[editor-core] speed up non-bmp detection (#34542)

* [editor-core] avoid duplicate non-bmp check on InsertOp.insertion

* [editor-core] remove non-bmp check from StringFileData

Existing content may contain non-bmp characters. New edits with non-bmp
input are rejected via non-bmp check in InsertOp constructor.

GitOrigin-RevId: e9ef3a4a8c4696e8dc751325608a788351791d1c

libraries/overleaf-editor-core/lib/operation/text_operation.js

@@ -9,7 +9,6 @@
  */
 
 'use strict'
-const containsNonBmpChars = require('../util').containsNonBmpChars
 const EditOperation = require('./edit_operation')
 const {
   RetainOp,
@@ -287,13 +286,6 @@ class TextOperation extends EditOperation {
   apply(file) {
     const str = file.getContent()
     const operation = this
-    if (containsNonBmpChars(str)) {
-      throw new TextOperation.ApplyError(
-        'The string contains non BMP characters.',
-        operation,
-        str
-      )
-    }
     if (str.length !== operation.baseLength) {
       throw new TextOperation.ApplyError(
         "The operation's base length must be equal to the string's length.",
@@ -317,9 +309,6 @@ class TextOperation extends EditOperation {
         result += str.slice(inputCursor, inputCursor + op.length)
         inputCursor += op.length
       } else if (op instanceof InsertOp) {
-        if (containsNonBmpChars(op.insertion)) {
-          throw new InvalidInsertionError(str, op.toJSON())
-        }
         file.comments.applyInsert(
           new Range(result.length, op.insertion.length),
           { commentIds: op.commentIds }

libraries/overleaf-editor-core/test/unit/string_file_data.test.js

@@ -9,16 +9,6 @@ const StringFileData = require('../../lib/file_data/string_file_data')
 const TextOperation = ot.TextOperation
 
 describe('StringFileData', function () {
-  it('throws when it contains non BMP chars', function () {
-    const content = '𝌆𝌆𝌆'
-    const fileData = new StringFileData(content)
-    const operation = new TextOperation()
-    operation.insert('aa')
-    expect(() => {
-      fileData.edit(operation)
-    }).to.throw(TextOperation.ApplyError, /string contains non BMP characters/)
-  })
-
   it('validates string length when edited', function () {
     const longString = _.repeat('a', TextOperation.MAX_STRING_LENGTH)
     const fileData = new StringFileData(longString)

libraries/overleaf-editor-core/test/unit/text_operation.test.js

@@ -193,17 +193,6 @@ describe('TextOperation', function () {
     )
   })
 
-  it('throws when base string contains non BMP chars', function () {
-    const operation = new TextOperation()
-    const str = '𝌆\n'
-    expect(() => {
-      operation.apply(new StringFileData(str))
-    }).to.throw(
-      TextOperation.UnprocessableError,
-      /string contains non BMP characters/
-    )
-  })
-
   it('throws at from JSON when it contains non BMP chars', function () {
     const operation = ['𝌆\n']
     expect(() => {