vidyasankarv commented on code in PR #383:
URL: https://github.com/apache/datafusion-comet/pull/383#discussion_r1600520823
##########
core/src/execution/datafusion/expressions/cast.rs:
##########
@@ -1444,13 +1483,136 @@ fn parse_str_to_time_only_timestamp(value: &str) ->
CometResult<Option<i64>> {
Ok(Some(timestamp))
}
+//a string to date parser - port of spark's SparkDateTimeUtils#stringToDate.
+fn date_parser(date_str: &str, eval_mode: EvalMode) ->
CometResult<Option<i32>> {
+ // local functions
+ fn get_trimmed_start(bytes: &[u8]) -> usize {
+ let mut start = 0;
+ while start < bytes.len() &&
is_whitespace_or_iso_control(bytes[start]) {
+ start += 1;
+ }
+ start
+ }
+
+ fn get_trimmed_end(start: usize, bytes: &[u8]) -> usize {
+ let mut end = bytes.len() - 1;
+ while end > start && is_whitespace_or_iso_control(bytes[end]) {
+ end -= 1;
+ }
+ end + 1
+ }
+
+ fn is_whitespace_or_iso_control(byte: u8) -> bool {
+ byte.is_ascii_whitespace() || byte.is_ascii_control()
+ }
+
+ fn is_valid_digits(segment: i32, digits: usize) -> bool {
+ // An integer is able to represent a date within [+-]5 million years.
+ let max_digits_year = 7;
+ //year (segment 0) can be between 4 to 7 digits,
+ //month and day (segment 1 and 2) can be between 1 to 2 digits
+ (segment == 0 && digits >= 4 && digits <= max_digits_year)
+ || (segment != 0 && digits > 0 && digits <= 2)
+ }
+
+ fn return_result(date_str: &str, eval_mode: EvalMode) ->
CometResult<Option<i32>> {
+ if eval_mode == EvalMode::Ansi {
+ Err(CometError::CastInvalidValue {
+ value: date_str.to_string(),
+ from_type: "STRING".to_string(),
+ to_type: "DATE".to_string(),
+ })
+ } else {
+ Ok(None)
+ }
+ }
+ // end local functions
+
+ if date_str.is_empty() {
+ return return_result(date_str, eval_mode);
+ }
+
+ //values of date segments year, month and day defaulting to 1
+ let mut date_segments = [1, 1, 1];
+ let mut sign = 1;
+ let mut current_segment = 0;
+ let mut current_segment_value = 0;
+ let mut current_segment_digits = 0;
+ let bytes = date_str.as_bytes();
+
+ let mut j = get_trimmed_start(bytes);
+ let str_end_trimmed = get_trimmed_end(j, bytes);
+
+ if j == str_end_trimmed {
+ return return_result(date_str, eval_mode);
+ }
+
+ //assign a sign to the date
+ if bytes[j] == b'-' || bytes[j] == b'+' {
+ sign = if bytes[j] == b'-' { -1 } else { 1 };
+ j += 1;
+ }
+
+ //loop to the end of string until we have processed 3 segments,
+ //exit loop on encountering any space ' ' or 'T' after the 3rd segment
+ while j < str_end_trimmed && (current_segment < 3 && !(bytes[j] == b' ' ||
bytes[j] == b'T')) {
+ let b = bytes[j];
+ if current_segment < 2 && b == b'-' {
+ //check for validity of year and month segments if current byte is
separator
+ if !is_valid_digits(current_segment, current_segment_digits) {
+ return return_result(date_str, eval_mode);
+ }
+ //if valid update corresponding segment with the current segment
value.
+ date_segments[current_segment as usize] = current_segment_value;
+ current_segment_value = 0;
+ current_segment_digits = 0;
+ current_segment += 1;
+ } else {
+ //increment value of current segment by the next digit
+ let parsed_value = (b - b'0') as i32;
Review Comment:
@andygrove added check for ascii digits and some negative test cases around
this in rust code. thank you
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: github-unsubscr...@datafusion.apache.org
For additional commands, e-mail: github-h...@datafusion.apache.org
