geektutu/7days-golang · error
invalid sql type %s (%s)
Error message
invalid sql type %s (%s)
What it means
Identical to error 103 but in the day3-save-query stage: the sqlite3 dialect panics with 'invalid sql type' when Schema.Parse encounters a field type it cannot map to a SQLite column type. Only basic kinds and time.Time are supported; all other kinds fail fast.
Source
Thrown at gee-orm/day3-save-query/dialect/sqlite3.go:38
case reflect.Bool:
return "bool"
case reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32,
reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uintptr:
return "integer"
case reflect.Int64, reflect.Uint64:
return "bigint"
case reflect.Float32, reflect.Float64:
return "real"
case reflect.String:
return "text"
case reflect.Array, reflect.Slice:
return "blob"
case reflect.Struct:
if _, ok := typ.Interface().(time.Time); ok {
return "datetime"
}
}
panic(fmt.Sprintf("invalid sql type %s (%s)", typ.Type().Name(), typ.Kind()))
}
// TableExistSQL returns SQL that judge whether the table exists in database
func (s *sqlite3) TableExistSQL(tableName string) (string, []interface{}) {
args := []interface{}{tableName}
return "SELECT name FROM sqlite_master WHERE type='table' and name = ?", args
}
View on GitHub (pinned to cf36443821)
Solutions
- Use a supported Go type for the field or serialize to string/[]byte
- Register a custom dialect mapping if you control the dialect code
- Remove the field from the persisted struct or move it to a related table
Example fix
// before
type Order struct {
Items []string
}
// after
type Order struct {
Items string // JSON-encoded []string
} Defensive patterns
Strategy: validation
Validate before calling
func checkSchemaTypes(v interface{}) error {
t := reflect.TypeOf(v).Elem()
for i := 0; i < t.NumField(); i++ {
k := t.Field(i).Type.Kind()
if k == reflect.Struct && t.Field(i).Type != reflect.TypeOf(time.Time{}) {
return fmt.Errorf("field %s: unsupported %s", t.Field(i).Name, t.Field(i).Type)
}
if k == reflect.Map || k == reflect.Interface || k == reflect.Ptr {
return fmt.Errorf("field %s: unsupported kind %s", t.Field(i).Name, k)
}
}
return nil
} Type guard
func isSQLiteMappable(t reflect.Type) bool {
if t == reflect.TypeOf(time.Time{}) { return true }
switch t.Kind() {
case reflect.Bool, reflect.Int, reflect.Int8, reflect.Int16, reflect.Int32, reflect.Int64,
reflect.Uint, reflect.Uint8, reflect.Uint16, reflect.Uint32, reflect.Uint64,
reflect.Float32, reflect.Float64, reflect.String, reflect.Slice:
return true
}
return false
} Try / catch
defer func() {
if r := recover(); r != nil {
if s, ok := r.(string); ok && strings.HasPrefix(s, "invalid sql type") {
err = fmt.Errorf("unsupported model field: %s", s)
} else { panic(r) }
}
}() Prevention
- Avoid map/slice-of-struct/interface fields in ORM models
- Encode complex values as JSON strings before persisting
- Add a schema sanity unit test per model
When it happens
Trigger: Calling geeorm.Open(db, &Model{...}) where Model has a field of unsupported kind (map, slice, embedded struct, custom kind) causing schema parse to call DataTypeOf with an unmappable type.
Common situations: Adding a new struct field for a feature and immediately opening a session against it; using pointer fields; storing collections directly in the row.
Related errors
- invalid sql type %s (%s)
- invalid sql type %s (%s)
- invalid sql type %s (%s)
- invalid sql type %s (%s)
- panic(p)
AI-assisted analysis of geektutu/7days-golang@cf36443821 (2026-09-03).
Data as JSON: /api/errors/1883d15c48d92830.
Report an issue: GitHub.