refactor: use functions to infer types in csv

arrow-csv/src/reader/mod.rs

@@ -133,7 +133,7 @@ use arrow_schema::*;
 use chrono::{TimeZone, Utc};
 use csv::StringRecord;
 use lazy_static::lazy_static;
-use regex::{Regex, RegexSet};
+use regex::Regex;
 use std::fmt::{self, Debug};
 use std::fs::File;
 use std::io::{BufRead, BufReader as StdBufReader, Read, Seek, SeekFrom};
@@ -145,16 +145,24 @@ use arrow_array::timezone::Tz;
 
 lazy_static! {
     /// Order should match [`InferredDataType`]
-    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{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].*)?$", //Timestamp(Millisecond)
-        r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,6}(?:[^\d].*)?$", //Timestamp(Microsecond)
-        r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,9}(?:[^\d].*)?$", //Timestamp(Nanosecond)
-    ]).unwrap();
+    static ref INFER_TYPE_FUNC: [fn(&str) -> bool; 8] = [
+        |s| s.eq_ignore_ascii_case("true") || s.eq_ignore_ascii_case("false"), // Boolean
+        |s| s.parse::<i64>().is_ok(), // Integer
+        // TODO: replace regex with functions.
+        |s| DECIMAL_REGEX.is_match(s), // Decimal
+        |s| DATE32_REGEX.is_match(s), // Date32
+        |s| TIMESTAMP_SECOND_REGEX.is_match(s), // Timestamp(Second)
+        |s| TIMESTAMP_MILLISECOND_REGEX.is_match(s), // Timestamp(Millisecond)
+        |s| TIMESTAMP_MICROSECOND_REGEX.is_match(s), // Timestamp(Microsecond)
+        |s| TIMESTAMP_NANOSECOND_REGEX.is_match(s), // Timestamp(Nanosecond)
+    ];
+
+    static ref DECIMAL_REGEX: Regex = Regex::new(r"(^-?((\d*\.\d+|\d+\.\d*)([eE][-+]?\d+)?|\d+([eE][-+]?\d+))$)|(^-?(\d+)$)").unwrap();
+    static ref DATE32_REGEX: Regex = Regex::new(r"^\d{4}-\d\d-\d\d$").unwrap();
+    static ref TIMESTAMP_SECOND_REGEX: Regex = Regex::new(r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d(?:[^\d\.].*)?$").unwrap();
+    static ref TIMESTAMP_MILLISECOND_REGEX: Regex = Regex::new(r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,3}(?:[^\d].*)?$").unwrap();
+    static ref TIMESTAMP_MICROSECOND_REGEX: Regex = Regex::new(r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,6}(?:[^\d].*)?$").unwrap();
+    static ref TIMESTAMP_NANOSECOND_REGEX: Regex = Regex::new(r"^\d{4}-\d\d-\d\d[T ]\d\d:\d\d:\d\d\.\d{1,9}(?:[^\d].*)?$").unwrap();
 }
 
 /// A wrapper over `Option<Regex>` to check if the value is `NULL`.
@@ -214,7 +222,7 @@ impl InferredDataType {
     fn update(&mut self, string: &str) {
         self.packed |= if string.starts_with('"') {
             1 << 8 // Utf8
-        } else if let Some(m) = REGEX_SET.matches(string).into_iter().next() {
+        } else if let Some(m) = INFER_TYPE_FUNC.iter().position(|f| f(string)) {
             1 << m
         } else {
             1 << 8 // Utf8