fix(transformer): Correctly trim JSX

crates/oxc_syntax/src/identifier.rs

@@ -28,6 +28,9 @@ pub const VT: char = '\u{b}';
 /// U+000C FORM FEED, abbreviated `<FF>`.
 pub const FF: char = '\u{c}';
 
+/// U+0020 SPACE, abbreviated `<SP>`.
+pub const SP: char = '\u{20}';
+
 /// U+00A0 NON-BREAKING SPACE, abbreviated `<NBSP>`.
 pub const NBSP: char = '\u{a0}';
 
@@ -39,6 +42,26 @@ pub fn is_irregular_whitespace(c: char) -> bool {
     )
 }
 
+const NEL: char = '\u{85}';
+const OGHAM_SPACE_MARK: char = '\u{1680}';
+const EN_QUAD: char = '\u{2000}';
+const ZWSP: char = '\u{200b}';
+
+// Narrow No-Break Space
+const NNBSP: char = '\u{202f}';
+
+// Medium Mathematical Space
+const MMSP: char = '\u{205f}';
+const IDEOGRAPHIC_SPACE: char = '\u{3000}';
+
+// https://github.com/microsoft/TypeScript/blob/b8e4ed8aeb0b228f544c5736908c31f136a9f7e3/src/compiler/scanner.ts#L556
+pub fn is_white_space_single_line(c: char) -> bool {
+    // Note: nextLine is in the Zs space, and should be considered to be a whitespace.
+    // It is explicitly not a line-break as it isn't in the exact set specified by EcmaScript.
+    matches!(c, |SP| TAB | VT | FF | NBSP | NEL | OGHAM_SPACE_MARK | EN_QUAD
+        ..=ZWSP | NNBSP | MMSP | IDEOGRAPHIC_SPACE | ZWNBSP)
+}
+
 // 11.3 Line Terminators
 
 ///  U+000A LINE FEED, abbreviated in the spec as `<LF>`.

crates/oxc_transformer/src/react/jsx.rs

@@ -93,7 +93,7 @@ use oxc_ast::{ast::*, AstBuilder, NONE};
 use oxc_ecmascript::PropName;
 use oxc_span::{Atom, GetSpan, Span, SPAN};
 use oxc_syntax::{
-    identifier::{is_irregular_whitespace, is_line_terminator},
+    identifier::{is_line_terminator, is_white_space_single_line},
     reference::ReferenceFlags,
     symbol::SymbolFlags,
     xml_entities::XML_ENTITIES,
@@ -909,7 +909,7 @@ impl<'a, 'ctx> ReactJsx<'a, 'ctx> {
                     acc = Some(Self::add_line_of_jsx_text(acc, &text[first..last]));
                 }
                 first_non_whitespace = None;
-            } else if c != ' ' && !is_irregular_whitespace(c) {
+            } else if !is_white_space_single_line(c) {
                 last_non_whitespace = Some(index + c.len_utf8());
                 if first_non_whitespace.is_none() {
                     first_non_whitespace.replace(index);