fix: Fixes and performance improvements in utils package (#331)

.github/workflows/e2e-tests.yml

@@ -33,6 +33,29 @@ jobs:
           name: docker-image
           path: /tmp/docker-image.tar
 
+  test-backend:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+      - uses: actions/setup-go@v5
+        with:
+          go-version-file: 'backend/go.mod'
+          cache-dependency-path: 'backend/go.sum'
+      - name: Install dependencies
+        working-directory: backend
+        run: |
+          go get ./...
+      - name: Run backend unit tests
+        working-directory: backend
+        run: |
+          go test -v ./... | tee /tmp/TestResults.log
+      - uses: actions/upload-artifact@v4
+        if: always()
+        with:
+          name: backend-unit-tests
+          path: /tmp/TestResults.log
+          retention-days: 15
+
   test-sqlite:
     runs-on: ubuntu-latest
     needs: build

backend/internal/utils/file_util.go

@@ -5,14 +5,16 @@ import (
 	"mime/multipart"
 	"os"
 	"path/filepath"
-	"strings"
 
 	"github.com/pocket-id/pocket-id/backend/resources"
 )
 
 func GetFileExtension(filename string) string {
-	splitted := strings.Split(filename, ".")
-	return splitted[len(splitted)-1]
+	ext := filepath.Ext(filename)
+	if len(ext) > 0 && ext[0] == '.' {
+		return ext[1:]
+	}
+	return filename
 }
 
 func GetImageMimeType(ext string) string {

backend/internal/utils/file_util_test.go

@@ -0,0 +1,73 @@
+package utils
+
+import (
+	"testing"
+)
+
+func TestGetFileExtension(t *testing.T) {
+	tests := []struct {
+		name     string
+		filename string
+		want     string
+	}{
+		{
+			name:     "Simple file with extension",
+			filename: "document.pdf",
+			want:     "pdf",
+		},
+		{
+			name:     "File with path",
+			filename: "/path/to/document.txt",
+			want:     "txt",
+		},
+		{
+			name:     "File with path (Windows style)",
+			filename: "C:\\path\\to\\document.jpg",
+			want:     "jpg",
+		},
+		{
+			name:     "Multiple extensions",
+			filename: "archive.tar.gz",
+			want:     "gz",
+		},
+		{
+			name:     "Hidden file with extension",
+			filename: ".config.json",
+			want:     "json",
+		},
+		{
+			name:     "Filename with dots",
+			filename: "version.1.2.3.txt",
+			want:     "txt",
+		},
+		{
+			name:     "File with uppercase extension",
+			filename: "image.JPG",
+			want:     "JPG",
+		},
+		{
+			name:     "File without extension",
+			filename: "README",
+			want:     "README",
+		},
+		{
+			name:     "Hidden file without extension",
+			filename: ".gitignore",
+			want:     "gitignore",
+		},
+		{
+			name:     "Empty filename",
+			filename: "",
+			want:     "",
+		},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			got := GetFileExtension(tt.filename)
+			if got != tt.want {
+				t.Errorf("GetFileExtension(%q) = %q, want %q", tt.filename, got, tt.want)
+			}
+		})
+	}
+}

backend/internal/utils/image/profile_picture.go

@@ -90,7 +90,7 @@ func CreateDefaultProfilePicture(firstName, lastName string) (*bytes.Buffer, err
 	var buf bytes.Buffer
 	err = imaging.Encode(&buf, img, imaging.PNG)
 	if err != nil {
-		return nil, fmt.Errorf("failed to encode image: %v", err)
+		return nil, fmt.Errorf("failed to encode image: %w", err)
 	}
 
 	return &buf, nil

backend/internal/utils/string_util.go

@@ -2,8 +2,9 @@ package utils
 
 import (
 	"crypto/rand"
+	"errors"
 	"fmt"
-	"math/big"
+	"io"
 	"net/url"
 	"regexp"
 	"strings"
@@ -13,23 +14,41 @@ import (
 // GenerateRandomAlphanumericString generates a random alphanumeric string of the given length
 func GenerateRandomAlphanumericString(length int) (string, error) {
 	const charset = "abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ0123456789"
-	const charsetLength = int64(len(charset))
 
 	if length <= 0 {
-		return "", fmt.Errorf("length must be a positive integer")
+		return "", errors.New("length must be a positive integer")
 	}
 
-	result := make([]byte, length)
+	// The algorithm below is adapted from https://stackoverflow.com/a/35615565
+	const (
+		letterIdxBits = 6                    // 6 bits to represent a letter index
+		letterIdxMask = 1<<letterIdxBits - 1 // All 1-bits, as many as letterIdxBits
+	)
 
-	for i := range result {
-		num, err := rand.Int(rand.Reader, big.NewInt(charsetLength))
-		if err != nil {
-			return "", err
+	result := strings.Builder{}
+	result.Grow(length)
+	// Because we discard a bunch of bytes, we read more in the buffer to minimize the changes of performing additional IO
+	bufferSize := int(float64(length) * 1.3)
+	randomBytes := make([]byte, bufferSize)
+	for i, j := 0, 0; i < length; j++ {
+		// Fill the buffer if needed
+		if j%bufferSize == 0 {
+			_, err := io.ReadFull(rand.Reader, randomBytes)
+			if err != nil {
+				return "", fmt.Errorf("failed to generate random bytes: %w", err)
+			}
+		}
+
+		// Discard bytes that are outside of the range
+		// This allows making sure that we maintain uniform distribution
+		idx := int(randomBytes[j%length] & letterIdxMask)
+		if idx < len(charset) {
+			result.WriteByte(charset[idx])
+			i++
 		}
-		result[i] = charset[num.Int64()]
 	}
 
-	return string(result), nil
+	return result.String(), nil
 }
 
 func GetHostnameFromURL(rawURL string) string {
@@ -45,30 +64,40 @@ func StringPointer(s string) *string {
 	return &s
 }
 
-func CapitalizeFirstLetter(s string) string {
-	if s == "" {
-		return s
+func CapitalizeFirstLetter(str string) string {
+	if str == "" {
+		return ""
 	}
-	runes := []rune(s)
-	runes[0] = unicode.ToUpper(runes[0])
-	return string(runes)
+
+	result := strings.Builder{}
+	result.Grow(len(str))
+	for i, r := range str {
+		if i == 0 {
+			result.WriteRune(unicode.ToUpper(r))
+		} else {
+			result.WriteRune(r)
+		}
+	}
+	return result.String()
 }
 
-func CamelCaseToSnakeCase(s string) string {
-	var result []rune
-	for i, r := range s {
+func CamelCaseToSnakeCase(str string) string {
+	result := strings.Builder{}
+	result.Grow(int(float32(len(str)) * 1.1))
+	for i, r := range str {
 		if unicode.IsUpper(r) && i > 0 {
-			result = append(result, '_')
+			result.WriteByte('_')
 		}
-		result = append(result, unicode.ToLower(r))
+		result.WriteRune(unicode.ToLower(r))
 	}
-	return string(result)
+	return result.String()
 }
 
+var camelCaseToScreamingSnakeCaseRe = regexp.MustCompile(`([a-z0-9])([A-Z])`)
+
 func CamelCaseToScreamingSnakeCase(s string) string {
 	// Insert underscores before uppercase letters (except the first one)
-	re := regexp.MustCompile(`([a-z0-9])([A-Z])`)
-	snake := re.ReplaceAllString(s, `${1}_${2}`)
+	snake := camelCaseToScreamingSnakeCaseRe.ReplaceAllString(s, `${1}_${2}`)
 
 	// Convert to uppercase
 	return strings.ToUpper(snake)

backend/internal/utils/string_util_test.go

@@ -0,0 +1,105 @@
+package utils
+
+import (
+	"regexp"
+	"testing"
+)
+
+func TestGenerateRandomAlphanumericString(t *testing.T) {
+	t.Run("valid length returns correct string", func(t *testing.T) {
+		const length = 10
+		str, err := GenerateRandomAlphanumericString(length)
+
+		if err != nil {
+			t.Errorf("Expected no error, got %v", err)
+		}
+		if len(str) != length {
+			t.Errorf("Expected length %d, got %d", length, len(str))
+		}
+
+		matched, err := regexp.MatchString(`^[a-zA-Z0-9]+$`, str)
+		if err != nil {
+			t.Errorf("Regex match failed: %v", err)
+		}
+		if !matched {
+			t.Errorf("String contains non-alphanumeric characters: %s", str)
+		}
+	})
+
+	t.Run("zero length returns error", func(t *testing.T) {
+		_, err := GenerateRandomAlphanumericString(0)
+		if err == nil {
+			t.Error("Expected error for zero length, got nil")
+		}
+	})
+
+	t.Run("negative length returns error", func(t *testing.T) {
+		_, err := GenerateRandomAlphanumericString(-1)
+		if err == nil {
+			t.Error("Expected error for negative length, got nil")
+		}
+	})
+
+	t.Run("generates different strings", func(t *testing.T) {
+		str1, _ := GenerateRandomAlphanumericString(10)
+		str2, _ := GenerateRandomAlphanumericString(10)
+		if str1 == str2 {
+			t.Error("Generated strings should be different")
+		}
+	})
+}
+
+func TestCapitalizeFirstLetter(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"empty string", "", ""},
+		{"lowercase first letter", "hello", "Hello"},
+		{"already capitalized", "Hello", "Hello"},
+		{"single lowercase letter", "h", "H"},
+		{"single uppercase letter", "H", "H"},
+		{"starts with number", "123abc", "123abc"},
+		{"unicode character", "étoile", "Étoile"},
+		{"special character", "_test", "_test"},
+		{"multi-word", "hello world", "Hello world"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := CapitalizeFirstLetter(tt.input)
+			if result != tt.expected {
+				t.Errorf("CapitalizeFirstLetter(%q) = %q, want %q", tt.input, result, tt.expected)
+			}
+		})
+	}
+}
+
+func TestCamelCaseToSnakeCase(t *testing.T) {
+	tests := []struct {
+		name     string
+		input    string
+		expected string
+	}{
+		{"empty string", "", ""},
+		{"simple camelCase", "camelCase", "camel_case"},
+		{"PascalCase", "PascalCase", "pascal_case"},
+		{"multipleWordsInCamelCase", "multipleWordsInCamelCase", "multiple_words_in_camel_case"},
+		{"consecutive uppercase", "HTTPRequest", "h_t_t_p_request"},
+		{"single lowercase word", "word", "word"},
+		{"single uppercase word", "WORD", "w_o_r_d"},
+		{"with numbers", "camel123Case", "camel123_case"},
+		{"with numbers in middle", "model2Name", "model2_name"},
+		{"mixed case", "iPhone6sPlus", "i_phone6s_plus"},
+	}
+
+	for _, tt := range tests {
+		t.Run(tt.name, func(t *testing.T) {
+			result := CamelCaseToSnakeCase(tt.input)
+			if result != tt.expected {
+				t.Errorf("CamelCaseToSnakeCase(%q) = %q, want %q", tt.input, result, tt.expected)
+			}
+		})
+	}
+}