Modify decimal regex to accept positive exponent specifier (#5649)

* Modify decimal regex to accept positive exponent specifier

* add test for positive exponent specifier

* nit

arrow-csv/src/reader/mod.rs

@@ -148,7 +148,7 @@ lazy_static! {
     static ref REGEX_SET: RegexSet = RegexSet::new([
         r"(?i)^(true)$|^(false)$(?-i)", //BOOLEAN
         r"^-?(\d+)$", //INTEGER
-        r"^-?((\d*\.\d+|\d+\.\d*)([eE]-?\d+)?|\d+([eE]-?\d+))$", //DECIMAL
+        r"^-?((\d*\.\d+|\d+\.\d*)([eE][-+]?\d+)?|\d+([eE][-+]?\d+))$", //DECIMAL
         r"^\d{4}-\d\d-\d\d$", //DATE32
         r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d(?:[^\d\.].*)?$", //Timestamp(Second)
         r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,3}(?:[^\d].*)?$", //Timestamp(Millisecond)
@@ -1658,6 +1658,27 @@ mod tests {
         assert_eq!(batch.schema().as_ref(), &expected_schema);
     }
 
+    #[test]
+    fn test_scientific_notation_with_inference() {
+        let mut file = File::open("test/data/scientific_notation_test.csv").unwrap();
+        let format = Format::default().with_header(false).with_delimiter(b',');
+
+        let (schema, _) = format.infer_schema(&mut file, None).unwrap();
+        file.rewind().unwrap();
+
+        let builder = ReaderBuilder::new(Arc::new(schema))
+            .with_format(format)
+            .with_batch_size(512)
+            .with_projection(vec![0, 1]);
+
+        let mut csv = builder.build(file).unwrap();
+        let batch = csv.next().unwrap().unwrap();
+
+        let schema = batch.schema();
+
+        assert_eq!(&DataType::Float64, schema.field(0).data_type());
+    }
+
     #[test]
     fn test_parse_invalid_csv() {
         let file = File::open("test/data/various_types_invalid.csv").unwrap();

arrow-csv/test/data/scientific_notation_test.csv

@@ -0,0 +1,19 @@
+1.439e+04, positive_exponent
+1.31e+04, positive_exponent
+1.2711e+0, positive_exponent
+1.44e+04, positive_exponent
+2.22e+04, positive_exponent
+1.149e+04, positive_exponent
+2.139e+04, positive_exponent
+7.322e+04, positive_exponent
+1.531e+04, positive_exponent
+2.206e-04, negative_exponent
+1.517e-04, negative_exponent
+2.332e-04, negative_exponent
+2.19e-04, negative_exponent
+2.087e-04, negative_exponent
+12683.18, no_exponent
+7134.6, no_exponent
+8540.17, no_exponent
+21462.36, no_exponent
+1120.76, no_exponent