feat(LanguageServiceProvider): refactored logic to be in addItem()

Signed-off-by: Kishan Patel <[email protected]>

Author: kpatl1

server/src/sas/LanguageServiceProvider.ts

@@ -216,8 +216,8 @@ export class LanguageServiceProvider {
   }
 
   private addCommentFolding(result: FoldingRange[]) {
-    // Get all comment ranges from the lexer
-    const commentRanges = this.syntaxProvider.getAllCommentRanges();
+    // Get multiline comments directly from lexer token processing
+    const commentRanges = this.syntaxProvider.getMultilineComments();
 
     for (const range of commentRanges) {
       result.push({

server/src/sas/SyntaxProvider.ts

@@ -29,6 +29,8 @@ export class SyntaxProvider {
   private tailUnchangedSyntaxTable: SyntaxToken[][] = [];
   private removedSyntaxTable: SyntaxToken[][] = [];
   private _tokenCallback: ((token: Token) => void) | undefined;
+  private _multilineComments: Array<{ startLine: number; endLine: number }> =
+    [];
 
   public blockComment = { start: "/*", end: "*/" };
   public lexer;
@@ -45,6 +47,7 @@ export class SyntaxProvider {
     let startLine = 0;
     this.currTokenIndex = 0;
     this.lastToken = null;
+    this._multilineComments = []; // Clear previous comments
     this.parsingState = 1; //LanguageService.ParsingState.STARTING;
     this.parsedRange = this.lexer.start(change);
 
@@ -345,6 +348,17 @@ export class SyntaxProvider {
         end: { line: token.end.line, column: token.end.column },
       });
     }
+
+    // Collect multiline comments as they're processed by the lexer
+    if (
+      (token.type === "comment" || token.type === "macro-comment") &&
+      token.end.line > token.start.line
+    ) {
+      this._multilineComments.push({
+        startLine: token.start.line,
+        endLine: token.end.line,
+      });
+    }
   }
 
   // public functions
@@ -426,55 +440,7 @@ export class SyntaxProvider {
   getTokenBlocks(): any[] {
     return this.lexer.tknBlks || [];
   }
-  getAllCommentRanges(): Array<{ startLine: number; endLine: number }> {
-    const ranges: Array<{ startLine: number; endLine: number }> = [];
-    const lineCount = this.model.getLineCount();
-
-    // First, get comment blocks from lexer token blocks
-    const tokenBlocks = this.getTokenBlocks();
-    for (const block of tokenBlocks) {
-      const isComment = block.blockComment === true || block.type === "comment";
-      if (isComment && block.endLine > block.startLine) {
-        ranges.push({
-          startLine: block.startLine,
-          endLine: block.endLine,
-        });
-      }
-    }
-
-    // Then scan for any multiline comment tokens that might not be in token blocks
-    for (let lineNum = 0; lineNum < lineCount; lineNum++) {
-      const tokens = this.getSyntax(lineNum);
-
-      for (const token of tokens) {
-        if (token.style === "comment" || token.style === "macro-comment") {
-          // Check if this token indicates a multiline span
-          if (
-            token.state &&
-            typeof token.state === "object" &&
-            token.state.line !== undefined
-          ) {
-            const endLine = token.state.line;
-            if (endLine > lineNum) {
-              // Check if this range is already covered
-              const isAlreadyCovered = ranges.some(
-                (range) =>
-                  range.startLine === lineNum && range.endLine === endLine,
-              );
-
-              if (!isAlreadyCovered) {
-                ranges.push({
-                  startLine: lineNum,
-                  endLine: endLine,
-                });
-              }
-            }
-          }
-          break; // Only need first comment token per line
-        }
-      }
-    }
-
-    return ranges;
+  getMultilineComments(): Array<{ startLine: number; endLine: number }> {
+    return this._multilineComments;
   }
 }