HaoYang670 commented on code in PR #1784:
URL: https://github.com/apache/arrow-rs/pull/1784#discussion_r890102648
##########
arrow/src/compute/kernels/substring.rs:
##########
@@ -1083,6 +1135,164 @@ mod tests {
generic_string_with_non_zero_offset::<i64>()
}
+ fn with_nulls_generic_string_by_char<O: OffsetSizeTrait>() -> Result<()> {
+ let input_vals = vec![Some("hello"), None, Some("Γ ⊢x:T")];
+ let cases = vec![
+ // all-nulls array is always identical
+ (vec![None, None, None], 0, None, vec![None, None, None]),
+ // identity
+ (
+ input_vals.clone(),
+ 0,
+ None,
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ // 0 length -> Nothing
+ (
+ input_vals.clone(),
+ 0,
+ Some(0),
+ vec![Some(""), None, Some("")],
+ ),
+ // high start -> Nothing
+ (
+ input_vals.clone(),
+ 1000,
+ Some(0),
+ vec![Some(""), None, Some("")],
+ ),
+ // high negative start -> identity
+ (
+ input_vals.clone(),
+ -1000,
+ None,
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ // high length -> identity
+ (
+ input_vals.clone(),
+ 0,
+ Some(1000),
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ ];
+
+ cases.into_iter().try_for_each::<_, Result<()>>(
+ |(array, start, length, expected)| {
+ let array = GenericStringArray::<O>::from(array);
+ let result = substring_by_char(&array, start, length)?;
+ assert_eq!(array.len(), result.len());
+
+ let expected = GenericStringArray::<O>::from(expected);
+ assert_eq!(expected, result);
+ Ok(())
+ },
+ )?;
+
+ Ok(())
+ }
+
+ #[test]
+ fn with_nulls_string_by_char() -> Result<()> {
+ with_nulls_generic_string_by_char::<i32>()
+ }
+
+ #[test]
+ fn with_nulls_large_string_by_char() -> Result<()> {
+ with_nulls_generic_string_by_char::<i64>()
+ }
+
+ fn without_nulls_generic_string_by_char<O: OffsetSizeTrait>() ->
Result<()> {
+ let input_vals = vec!["hello", "", "Γ ⊢x:T"];
+ let cases = vec![
+ // empty array is always identical
+ (vec!["", "", ""], 0, None, vec!["", "", ""]),
+ // increase start
+ (input_vals.clone(), 0, None, vec!["hello", "", "Γ ⊢x:T"]),
+ (input_vals.clone(), 1, None, vec!["ello", "", " ⊢x:T"]),
+ (input_vals.clone(), 2, None, vec!["llo", "", "⊢x:T"]),
+ (input_vals.clone(), 3, None, vec!["lo", "", "x:T"]),
+ (input_vals.clone(), 10, None, vec!["", "", ""]),
+ // increase start negatively
+ (input_vals.clone(), -1, None, vec!["o", "", "T"]),
+ (input_vals.clone(), -2, None, vec!["lo", "", ":T"]),
+ (input_vals.clone(), -4, None, vec!["ello", "", "⊢x:T"]),
+ (input_vals.clone(), -10, None, vec!["hello", "", "Γ ⊢x:T"]),
+ // increase length
+ (input_vals.clone(), 1, Some(1), vec!["e", "", " "]),
+ (input_vals.clone(), 1, Some(2), vec!["el", "", " ⊢"]),
+ (input_vals.clone(), 1, Some(3), vec!["ell", "", " ⊢x"]),
+ (input_vals.clone(), 1, Some(6), vec!["ello", "", " ⊢x:T"]),
+ (input_vals.clone(), -4, Some(1), vec!["e", "", "⊢"]),
+ (input_vals.clone(), -4, Some(2), vec!["el", "", "⊢x"]),
+ (input_vals.clone(), -4, Some(3), vec!["ell", "", "⊢x:"]),
+ (input_vals.clone(), -4, Some(4), vec!["ello", "", "⊢x:T"]),
+ ];
+
+ cases.into_iter().try_for_each::<_, Result<()>>(
Review Comment:
Tracked by #1801
##########
arrow/src/compute/kernels/substring.rs:
##########
@@ -1083,6 +1135,164 @@ mod tests {
generic_string_with_non_zero_offset::<i64>()
}
+ fn with_nulls_generic_string_by_char<O: OffsetSizeTrait>() -> Result<()> {
+ let input_vals = vec![Some("hello"), None, Some("Γ ⊢x:T")];
+ let cases = vec![
+ // all-nulls array is always identical
+ (vec![None, None, None], 0, None, vec![None, None, None]),
+ // identity
+ (
+ input_vals.clone(),
+ 0,
+ None,
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ // 0 length -> Nothing
+ (
+ input_vals.clone(),
+ 0,
+ Some(0),
+ vec![Some(""), None, Some("")],
+ ),
+ // high start -> Nothing
+ (
+ input_vals.clone(),
+ 1000,
+ Some(0),
+ vec![Some(""), None, Some("")],
+ ),
+ // high negative start -> identity
+ (
+ input_vals.clone(),
+ -1000,
+ None,
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ // high length -> identity
+ (
+ input_vals.clone(),
+ 0,
+ Some(1000),
+ vec![Some("hello"), None, Some("Γ ⊢x:T")],
+ ),
+ ];
+
+ cases.into_iter().try_for_each::<_, Result<()>>(
+ |(array, start, length, expected)| {
+ let array = GenericStringArray::<O>::from(array);
+ let result = substring_by_char(&array, start, length)?;
+ assert_eq!(array.len(), result.len());
+
+ let expected = GenericStringArray::<O>::from(expected);
+ assert_eq!(expected, result);
+ Ok(())
+ },
+ )?;
+
+ Ok(())
+ }
+
+ #[test]
+ fn with_nulls_string_by_char() -> Result<()> {
+ with_nulls_generic_string_by_char::<i32>()
+ }
+
+ #[test]
+ fn with_nulls_large_string_by_char() -> Result<()> {
+ with_nulls_generic_string_by_char::<i64>()
+ }
+
+ fn without_nulls_generic_string_by_char<O: OffsetSizeTrait>() ->
Result<()> {
+ let input_vals = vec!["hello", "", "Γ ⊢x:T"];
+ let cases = vec![
+ // empty array is always identical
+ (vec!["", "", ""], 0, None, vec!["", "", ""]),
+ // increase start
+ (input_vals.clone(), 0, None, vec!["hello", "", "Γ ⊢x:T"]),
+ (input_vals.clone(), 1, None, vec!["ello", "", " ⊢x:T"]),
+ (input_vals.clone(), 2, None, vec!["llo", "", "⊢x:T"]),
+ (input_vals.clone(), 3, None, vec!["lo", "", "x:T"]),
+ (input_vals.clone(), 10, None, vec!["", "", ""]),
+ // increase start negatively
+ (input_vals.clone(), -1, None, vec!["o", "", "T"]),
+ (input_vals.clone(), -2, None, vec!["lo", "", ":T"]),
+ (input_vals.clone(), -4, None, vec!["ello", "", "⊢x:T"]),
+ (input_vals.clone(), -10, None, vec!["hello", "", "Γ ⊢x:T"]),
+ // increase length
+ (input_vals.clone(), 1, Some(1), vec!["e", "", " "]),
+ (input_vals.clone(), 1, Some(2), vec!["el", "", " ⊢"]),
+ (input_vals.clone(), 1, Some(3), vec!["ell", "", " ⊢x"]),
+ (input_vals.clone(), 1, Some(6), vec!["ello", "", " ⊢x:T"]),
+ (input_vals.clone(), -4, Some(1), vec!["e", "", "⊢"]),
+ (input_vals.clone(), -4, Some(2), vec!["el", "", "⊢x"]),
+ (input_vals.clone(), -4, Some(3), vec!["ell", "", "⊢x:"]),
+ (input_vals.clone(), -4, Some(4), vec!["ello", "", "⊢x:T"]),
+ ];
+
+ cases.into_iter().try_for_each::<_, Result<()>>(
+ |(array, start, length, expected)| {
+ let array = GenericStringArray::<O>::from(array);
+ let result = substring_by_char(&array, start, length)?;
+ assert_eq!(array.len(), result.len());
+ let expected = GenericStringArray::<O>::from(expected);
+ assert_eq!(expected, result);
+ Ok(())
+ },
+ )?;
+
+ Ok(())
+ }
+
+ #[test]
+ fn without_nulls_string_by_char() -> Result<()> {
+ without_nulls_generic_string_by_char::<i32>()
+ }
+
+ #[test]
+ fn without_nulls_large_string_by_char() -> Result<()> {
+ without_nulls_generic_string_by_char::<i64>()
+ }
+
+ fn generic_string_by_char_with_non_zero_offset<O: OffsetSizeTrait>() ->
Result<()> {
+ let values = "S→T = Πx:S.T";
Review Comment:
Tracked by #1801
--
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: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
