NL/amount-of-money (#504)

Summary:
Currently values like 1000.000 (in Dutch . is thousand separator) are not recognised, as the ruleDecimalWithThousandsSeparator requires the decimal part (e.g. 1000.000,34) to be present. This PR adds some data and changes the ruleDecimalWithThousandsSeparator to make the decimal part optional.

Pull Request resolved: https://github.com/facebook/duckling/pull/504

Reviewed By: patapizza, girifb

Differential Revision: D26078885

Pulled By: chessai

fbshipit-source-id: b1679c713e1d17a168d34a3cc556b6c36a571d75

Duckling/AmountOfMoney/NL/Corpus.hs

@@ -46,6 +46,8 @@ allExamples = concat
              [ "$10K"
              , "10k$"
              , "$10000"
+             , "$10.000"
+             , "$10.000,00"
              , "10000,00 $"
              ]
   , examples (simple USD 3.14)

Duckling/Numeral/NL/Corpus.hs

@@ -76,6 +76,14 @@ allExamples = concat
              [ "0,77"
              , ",77"
              ]
+  , examples (NumeralValue 100000)
+             [ "100.000"
+             , "100.000,0"
+             , "100000"
+             , "100K"
+             , "100k"
+             , "honderd duizend"
+             ]
   , examples (NumeralValue 300)
              [ "3 honderd"
              , "drie honderd"
@@ -83,6 +91,8 @@ allExamples = concat
   , examples (NumeralValue 5000)
              [ "5 duizend"
              , "vijf duizend"
+             , "5.000"
+             , "5.000,00"
              ]
   , examples (NumeralValue 144)
              [ "gros"
@@ -94,6 +104,9 @@ allExamples = concat
              ]
   , examples (NumeralValue 20000)
              [ "twintig duizend"
+             , "20.000"
+             , "20000"
+             , "20.000,00"
              ]
   , examples (NumeralValue 0.2)
              [ "1/5"

Duckling/Numeral/NL/Rules.hs

@@ -62,7 +62,7 @@ ruleDecimalWithThousandsSeparator :: Rule
 ruleDecimalWithThousandsSeparator = Rule
   { name = "decimal with thousands separator"
   , pattern =
-    [ regex "(\\d+(\\.\\d\\d\\d)+,\\d+)"
+    [ regex "(\\d+(\\.\\d\\d\\d)+(,\\d+)?)"
     ]
   , prod = \tokens -> case tokens of
       (Token RegexMatch (GroupMatch (match:_)):