Removed svg from the binary file filtering and fixed the rename/copy status parsing

Author: vinyas-bharadwaj

internal/git/operations.go

@@ -22,63 +22,106 @@ func IsRepository(path string) bool {
 	return strings.TrimSpace(string(output)) == "true"
 }
 
-// filterBinaryFiles filters out binary files from git diff --name-status output
-func filterBinaryFiles(nameStatusOutput string) string {
-	if nameStatusOutput == "" {
-		return ""
+// parseGitStatusLine represents a parsed git status line
+type parseGitStatusLine struct {
+	status    string
+	filenames []string
+}
+
+// parseGitNameStatus parses a single line from git diff --name-status output
+// Handles various git status codes including rename (R) and copy (C) operations
+func parseGitNameStatus(line string) parseGitStatusLine {
+	if line == "" {
+		return parseGitStatusLine{}
 	}
 
-	lines := strings.Split(strings.TrimSpace(nameStatusOutput), "\n")
-	var filteredLines []string
+	parts := strings.Fields(line)
+	if len(parts) < 2 {
+		return parseGitStatusLine{}
+	}
 
-	for _, line := range lines {
-		if line == "" {
-			continue
-		}
-		
-		// Parse git diff --name-status format (e.g., "M filename" or "A filename")
-		parts := strings.Fields(line)
-		if len(parts) >= 2 {
-			filename := strings.Join(parts[1:], " ") // Handle filenames with spaces
-			if !utils.IsBinaryFile(filename) {
-				filteredLines = append(filteredLines, line)
+	status := parts[0]
+	
+	// Handle rename/copy status codes (e.g., "R100", "C75")
+	if len(status) > 1 && (status[0] == 'R' || status[0] == 'C') {
+		// For rename/copy, we expect: "R100 oldname newname" or "C75 oldname newname"
+		if len(parts) >= 3 {
+			// For renames/copies, both old and new filenames need to be checked
+			oldFile := parts[1]
+			newFile := strings.Join(parts[2:], " ") // Handle spaces in new filename
+			return parseGitStatusLine{
+				status:    status,
+				filenames: []string{oldFile, newFile},
 			}
 		}
 	}
 
-	if len(filteredLines) == 0 {
-		return ""
+	// Handle regular status codes (M, A, D, etc.)
+	filename := strings.Join(parts[1:], " ") // Handle filenames with spaces
+	return parseGitStatusLine{
+		status:    status,
+		filenames: []string{filename},
 	}
-	
-	return strings.Join(filteredLines, "\n")
 }
 
-// extractNonBinaryFiles extracts non-binary filenames from git diff --name-status output
-func extractNonBinaryFiles(nameStatusOutput string) []string {
+// processGitStatusOutput processes git diff --name-status output and returns filtered results
+func processGitStatusOutput(nameStatusOutput string, returnFilenames bool) ([]string, []string) {
 	if nameStatusOutput == "" {
-		return nil
+		return nil, nil
 	}
 
 	lines := strings.Split(strings.TrimSpace(nameStatusOutput), "\n")
+	var filteredLines []string
 	var nonBinaryFiles []string
 
 	for _, line := range lines {
 		if line == "" {
 			continue
 		}
 
-		// Parse git diff --name-status format (e.g., "M filename" or "A filename")
-		parts := strings.Fields(line)
-		if len(parts) >= 2 {
-			filename := strings.Join(parts[1:], " ") // Handle filenames with spaces
-			if !utils.IsBinaryFile(filename) {
-				nonBinaryFiles = append(nonBinaryFiles, filename)
+		parsed := parseGitNameStatus(line)
+		if len(parsed.filenames) == 0 {
+			continue
+		}
+		
+		// Check if any of the filenames are binary
+		hasBinaryFile := false
+		for _, filename := range parsed.filenames {
+			if utils.IsBinaryFile(filename) {
+				hasBinaryFile = true
+				break
+			}
+		}
+		
+		// If no binary files found, include this line/files
+		if !hasBinaryFile {
+			filteredLines = append(filteredLines, line)
+			if returnFilenames {
+				nonBinaryFiles = append(nonBinaryFiles, parsed.filenames...)
 			}
 		}
 	}
 
+	return filteredLines, nonBinaryFiles
+}
+
+// filterBinaryFiles filters out binary files from git diff --name-status output
+func filterBinaryFiles(nameStatusOutput string) string {
+	filteredLines, _ := processGitStatusOutput(nameStatusOutput, false)
+	
+	if len(filteredLines) == 0 {
+		return ""
+	}
+	
+	return strings.Join(filteredLines, "\n")
+}
+
+// extractNonBinaryFiles extracts non-binary filenames from git diff --name-status output
+func extractNonBinaryFiles(nameStatusOutput string) []string {
+	_, nonBinaryFiles := processGitStatusOutput(nameStatusOutput, true)
 	return nonBinaryFiles
 }
+}
 
 // GetChanges retrieves all Git changes including staged, unstaged, and untracked files
 func GetChanges(config *types.RepoConfig) (string, error) {

internal/utils/utils.go

@@ -24,7 +24,7 @@ func IsTextFile(filename string) bool {
 		".ts", ".tsx", ".jsx", ".php", ".rb", ".rs", ".dart", ".sql", ".r",
 		".scala", ".kt", ".swift", ".m", ".pl", ".lua", ".vim", ".csv", 
 		".log", ".cfg", ".conf", ".ini", ".toml", ".lock", ".gitignore",
-		".dockerfile", ".makefile", ".cmake", ".pro", ".pri",
+		".dockerfile", ".makefile", ".cmake", ".pro", ".pri", ".svg",
 	}
 
 	ext := strings.ToLower(filepath.Ext(filename))
@@ -34,15 +34,21 @@ func IsTextFile(filename string) bool {
 		}
 	}
 
+	// Files without extensions (like README, Dockerfile, Makefile) are treated as text
+	// This ensures consistent behavior for common configuration and documentation files
+	if ext == "" {
+		return true
+	}
+
 	return false
 }
 
 // IsBinaryFile checks if a file is likely to be a binary file that should be excluded from diffs
 func IsBinaryFile(filename string) bool {
 	// List of common binary file extensions
 	binaryExtensions := []string{
-		// Images
-		".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".tif", ".svg", ".ico", ".webp",
+		// Images (excluding SVG which is XML text)
+		".jpg", ".jpeg", ".png", ".gif", ".bmp", ".tiff", ".tif", ".ico", ".webp",
 		// Audio/Video
 		".mp3", ".mp4", ".avi", ".mkv", ".mov", ".wmv", ".flv", ".wav", ".ogg", ".m4a",
 		// Archives/Compressed
@@ -65,6 +71,8 @@ func IsBinaryFile(filename string) bool {
 		}
 	}
 
+	// Note: Files with unknown extensions are not considered binary by default
+	// This allows them to be processed as text files for diff analysis
 	return false
 }