
+ truth = s"tok_is_user() == false",
+ token = tkn(id = "nlpcraft:nlp")
+ ),
+ mkMeta(truth = s"tok_is_wordnet() ==
${meta("nlpcraft:nlp:dict")}"),
+ TestDesc(
+ truth = s"tok_ancestors() == list('1', '2')",
+ token = tkn(ancestors = Seq("1", "2"))
+ ),
+ TestDesc(
+ truth = s"tok_parent() == 'parentId'",
+ token = tkn(parentId = "parentId")
+ ),
+ TestDesc(
+ truth = "tok_groups() == list('1', '2')",
+ token = tkn(groups = Seq("1", "2"))
+ ),
TestDesc(
- truth = "tok_parent() == 'a'",
- token = tkn(parentId = "a")
- )
+ truth = "tok_value() == 'value'",
+ token = tkn(value = "value")
+ ),
+ TestDesc(
+ truth = "tok_value() == null",
+ token = tkn()
+ ),
+ TestDesc(
+ truth = "tok_start_idx() == 123",
+ token = tkn(start = 123)
+ ),
+ TestDesc(
+ truth = "tok_end_idx() == 123",
+ token = tkn(end = 123)
+ ),
+ // TODO: fix following.
+// TestDesc(
+// truth = "tok_aliases() == list('1')",
+// token = tkn(aliases = Set("1"))
+// ),
+// TestDesc(
+// truth = "tok_this() == tok_this()"
+// ),
+// TestDesc(
+// truth = "tok_find_part() != null",
+// token = tkn(partTokens = Seq(tkn("part1"), tkn("part2")))
+// ),
+// TestDesc(
+// truth = "tok_find_parts() != null",
+// token = tkn(partTokens = Seq(tkn("part1"), tkn("part2")))
+// ),
+
)
}
