Improve function name

Author: neilconway

datafusion/functions/src/unicode/common.rs

@@ -88,26 +88,26 @@ pub(crate) fn byte_offset_of_char(string: &str, n: usize) -> usize {
         .map_or(string.len(), |(i, _)| i)
 }
 
-/// Measures the character length of `string` in a single pass, returning
-/// early with the truncation byte offset if the string exceeds `n` chars.
+/// If `string` has more than `n` codepoints, returns the byte offset of
+/// the `n`-th codepoint boundary. Otherwise returns the total codepoint count.
 #[inline]
-pub(crate) fn measure_char_count(string: &str, n: usize) -> StringCharLen {
+pub(crate) fn char_count_or_boundary(string: &str, n: usize) -> StringCharLen {
     let mut count = 0;
     for (byte_idx, _) in string.char_indices() {
         if count == n {
-            return StringCharLen::TruncateAt(byte_idx);
+            return StringCharLen::ByteOffset(byte_idx);
         }
         count += 1;
     }
     StringCharLen::CharCount(count)
 }
 
-/// Result of [`measure_char_count`].
+/// Result of [`char_count_or_boundary`].
 pub(crate) enum StringCharLen {
-    /// The string has more than `n` chars; contains the byte offset at the
-    /// `n`-th character boundary (suitable for slicing `&string[..offset]`).
-    TruncateAt(usize),
-    /// The string has `n` or fewer chars; contains the exact character count.
+    /// The string has more than `n` codepoints; contains the byte offset
+    /// at the `n`-th codepoint boundary.
+    ByteOffset(usize),
+    /// The string has `n` or fewer codepoints; contains the exact count.
     CharCount(usize),
 }
 

datafusion/functions/src/unicode/lpad.rs

@@ -178,7 +178,7 @@ impl ScalarUDFImpl for LPadFunc {
 }
 
 use super::common::{
-    StringCharLen, measure_char_count, try_as_scalar_i64, try_as_scalar_str,
+    StringCharLen, char_count_or_boundary, try_as_scalar_i64, try_as_scalar_str,
 };
 
 /// Optimized lpad for constant target_len and fill arguments.
@@ -275,8 +275,8 @@ fn lpad_scalar_unicode<'a, V: StringArrayType<'a> + Copy, T: OffsetSizeTrait>(
     for maybe_string in string_array.iter() {
         match maybe_string {
             Some(string) => {
-                match measure_char_count(string, target_len) {
-                    StringCharLen::TruncateAt(offset) => {
+                match char_count_or_boundary(string, target_len) {
+                    StringCharLen::ByteOffset(offset) => {
                         builder.append_value(&string[..offset]);
                     }
                     StringCharLen::CharCount(char_count) => {
@@ -427,8 +427,8 @@ where
                     fill_chars_buf.clear();
                     fill_chars_buf.extend(fill.chars());
 
-                    match measure_char_count(string, target_len) {
-                        StringCharLen::TruncateAt(offset) => {
+                    match char_count_or_boundary(string, target_len) {
+                        StringCharLen::ByteOffset(offset) => {
                             builder.append_value(&string[..offset]);
                         }
                         StringCharLen::CharCount(char_count) => {
@@ -484,8 +484,8 @@ where
                         builder.append_value(string);
                     }
                 } else {
-                    match measure_char_count(string, target_len) {
-                        StringCharLen::TruncateAt(offset) => {
+                    match char_count_or_boundary(string, target_len) {
+                        StringCharLen::ByteOffset(offset) => {
                             builder.append_value(&string[..offset]);
                         }
                         StringCharLen::CharCount(char_count) => {

datafusion/functions/src/unicode/rpad.rs

@@ -178,7 +178,7 @@ impl ScalarUDFImpl for RPadFunc {
 }
 
 use super::common::{
-    StringCharLen, measure_char_count, try_as_scalar_i64, try_as_scalar_str,
+    StringCharLen, char_count_or_boundary, try_as_scalar_i64, try_as_scalar_str,
 };
 
 /// Optimized rpad for constant target_len and fill arguments.
@@ -276,8 +276,8 @@ fn rpad_scalar_unicode<'a, V: StringArrayType<'a> + Copy, T: OffsetSizeTrait>(
     for maybe_string in string_array.iter() {
         match maybe_string {
             Some(string) => {
-                match measure_char_count(string, target_len) {
-                    StringCharLen::TruncateAt(offset) => {
+                match char_count_or_boundary(string, target_len) {
+                    StringCharLen::ByteOffset(offset) => {
                         builder.append_value(&string[..offset]);
                     }
                     StringCharLen::CharCount(char_count) => {
@@ -427,8 +427,8 @@ where
                     fill_chars_buf.clear();
                     fill_chars_buf.extend(fill.chars());
 
-                    match measure_char_count(string, target_len) {
-                        StringCharLen::TruncateAt(offset) => {
+                    match char_count_or_boundary(string, target_len) {
+                        StringCharLen::ByteOffset(offset) => {
                             builder.append_value(&string[..offset]);
                         }
                         StringCharLen::CharCount(char_count) => {
@@ -486,8 +486,8 @@ where
                         builder.append_value("");
                     }
                 } else {
-                    match measure_char_count(string, target_len) {
-                        StringCharLen::TruncateAt(offset) => {
+                    match char_count_or_boundary(string, target_len) {
+                        StringCharLen::ByteOffset(offset) => {
                             builder.append_value(&string[..offset]);
                         }
                         StringCharLen::CharCount(char_count) => {