Files
zitadel/backend/v3/storage/database/errors.go
Silvan cccfc816f6 refactor: database interaction and error handling (#10762)
This pull request introduces a significant refactoring of the database
interaction layer, focusing on improving explicitness, transactional
control, and error handling. The core change is the removal of the
stateful `QueryExecutor` from repository instances. Instead, it is now
passed as an argument to each method that interacts with the database.

This change makes transaction management more explicit and flexible, as
the same repository instance can be used with a database pool or a
specific transaction without needing to be re-instantiated.

### Key Changes

- **Explicit `QueryExecutor` Passing:**
- All repository methods (`Get`, `List`, `Create`, `Update`, `Delete`,
etc.) in `InstanceRepository`, `OrganizationRepository`,
`UserRepository`, and their sub-repositories now require a
`database.QueryExecutor` (e.g., a `*pgxpool.Pool` or `pgx.Tx`) as the
first argument.
- Repository constructors no longer accept a `QueryExecutor`. For
example, `repository.InstanceRepository(pool)` is now
`repository.InstanceRepository()`.

- **Enhanced Error Handling:**
- A new `database.MissingConditionError` is introduced to enforce
required query conditions, such as ensuring an `instance_id` is always
present in `UPDATE` and `DELETE` operations.
- The database error wrapper in the `postgres` package now correctly
identifies and wraps `pgx.ErrTooManyRows` and similar errors from the
`scany` library into a `database.MultipleRowsFoundError`.

- **Improved Database Conditions:**
- The `database.Condition` interface now includes a
`ContainsColumn(Column) bool` method. This allows for runtime checks to
ensure that critical filters (like `instance_id`) are included in a
query, preventing accidental cross-tenant data modification.
- A new `database.Exists()` condition has been added to support `EXISTS`
subqueries, enabling more complex filtering logic, such as finding an
organization that has a specific domain.

- **Repository and Interface Refactoring:**
- The method for loading related entities (e.g., domains for an
organization) has been changed from a boolean flag (`Domains(true)`) to
a more explicit, chainable method (`LoadDomains()`). This returns a new
repository instance configured to load the sub-resource, promoting
immutability.
- The custom `OrgIdentifierCondition` has been removed in favor of using
the standard `database.Condition` interface, simplifying the API.

- **Code Cleanup and Test Updates:**
  - Unnecessary struct embeddings and metadata have been removed.
- All integration and repository tests have been updated to reflect the
new method signatures, passing the database pool or transaction object
explicitly.
- New tests have been added to cover the new `ExistsDomain`
functionality and other enhancements.

These changes make the data access layer more robust, predictable, and
easier to work with, especially in the context of database transactions.
2025-09-24 10:12:31 +00:00

274 lines
6.8 KiB
Go

package database
import (
"errors"
"fmt"
)
var (
ErrNoChanges = errors.New("update must contain a change")
)
type MissingConditionError struct {
col Column
}
func NewMissingConditionError(col Column) error {
return &MissingConditionError{
col: col,
}
}
func (e *MissingConditionError) Error() string {
var builder StatementBuilder
builder.WriteString("missing condition for column")
if e.col != nil {
builder.WriteString(" on ")
e.col.WriteQualified(&builder)
}
return builder.String()
}
func (e *MissingConditionError) Is(target error) bool {
_, ok := target.(*MissingConditionError)
return ok
}
// NoRowFoundError is returned when QueryRow does not find any row.
// It wraps the dialect specific original error to provide more context.
type NoRowFoundError struct {
original error
}
func NewNoRowFoundError(original error) error {
return &NoRowFoundError{
original: original,
}
}
func (e *NoRowFoundError) Error() string {
if e.original != nil {
return fmt.Sprintf("no row found: %v", e.original)
}
return "no row found"
}
func (e *NoRowFoundError) Is(target error) bool {
_, ok := target.(*NoRowFoundError)
return ok
}
func (e *NoRowFoundError) Unwrap() error {
return e.original
}
// MultipleRowsFoundError is returned when QueryRow finds multiple rows.
// It wraps the dialect specific original error to provide more context.
type MultipleRowsFoundError struct {
original error
}
func NewMultipleRowsFoundError(original error) error {
return &MultipleRowsFoundError{
original: original,
}
}
func (e *MultipleRowsFoundError) Error() string {
if e.original != nil {
return fmt.Sprintf("multiple rows found: %v", e.original)
}
return "multiple rows found"
}
func (e *MultipleRowsFoundError) Is(target error) bool {
_, ok := target.(*MultipleRowsFoundError)
return ok
}
func (e *MultipleRowsFoundError) Unwrap() error {
return e.original
}
type IntegrityType string
const (
IntegrityTypeCheck IntegrityType = "check"
IntegrityTypeUnique IntegrityType = "unique"
IntegrityTypeForeign IntegrityType = "foreign"
IntegrityTypeNotNull IntegrityType = "not null"
)
// IntegrityViolationError represents a generic integrity violation error.
// It wraps the dialect specific original error to provide more context.
type IntegrityViolationError struct {
integrityType IntegrityType
table string
constraint string
original error
}
func newIntegrityViolationError(typ IntegrityType, table, constraint string, original error) IntegrityViolationError {
return IntegrityViolationError{
integrityType: typ,
table: table,
constraint: constraint,
original: original,
}
}
func (e *IntegrityViolationError) Error() string {
if e.original != nil {
return fmt.Sprintf("integrity violation of type %q on %q (constraint: %q): %v", e.integrityType, e.table, e.constraint, e.original)
}
return fmt.Sprintf("integrity violation of type %q on %q (constraint: %q)", e.integrityType, e.table, e.constraint)
}
func (e *IntegrityViolationError) Is(target error) bool {
_, ok := target.(*IntegrityViolationError)
return ok
}
func (e *IntegrityViolationError) Unwrap() error {
return e.original
}
// CheckError is returned when a check constraint fails.
// It wraps the [IntegrityViolationError] to provide more context.
// It is used to indicate that a check constraint was violated during an insert or update operation.
type CheckError struct {
IntegrityViolationError
}
func NewCheckError(table, constraint string, original error) error {
return &CheckError{
IntegrityViolationError: newIntegrityViolationError(IntegrityTypeCheck, table, constraint, original),
}
}
func (e *CheckError) Is(target error) bool {
_, ok := target.(*CheckError)
return ok
}
func (e *CheckError) Unwrap() error {
return &e.IntegrityViolationError
}
// UniqueError is returned when a unique constraint fails.
// It wraps the [IntegrityViolationError] to provide more context.
// It is used to indicate that a unique constraint was violated during an insert or update operation.
type UniqueError struct {
IntegrityViolationError
}
func NewUniqueError(table, constraint string, original error) error {
return &UniqueError{
IntegrityViolationError: newIntegrityViolationError(IntegrityTypeUnique, table, constraint, original),
}
}
func (e *UniqueError) Is(target error) bool {
_, ok := target.(*UniqueError)
return ok
}
func (e *UniqueError) Unwrap() error {
return &e.IntegrityViolationError
}
// ForeignKeyError is returned when a foreign key constraint fails.
// It wraps the [IntegrityViolationError] to provide more context.
// It is used to indicate that a foreign key constraint was violated during an insert or update operation
type ForeignKeyError struct {
IntegrityViolationError
}
func NewForeignKeyError(table, constraint string, original error) error {
return &ForeignKeyError{
IntegrityViolationError: newIntegrityViolationError(IntegrityTypeForeign, table, constraint, original),
}
}
func (e *ForeignKeyError) Is(target error) bool {
_, ok := target.(*ForeignKeyError)
return ok
}
func (e *ForeignKeyError) Unwrap() error {
return &e.IntegrityViolationError
}
// NotNullError is returned when a not null constraint fails.
// It wraps the [IntegrityViolationError] to provide more context.
// It is used to indicate that a not null constraint was violated during an insert or update operation.
type NotNullError struct {
IntegrityViolationError
}
func NewNotNullError(table, constraint string, original error) error {
return &NotNullError{
IntegrityViolationError: newIntegrityViolationError(IntegrityTypeNotNull, table, constraint, original),
}
}
func (e *NotNullError) Is(target error) bool {
_, ok := target.(*NotNullError)
return ok
}
func (e *NotNullError) Unwrap() error {
return &e.IntegrityViolationError
}
// ScanError is returned when scanning rows into objects failed.
// It wraps the original error to provide more context.
type ScanError struct {
original error
}
func NewScanError(original error) error {
return &ScanError{
original: original,
}
}
func (e *ScanError) Error() string {
return fmt.Sprintf("Scan error: %v", e.original)
}
func (e *ScanError) Is(target error) bool {
_, ok := target.(*ScanError)
return ok
}
func (e *ScanError) Unwrap() error {
return e.original
}
// UnknownError is returned when an unknown error occurs.
// It wraps the dialect specific original error to provide more context.
// It is used to indicate that an error occurred that does not fit into any of the other categories.
type UnknownError struct {
original error
}
func NewUnknownError(original error) error {
return &UnknownError{
original: original,
}
}
func (e *UnknownError) Error() string {
return fmt.Sprintf("unknown database error: %v", e.original)
}
func (e *UnknownError) Is(target error) bool {
_, ok := target.(*UnknownError)
return ok
}
func (e *UnknownError) Unwrap() error {
return e.original
}