fix(editor): enhance markdown latex preprocessing (#11597)

Close [BS-2440](https://linear.app/affine-design/issue/BS-2440/inline-latex-markdown-adapter-需要更精确的处理)

blocksuite/affine/all/src/__tests__/adapters/markdown.unit.spec.ts

@@ -3769,6 +3769,48 @@ bbb
       });
       expect(nanoidReplacement(rawBlockSnapshot)).toEqual(blockSnapshot);
     });
+
+    test('escapes dollar signs followed by a digit or space and digit', async () => {
+      const markdown =
+        'The price of the T-shirt is $9.15 and the price of the hat is $ 8\n';
+      const blockSnapshot: BlockSnapshot = {
+        type: 'block',
+        id: 'matchesReplaceMap[0]',
+        flavour: 'affine:note',
+        props: {
+          xywh: '[0,0,800,95]',
+          background: DefaultTheme.noteBackgrounColor,
+          index: 'a0',
+          hidden: false,
+          displayMode: NoteDisplayMode.DocAndEdgeless,
+        },
+        children: [
+          {
+            type: 'block',
+            id: 'matchesReplaceMap[1]',
+            flavour: 'affine:paragraph',
+            props: {
+              type: 'text',
+              text: {
+                '$blocksuite:internal:text$': true,
+                delta: [
+                  {
+                    insert:
+                      'The price of the T-shirt is $9.15 and the price of the hat is $ 8',
+                  },
+                ],
+              },
+            },
+            children: [],
+          },
+        ],
+      };
+      const mdAdapter = new MarkdownAdapter(createJob(), provider);
+      const rawBlockSnapshot = await mdAdapter.toBlockSnapshot({
+        file: markdown,
+      });
+      expect(nanoidReplacement(rawBlockSnapshot)).toEqual(blockSnapshot);
+    });
   });
 
   test('reference', async () => {

blocksuite/affine/blocks/latex/src/adapters/markdown/preprocessor.ts

@@ -3,21 +3,84 @@ import {
   MarkdownPreprocessorExtension,
 } from '@blocksuite/affine-shared/adapters';
 
+function escapeBrackets(text: string) {
+  const pattern =
+    /(```[\S\s]*?```|`.*?`)|\\\[([\S\s]*?[^\\])\\]|\\\((.*?)\\\)/g;
+  return text.replaceAll(
+    pattern,
+    (match, codeBlock, squareBracket, roundBracket) => {
+      if (codeBlock) {
+        return codeBlock;
+      } else if (squareBracket) {
+        return `$$${squareBracket}$$`;
+      } else if (roundBracket) {
+        return `$${roundBracket}$`;
+      }
+      return match;
+    }
+  );
+}
+
+function escapeMhchem(text: string) {
+  return text.replaceAll('$\\ce{', '$\\\\ce{').replaceAll('$\\pu{', '$\\\\pu{');
+}
+
+/**
+ * Preprocess the content to protect code blocks and LaTeX expressions
+ * reference issue: https://github.com/remarkjs/react-markdown/issues/785
+ * reference comment: https://github.com/remarkjs/react-markdown/issues/785#issuecomment-2307567823
+ * @param content - The content to preprocess
+ * @returns The preprocessed content
+ */
+function preprocessLatex(content: string) {
+  // Protect code blocks
+  const codeBlocks: string[] = [];
+  let preprocessedContent = content;
+  preprocessedContent = preprocessedContent.replace(
+    /(```[\s\S]*?```|`[^`\n]+`)/g,
+    (_, code) => {
+      codeBlocks.push(code);
+      return `<<CODE_BLOCK_${codeBlocks.length - 1}>>`;
+    }
+  );
+
+  // Protect existing LaTeX expressions
+  const latexExpressions: string[] = [];
+  preprocessedContent = preprocessedContent.replace(
+    /(\$\$[\s\S]*?\$\$|\\\[[\s\S]*?\\\]|\\\(.*?\\\))/g,
+    match => {
+      latexExpressions.push(match);
+      return `<<LATEX_${latexExpressions.length - 1}>>`;
+    }
+  );
+
+  // Escape dollar signs that are likely currency indicators
+  preprocessedContent = preprocessedContent.replace(/\$(?=\d)/g, '\\$');
+
+  // Restore LaTeX expressions
+  preprocessedContent = preprocessedContent.replace(
+    /<<LATEX_(\d+)>>/g,
+    (_, index) => latexExpressions[parseInt(index)]
+  );
+
+  // Restore code blocks
+  preprocessedContent = preprocessedContent.replace(
+    /<<CODE_BLOCK_(\d+)>>/g,
+    (_, index) => codeBlocks[parseInt(index)]
+  );
+
+  // Apply additional escaping functions
+  preprocessedContent = escapeBrackets(preprocessedContent);
+  preprocessedContent = escapeMhchem(preprocessedContent);
+
+  return preprocessedContent;
+}
+
 const latexPreprocessor: MarkdownAdapterPreprocessor = {
   name: 'latex',
   levels: ['block', 'slice', 'doc'],
   preprocess: content => {
-    // Replace block-level LaTeX delimiters \[ \] with $$ $$
+    return preprocessLatex(content);
-    const blockProcessedContent = content.replace(
-      /\\\[(.*?)\\\]/gs,
-      (_, equation) => `$$${equation}$$`
-    );
-    // Replace inline LaTeX delimiters \( \) with $ $
-    const inlineProcessedContent = blockProcessedContent.replace(
-      /\\\((.*?)\\\)/gs,
-      (_, equation) => `$${equation}$`
-    );
-    return inlineProcessedContent;
   },
 };