[KO NumberWithUnit] Korean Temperature refinements (#2830)

* Korean Temperature refinements * Fixed failing Duration cases due to unit change Co-authored-by: aitelint <[email protected]>
microsoft · Jan 11, 2022 · e6d72a0 · e6d72a0
1 parent f4744ce
commit e6d72a0
Show file tree

Hide file tree

Showing 6 changed files with 6 additions and 51 deletions.
diff --git a/.NET/Microsoft.Recognizers.Definitions.Common/Korean/NumbersDefinitions.cs b/.NET/Microsoft.Recognizers.Definitions.Common/Korean/NumbersDefinitions.cs
@@ -258,7 +258,7 @@ public static class NumbersDefinitions
       public static readonly string TwoNumberRangeRegex5 = $@"(?<number1>([십백천만억조경열]|(영|령|공|일|이(?!다)|두|삼|사|오|육|(?<!며)칠|팔|구))+)\s*{TillRegex}\s*(?<number2>([십백천만억조경열]|(영|령|공|일|이(?!다)|두|삼|사|오|육|칠|팔|구))+)\s*([십백천만억조경열]|((미만|적|낮|작|더적|더낮|더적|이하이다|까지|아래))+)";
       public static readonly string TwoNumberRangeRegex6 = $@"((?<number1>((?!((，(?!\d+))|(,(?!\d+))|。|\D)).)+)\s*{TillRegex}+\s*(?<number2>((?!((，(?!\d+))|(,(?!\d+))|。)).)+)\s*(까지))";
       public static readonly string TwoNumberRangeRegex7 = $@"({OneNumberRangeMoreRegex2}\s*{OneNumberRangeLessRegex5})";
-      public const string InexactNumberUnitRegex = @"(몇|며|여러)";
+      public const string InexactNumberUnitRegex = @"(몇(?!.+\?)|며|여러)";
       public static readonly Dictionary<string, string> RelativeReferenceOffsetMap = new Dictionary<string, string>
         {
             { @"마지막", @"0" },

diff --git a/.NET/Microsoft.Recognizers.Definitions.Common/Korean/NumbersWithUnitDefinitions.cs b/.NET/Microsoft.Recognizers.Definitions.Common/Korean/NumbersWithUnitDefinitions.cs
@@ -38,6 +38,7 @@ public static class NumbersWithUnitDefinitions
       public const string BuildPrefix = @"";
       public const string BuildSuffix = @"";
       public const string ConnectorToken = @"";
+      public const bool CheckFirstSuffix = true;
       public static readonly Dictionary<string, string> CurrencySuffixList = new Dictionary<string, string>
         {
             { @"Afghan afghani", @"아프가니" },

diff --git a/...icrosoft.Recognizers.Text.NumberWithUnit/Korean/Parsers/TemperatureParserConfiguration.cs b/...icrosoft.Recognizers.Text.NumberWithUnit/Korean/Parsers/TemperatureParserConfiguration.cs
@@ -19,6 +19,7 @@ public TemperatureParserConfiguration(CultureInfo ci)
         {
             this.BindDictionary(NumbersWithUnitDefinitions.TemperaturePrefixList);
             this.BindDictionary(NumbersWithUnitDefinitions.TemperatureSuffixList);
+            this.CheckFirstSuffix = NumbersWithUnitDefinitions.CheckFirstSuffix;
         }
     }
 }
diff --git a/Patterns/Korean/Korean-Numbers.yaml b/Patterns/Korean/Korean-Numbers.yaml
@@ -431,7 +431,7 @@ TwoNumberRangeRegex7: !nestedRegex
   def: ({OneNumberRangeMoreRegex2}\s*{OneNumberRangeLessRegex5})
   references: [OneNumberRangeMoreRegex2, OneNumberRangeLessRegex5]
 InexactNumberUnitRegex: !simpleRegex
-  def: (몇|며|여러)
+  def: (몇(?!.+\?)|며|여러)
 RelativeReferenceOffsetMap: !dictionary
   types: [ string, string ]
   entries:

diff --git a/Patterns/Korean/Korean-NumbersWithUnit.yaml b/Patterns/Korean/Korean-NumbersWithUnit.yaml
@@ -18,6 +18,7 @@ AgeSuffixList: !dictionary
 BuildPrefix: ''
 BuildSuffix: ''
 ConnectorToken: ''
+CheckFirstSuffix: !bool true
 #CurrencyExtractorConfiguration
 CurrencySuffixList: !dictionary
   types: [string, string]

diff --git a/Specs/NumberWithUnit/Korean/TemperatureModel.json b/Specs/NumberWithUnit/Korean/TemperatureModel.json
@@ -1,8 +1,6 @@
 [
   {
     "Input": "외부 온도는 섭씨 40도입니다. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -19,8 +17,6 @@
   },
   {
     "Input": "텍사스는 화씨 90도입니다. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -37,8 +33,6 @@
   },
   {
     "Input": "화씨 마이너스 5도",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -55,8 +49,6 @@
   },
   {
     "Input": "섭씨 6도",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -73,8 +65,6 @@
   },
   {
     "Input": "화씨 98.6도는 정상 온도입니다. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -91,8 +81,6 @@
   },
   {
     "Input": "온도를 섭씨 30도로 설정하십시오. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -109,8 +97,6 @@
   },
   {
     "Input": "정상 온도는 화씨 98.6도입니다. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -127,8 +113,6 @@
   },
   {
     "Input": "화씨 100 도 ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -145,8 +129,6 @@
   },
   {
     "Input": "섭씨 20 도 ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -163,8 +145,6 @@
   },
   {
     "Input": "화씨 100.2도는 낮다. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -181,16 +161,14 @@
   },
   {
     "Input": "10.5도 ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
         "Text": "10.5도",
         "TypeName": "temperature",
         "Resolution": {
           "value": "10.5",
-          "unit": "C"
+          "unit": "Degree"
         },
         "Start": 0,
         "End": 4
@@ -199,8 +177,6 @@
   },
   {
     "Input": "섭씨 20도 ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -297,8 +273,6 @@
   },
   {
     "Input": "온도를 화씨 70도로 설정하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -347,8 +321,6 @@
   },
   {
     "Input": "온도를 화씨 75도로 유지하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -365,8 +337,6 @@
   },
   {
     "Input": "온도를 섭씨 40도로 하세요.",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -399,8 +369,6 @@
   },
   {
     "Input": "섭씨 10도를 화씨로 변환하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -427,8 +395,6 @@
   },
   {
     "Input": "섭씨 34.9도를 화씨로",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -455,8 +421,6 @@
   },
   {
     "Input": "섭씨 200도를 화씨로 변환하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -483,8 +447,6 @@
   },
   {
     "Input": "화씨를 섭씨로 화씨 101도는 섭씨 몇 도입니까? ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -531,8 +493,6 @@
   },
   {
     "Input": "섭씨 50도 섭씨를 화씨로 ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -569,8 +529,6 @@
   },
   {
     "Input": "화씨 51도를 섭씨로 변환해 주시겠습니까?",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -597,8 +555,6 @@
   },
   {
     "Input": "화씨 106도를 섭씨로 변환하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -625,8 +581,6 @@
   },
   {
     "Input": "화씨 45도를 섭씨로 변환하세요. ",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {
@@ -653,8 +607,6 @@
   },
   {
     "Input": "화씨 마이너스 20도를 섭씨로 어떻게 변환하나요?",
-    "Comment": "The unit is split around the number (e.g. '섭씨 40도' = 'Celsius 40 degree') but only the last unit in the suffix is used to determine the unit type.",
-    "IgnoreResolution": true,
     "NotSupportedByDesign": "javascript,python,java",
     "Results": [
       {