Function extractFrontmatterAndBodyText

pkg/parser/frontmatter_hash.go:253–289

extractFrontmatterAndBodyText extracts frontmatter as raw text without parsing YAML Returns: frontmatterText, markdownBody, error

(content string)

// extractFrontmatterAndBodyText extracts frontmatter as raw text without parsing YAML
// Returns: frontmatterText, markdownBody, error
func extractFrontmatterAndBodyText(content string) (string, string, error) {
	// Normalize CRLF to LF so that files with Windows line-endings produce the
	// same frontmatter text (and therefore the same hash) as equivalent LF files.
	content = strings.ReplaceAll(content, "\r\n", "\n")

	lines := strings.Split(content, "\n")

	// Check if content starts with frontmatter delimiter
	if len(lines) == 0 || strings.TrimSpace(lines[0]) != "---" {
		// No frontmatter
		return "", content, nil
	}

	// Find end of frontmatter
	endIndex := -1
	for i := 1; i < len(lines); i++ {
		if strings.TrimSpace(lines[i]) == "---" {
			endIndex = i
			break
		}
	}

	if endIndex == -1 {
		return "", "", errors.New("frontmatter not properly closed")
	}

	// Extract frontmatter text (lines between --- delimiters)
	frontmatterText := strings.Join(lines[1:endIndex], "\n")

	// Extract markdown body (everything after closing ---)
	var markdown string
	if endIndex+1 < len(lines) {
		markdown = strings.Join(lines[endIndex+1:], "\n")
	}

	return frontmatterText, markdown, nil
}

// normalizeFrontmatterText normalizes frontmatter text for consistent hashing
// Removes leading/trailing whitespace and normalizes line endings