[Bug]: Nil pointer dereference in Arabic translations

[blacksud0]

What happened?

When using Arabic translations with the validator, a nil pointer dereference occurs in the translator. Line 177 (ar.go:177) C method during a struct validation especially (min or max) tags.

Error Stack Trace

runtime error: invalid memory address or nil pointer dereference
goroutine 16 [running]:
github.com/go-playground/universal-translator.(translator).C(0xc0004c0870, {0xc18a20, 0x11434d0}, 0x4020000000000000, 0x0, {0x1165f80, 0x1})
/home/blacksudo/go/pkg/mod/github.com/go-playground/universal-translator@v0.18.1/translator.go:335 +0xdf
github.com/go-playground/validator/v10/translations/ar.RegisterDefaultTranslations.func4({0x115a980, 0xc0004c0870}, {0x1150cb0, 0xc00044e510})

go/pkg/mod/github.com/go-playground/validator/v10@v10.15.4/translations/ar/ar.go:177 +0x3ac

Environment & Version

• Go version: 1.24
• echo web framework v4.13.3
• validator version: v10.15.4
• universal-translator version: v0.18.1

Steps to Reproduce

1. Initialize a validator with Arabic language support
2. Register default Arabic translations
3. Attempt to validate a struct with validation tags eg. (min, max)

Example Code

Login struct {
		Username string `json:"username" validate:"required,min=3,max=50"`
		Password string `json:"password" validate:"required,min=8"`
	}

[nodivbyzero]

I can confirm that the Arabic translation caused panic.

I was able to resolve it by adding ut.AddCardinal("len-string-character", "{0} حرف", locales.PluralRuleTwo, false); to the Arabic translation file, but I'm unsure if the translation is accurate.

@blacksud0 can you verify it?

[blacksud0]

Hi @nodivbyzero

My Solution to Arabic Translation Panic

The panic was caused by using the cardinal number system (ut.AddCardinal() and ut.C()) which wasn't properly initialized for Arabic translations. Here's how to fix it by using simple singular/plural translations instead:

The Fix

Replace the cardinal number system with simple singular/plural translations. Here's an example for the "len" validation tag:

// Before (problematic code)
customRegisFunc: func(ut ut.Translator) (err error) {
if err = ut.Add("len-string", "يجب أن يكون طول {0} مساويا ل {1}", false); err != nil {
return
}
if err = ut.AddCardinal("len-string-character", "{0} حرف", locales.PluralRuleOne, false); err != nil {
return
}
if err = ut.AddCardinal("len-string-character", "{0} أحرف", locales.PluralRuleOther, false); err != nil {
return
}
// ... more AddCardinal calls ...
}


// After (fixed code)
customRegisFunc: func(ut ut.Translator) (err error) {
if err = ut.Add("len-string", "يجب أن يكون طول {0} مساويا ل {1}", false); err != nil {
return
}
if err = ut.Add("len-string-character-one", "{0} حرف", false); err != nil {
return
}
if err = ut.Add("len-string-character-other", "{0} أحرف", false); err != nil {
return
}
// ... more simple Add calls ...
}

And in the translation function:

// Before (problematic code)
c, err = ut.C("len-string-character", f64, digits, ut.FmtNumber(f64, digits))

// After (fixed code)
if f64 == 1 {
c, err = ut.T("len-string-character-one", ut.FmtNumber(f64, digits))
} else {
c, err = ut.T("len-string-character-other", ut.FmtNumber(f64, digits))
}

Why This Works

1. Avoids the nil pointer dereference by not using the cardinal number system
2. Still maintains proper Arabic plural forms using simple singular/plural logic
3. More reliable as it doesn't depend on complex plural rules initialization

Implementation Notes

• Apply this pattern to all validation tags that use numbers (len, min, max, lt, lte, etc.)
• Use -one suffix for singular form and -other suffix for plural form
• Keep the same Arabic translations, just change how they're registered and accessed

The complete solution can be found in this gist ar.go.

[blacksud0]

Testing the implementation

Code:

Result:

[nodivbyzero]

Hey @blacksud0

Could you please create a new PR with the changes you’re suggesting?