From ea9a9fbe1cfbfc3e897cc88e6928740f91aba486 Mon Sep 17 00:00:00 2001
From: Alf Eaton <alf.eaton@overleaf.com>
Date: Fri, 4 Aug 2023 08:37:00 +0100
Subject: [PATCH] [cm6] Use less string casting, refactor tokenizers (#13898)

* Use more char definitions in tokenizers
* Refactor tokenizers

GitOrigin-RevId: 89f49da15f2bd64419fd2f19b0f255625e1d75cc
---
 .../source-editor/lezer-latex/tokens.mjs      | 57 ++++---------------
 1 file changed, 11 insertions(+), 46 deletions(-)

diff --git a/services/web/frontend/js/features/source-editor/lezer-latex/tokens.mjs b/services/web/frontend/js/features/source-editor/lezer-latex/tokens.mjs
index 933bbd7a71..0cde16787f 100644
--- a/services/web/frontend/js/features/source-editor/lezer-latex/tokens.mjs
+++ b/services/web/frontend/js/features/source-editor/lezer-latex/tokens.mjs
@@ -197,73 +197,38 @@ const matchForward = (input, expected, offset = 0) => {
 export const verbatimTokenizer = new ExternalTokenizer(
   (input, stack) => {
     const delimiter = '\\end{' + stack.context.name + '}'
-    let offset = 0
-    let end = -1
-    for (;;) {
+    for (let offset = 0; ; offset++) {
       const next = input.peek(offset)
-      if (next === -1) {
-        end = offset - 1
-        break
+      if (next === -1 || matchForward(input, delimiter, offset)) {
+        return input.acceptToken(VerbatimContent, offset)
       }
-      if (matchForward(input, delimiter, offset)) {
-        // Found the end marker
-        end = offset - 1
-        break
-      }
-      offset++
     }
-    return input.acceptToken(VerbatimContent, end + 1)
   },
   { contextual: false }
 )
 
 // tokenizer for \href{...} and similar commands
 export const literalArgTokenizer = new ExternalTokenizer(
-  (input, stack) => {
-    const delimiter = '}'
-    let content = ''
-    let offset = 0
-    let end = -1
-    for (;;) {
+  input => {
+    for (let offset = 0; ; offset++) {
       const next = input.peek(offset)
-      if (next === -1) {
-        end = offset - 1
-        break
+      if (next === -1 || next === CHAR_CLOSE_BRACE) {
+        return input.acceptToken(LiteralArgContent, offset)
       }
-      content += String.fromCharCode(next)
-      if (content.slice(-delimiter.length) === delimiter) {
-        // found the '}'
-        end = offset - delimiter.length
-        break
-      }
-      offset++
     }
-    return input.acceptToken(LiteralArgContent, end + 1)
   },
   { contextual: false }
 )
 
 // tokenizer for literal content delimited by whitespace, such as in `\input foo.tex`
 export const spaceDelimitedLiteralArgTokenizer = new ExternalTokenizer(
-  (input, stack) => {
-    let content = ''
-    let offset = 0
-    let end = -1
-    for (;;) {
+  input => {
+    for (let offset = 0; ; offset++) {
       const next = input.peek(offset)
-      if (next === -1) {
-        end = offset - 1
-        break
+      if (next === -1 || next === CHAR_SPACE || next === CHAR_NEWLINE) {
+        return input.acceptToken(SpaceDelimitedLiteralArgContent, offset)
       }
-      content += String.fromCharCode(next)
-      if (content.slice(-1) === ' ' || content.slice(-1) === '\n') {
-        // found the whitespace
-        end = offset - 1
-        break
-      }
-      offset++
     }
-    return input.acceptToken(SpaceDelimitedLiteralArgContent, end + 1)
   },
   { contextual: false }
 )