- Walk original text across node boundaries while tracking normalized positions
- Identify exact start/end nodes and offsets for the match
- Compute combined indices from node spans to create accurate DOM Range
- Eliminates artifacts like 'We b' by preventing whitespace from splitting words
- Keeps strict bounds checks and graceful failures
- Strip all existing mark elements from HTML before re-highlighting
- Prevents old broken highlights from persisting in the DOM
- Ensures clean text is used as the base for new highlight application
- Fixes 'We b' spacing issue caused by corrupted marks from previous buggy renders
- Remove debug logging now that position mapping is working correctly
- Log search text, match indices, and extracted text during position mapping
- Show sample of combined text around the extracted range
- Help identify where position mapping is going wrong for 'We b' issue
- Build explicit position map array from normalized to original text indices
- Properly handle whitespace sequences in position mapping
- Ensure each normalized character position maps to correct original position
- Validate mapped positions are within bounds before using
- Fixes spacing issues like 'We b' appearing instead of 'Web' in highlights
- Validate extracted range text matches search text before highlighting
- Check single-node matches are not empty or whitespace-only
- Compare both exact and normalized text to handle whitespace variations
- Prevent broken/corrupted highlights from being applied to DOM
- Add detailed logging for validation failures to aid debugging
- Implement proper normalized-to-original text position mapping
- Add comprehensive validation for range indices and node offsets
- Verify range is not collapsed before extracting content
- Add try-catch block to handle DOM manipulation errors gracefully
- Add detailed warning logs for debugging failed highlight matches
- Prevent invalid ranges from corrupting the DOM structure
- Fix broken text nodes and visual artifacts in highlighted content
- Use DOM Range API to extract and wrap content in a single mark element
- Preserves internal DOM structure (links, formatting) within the highlight
- Eliminates visual breaks between multiple mark elements
- Ensures highlight appears as one continuous selection even across inline elements
- Add tryMultiNodeMatch function to find text spanning multiple DOM nodes
- Build combined text from all text nodes for comprehensive matching
- Handle highlighting across node boundaries with proper offsets
- Falls back to multi-node matching when single-node match fails
- Fixes issue where selections with inline formatting couldn't be matched
- Create textMatching module for text search utilities
- Create domUtils module for DOM manipulation helpers
- Create htmlMatching module for HTML highlight application
- Reduce highlightMatching.tsx from 217 lines to 59 lines
- All files now under 210 lines
