Documentation ¶
Overview ¶
Package topdown provides query evaluation support.
The topdown implementation is a (slightly modified) version of the standard "top-down evaluation" algorithm used in query languages such as Datalog. The main difference in the implementation is in the handling of Rego references.
At a high level, the topdown implementation consists of (1) term evaluation and (2) expression evaluation. Term evaluation involves handling expression terms in isolation whereas expression evaluation involves evaluating the terms in relation to each other.
During the term evaluation phase, the topdown implementation will evaluate references found in the expression to produce bindings for:
- Variables appearing in references
- References to Virtual Documents
- Comprehensions
Once terms have been evaluated in isolation, the overall expression can be evaluated. If the expression is simply a term (e.g., string, reference, variable, etc.) then it is compared against the boolean "false" value. If the value IS NOT "false", evaluation continues. On the other hand, if the expression is defined by a built-in operator (e.g., =, !=, min, max, etc.) then the appropriate built-in function is invoked to determine if evaluation continues and bind variables accordingly.
Index ¶
- Constants
- func Continue(t *Topdown, key, value ast.Value, iter Iterator) error
- func Eval(t *Topdown, iter Iterator) error
- func MakeInput(pairs [][2]*ast.Term) (ast.Value, error)
- func PlugExpr(expr *ast.Expr, binding Binding) *ast.Expr
- func PlugHead(head *ast.Head, binding Binding) *ast.Head
- func PlugTerm(term *ast.Term, binding Binding) *ast.Term
- func PlugValue(v ast.Value, binding func(ast.Value) ast.Value) ast.Value
- func PrettyTrace(w io.Writer, trace []*Event)
- func RegisterBuiltinFunc(name ast.Var, fun BuiltinFunc)
- func RegisterFunctionalBuiltin1(name ast.Var, fun FunctionalBuiltin1)
- func RegisterFunctionalBuiltin2(name ast.Var, fun FunctionalBuiltin2)
- func RegisterFunctionalBuiltin3(name ast.Var, fun FunctionalBuiltin3)
- func RegisterFunctionalBuiltinVoid2(name ast.Var, fun FunctionalBuiltinVoid2)
- func ResetQueryIDs()
- func ResolveRefs(v ast.Value, t *Topdown) (ast.Value, error)
- func ValueToInterface(v ast.Value, resolver Resolver) (interface{}, error)
- type Binding
- type BufferTracer
- type BuiltinEmpty
- type BuiltinFunc
- type Error
- type Event
- type FunctionalBuiltin1
- type FunctionalBuiltin2
- type FunctionalBuiltin3
- type FunctionalBuiltinVoid2
- type Iterator
- type Op
- type QueryParams
- type QueryResult
- type QueryResultSet
- type Resolver
- type Topdown
- func (t *Topdown) Bind(key ast.Value, value ast.Value, prev *Undo) *Undo
- func (t *Topdown) Binding(k ast.Value) ast.Value
- func (t *Topdown) Child(query ast.Body) *Topdown
- func (t *Topdown) Closure(query ast.Body) *Topdown
- func (t *Topdown) Current() *ast.Expr
- func (t *Topdown) Resolve(ref ast.Ref) (interface{}, error)
- func (t *Topdown) Step() *Topdown
- func (t *Topdown) Unbind(undo *Undo)
- func (t *Topdown) Vars() map[ast.Var]ast.Value
- func (t *Topdown) WithInput(input ast.Value) *Topdown
- func (t *Topdown) WithTracer(tracer Tracer) *Topdown
- type Tracer
- type Undo
- type Vars
Examples ¶
Constants ¶
const ( // InternalErr represents an unknown evaluation error. InternalErr = iota // ConflictErr indicates a conflict was encountered during evaluation. For // instance, a conflict occurs if a rule produces multiple, differing values // for the same key in an object. Conflict errors indicate the policy does // not account for the data loaded into the policy engine. ConflictErr = iota // TypeErr indicates evaluation stopped because an expression was applied to // a value of an inappropriate type. TypeErr = iota )
Variables ¶
This section is empty.
Functions ¶
func Continue ¶
Continue binds key to value in t and calls the iterator. This is a helper function for simple cases where a single value (e.g., a variable) needs to be bound to a value in order for the evaluation the proceed.
func Eval ¶
Eval evaluates the query in t and calls iter once for each set of bindings that satisfy all of the expressions in the query.
Example ¶
package main import ( "bytes" "context" "encoding/json" "fmt" "github.com/open-policy-agent/opa/ast" "github.com/open-policy-agent/opa/storage" "github.com/open-policy-agent/opa/topdown" ) func main() { // Initialize context for the example. Normally the caller would obtain the // context from an input parameter or instantiate their own. ctx := context.Background() compiler := ast.NewCompiler() // Define a dummy query and some data that the query will execute against. query, err := compiler.QueryCompiler().Compile(ast.MustParseBody(`data.a[_] = x; x >= 2`)) if err != nil { // Handle error. } var data map[string]interface{} // OPA uses Go's standard JSON library but assumes that numbers have been // decoded as json.Number instead of float64. You MUST decode with UseNumber // enabled. decoder := json.NewDecoder(bytes.NewBufferString(`{"a": [1,2,3,4]}`)) decoder.UseNumber() if err := decoder.Decode(&data); err != nil { // Handle error. } // Instantiate the policy engine's storage layer. store := storage.New(storage.InMemoryWithJSONConfig(data)) // Create a new transaction. Transactions allow the policy engine to // evaluate the query over a consistent snapshot fo the storage layer. txn, err := store.NewTransaction(ctx) if err != nil { // Handle error. } defer store.Close(ctx, txn) // Prepare the evaluation parameters. Evaluation executes against the policy // engine's storage. In this case, we seed the storage with a single array // of number. Other parameters such as the input, tracing configuration, // etc. can be set on the Topdown object. t := topdown.New(ctx, query, compiler, store, txn) result := []interface{}{} // Execute the query and provide a callbakc function to accumulate the results. err = topdown.Eval(t, func(t *topdown.Topdown) error { // Each variable in the query will have an associated "binding". x := t.Binding(ast.Var("x")) // Alternatively, you can get a mapping of all bound variables. x = t.Vars()[ast.Var("x")] // The bindings are ast.Value types so we will convert to a native Go value here. v, err := topdown.ValueToInterface(x, t) if err != nil { return err } result = append(result, v) return nil }) // Inspect the query result. fmt.Println("result:", result) fmt.Println("err:", err) }
Output: result: [2 3 4] err: <nil>
func MakeInput ¶ added in v0.4.0
MakeInput converts the slice of key/value pairs into a single input value. The keys define an object hierarchy.
func PlugHead ¶ added in v0.2.0
PlugHead returns a copy of head with bound terms substituted for the binding.
func PrettyTrace ¶ added in v0.2.0
PrettyTrace pretty prints the trace to the writer.
func RegisterBuiltinFunc ¶
func RegisterBuiltinFunc(name ast.Var, fun BuiltinFunc)
RegisterBuiltinFunc adds a new built-in function to the evaluation engine.
func RegisterFunctionalBuiltin1 ¶ added in v0.4.0
func RegisterFunctionalBuiltin1(name ast.Var, fun FunctionalBuiltin1)
RegisterFunctionalBuiltin1 adds a new built-in function to the evaluation engine.
Example ¶
package main import ( "context" "fmt" "strings" "github.com/open-policy-agent/opa/ast" "github.com/open-policy-agent/opa/topdown" "github.com/open-policy-agent/opa/topdown/builtins" ) func main() { // Rego includes a number of built-in functions ("built-ins") for performing // standard operations like string manipulation, regular expression // matching, and computing aggregates. // // This test shows how to add a new built-in to Rego and OPA. // Initialize context for the example. Normally the caller would obtain the // context from an input parameter or instantiate their own. ctx := context.Background() // The ast package contains a registry that enumerates the built-ins // included in Rego. When adding a new built-in, you must update the // registry to include your built-in. Otherwise, the compiler will complain // when it encounters your built-in. builtin := &ast.Builtin{ Name: ast.Var("selective_upper"), NumArgs: 2, TargetPos: []int{1}, } ast.RegisterBuiltin(builtin) // This is the implementation of the built-in that will be called during // query evaluation. builtinImpl := func(a ast.Value) (ast.Value, error) { str, err := builtins.StringOperand(a, 1) if err != nil { return nil, err } if str.Equal(ast.String("magic")) { // topdown.BuiltinEmpty indicates to the evaluation engine that the // expression is false/not defined. return nil, topdown.BuiltinEmpty{} } return ast.String(strings.ToUpper(string(str))), nil } // See documentation for registering functions that take different numbers // of arguments. topdown.RegisterFunctionalBuiltin1(builtin.Name, builtinImpl) // At this point, the new built-in has been registered and can be used in // queries. Our custom built-in converts strings to upper case but is not // defined for the input "magic". compiler := ast.NewCompiler() query, err := compiler.QueryCompiler().Compile(ast.MustParseBody(`selective_upper("custom", x); not selective_upper("magic", "MAGIC")`)) if err != nil { // Handle error. } // Evaluate the query. t := topdown.New(ctx, query, compiler, nil, nil) topdown.Eval(t, func(t *topdown.Topdown) error { fmt.Println("x:", t.Binding(ast.Var("x"))) return nil }) // If you are adding new built-in functions to upstream OPA, you must also // update the [Language // Reference](http://www.openpolicyagent.org/documentation/references/language/) // and [How Do I Write // Policies](http://www.openpolicyagent.org/documentation/how-do-i-write-policies/) // documents. In addition, you must add tests for your new built-in. See the // existing integration tests in the topdown package. }
Output: x: "CUSTOM"
func RegisterFunctionalBuiltin2 ¶ added in v0.4.0
func RegisterFunctionalBuiltin2(name ast.Var, fun FunctionalBuiltin2)
RegisterFunctionalBuiltin2 adds a new built-in function to the evaluation engine.
func RegisterFunctionalBuiltin3 ¶ added in v0.4.0
func RegisterFunctionalBuiltin3(name ast.Var, fun FunctionalBuiltin3)
RegisterFunctionalBuiltin3 adds a new built-in function to the evaluation engine.
func RegisterFunctionalBuiltinVoid2 ¶ added in v0.4.0
func RegisterFunctionalBuiltinVoid2(name ast.Var, fun FunctionalBuiltinVoid2)
RegisterFunctionalBuiltinVoid2 adds a new built-in function to the evaluation engine.
func ResetQueryIDs ¶ added in v0.2.0
func ResetQueryIDs()
ResetQueryIDs resets the query ID generator. This is only for test purposes.
func ResolveRefs ¶ added in v0.2.0
ResolveRefs returns the AST value obtained by resolving references to base documents.
func ValueToInterface ¶
ValueToInterface returns the underlying Go value associated with an AST value. If the value is a reference, the reference is fetched from storage. Composite AST values such as objects and arrays are converted recursively.
Types ¶
type Binding ¶ added in v0.2.0
Binding defines the interface used to apply term bindings to terms, expressions, etc.
type BufferTracer ¶ added in v0.2.0
type BufferTracer []*Event
BufferTracer implements the Tracer interface by simply buffering all events received.
func NewBufferTracer ¶ added in v0.2.0
func NewBufferTracer() *BufferTracer
NewBufferTracer returns a new BufferTracer.
func (*BufferTracer) Enabled ¶ added in v0.2.0
func (b *BufferTracer) Enabled() bool
Enabled always returns true.
func (*BufferTracer) Trace ¶ added in v0.2.0
func (b *BufferTracer) Trace(t *Topdown, evt *Event)
Trace adds the event to the buffer.
type BuiltinEmpty ¶ added in v0.4.0
type BuiltinEmpty struct{}
BuiltinEmpty is used to signal that the built-in function evaluated, but the result is undefined so evaluation should not continue.
func (BuiltinEmpty) Error ¶ added in v0.4.0
func (BuiltinEmpty) Error() string
type BuiltinFunc ¶
BuiltinFunc defines the interface that the evaluation engine uses to invoke built-in functions (built-ins). In most cases, custom built-ins can be implemented using the FunctionalBuiltin interfaces (which provide less control but are much simpler).
Users can implement their own built-ins and register them with OPA.
Built-ins are given the current evaluation context t with the expression expr to be evaluated. Built-ins can assume that the expression has been plugged with bindings from the current context however references to base documents will not have been resolved. If the built-in determines that the expression has evaluated successfully it should bind any output variables and invoke the iterator with the context produced by binding the output variables. Built-ins must be sure to unbind the outputs after the iterator returns.
type Error ¶
Error is the error type returned by the Eval and Query functions when an evaluation error occurs.
type Event ¶ added in v0.2.0
type Event struct { Op Op // Identifies type of event. Node interface{} // Contains AST node relevant to the event. QueryID uint64 // Identifies the query this event belongs to. ParentID uint64 // Identifies the parent query this event belongs to. Locals *ast.ValueMap // Contains local variable bindings from the query context. }
Event contains state associated with a tracing event.
type FunctionalBuiltin1 ¶ added in v0.4.0
FunctionalBuiltin1 defines an interface for simple functional built-ins.
Implement this interface if your built-in function takes one input and produces one output.
If an error occurs, the functional built-in should return a descriptive message. The message should not be prefixed with the built-in name as the framework takes care of this.
type FunctionalBuiltin2 ¶ added in v0.4.0
FunctionalBuiltin2 defines an interface for simple functional built-ins.
Implement this interface if your built-in function takes two inputs and produces one output.
If an error occurs, the functional built-in should return a descriptive message. The message should not be prefixed with the built-in name as the framework takes care of this.
type FunctionalBuiltin3 ¶ added in v0.4.0
FunctionalBuiltin3 defines an interface for simple functional built-ins.
Implement this interface if your built-in function takes three inputs and produces one output.
If an error occurs, the functional built-in should return a descriptive message. The message should not be prefixed with the built-in name as the framework takes care of this.
type FunctionalBuiltinVoid2 ¶ added in v0.4.0
FunctionalBuiltinVoid2 defines an interface for simple functional built-ins.
Implement this interface if your built-in function takes two inputs and produces no outputs.
If an error occurs, the functional built-in should return a descriptive message. The message should not be prefixed with the built-in name as the framework takes care of this.
type Op ¶ added in v0.2.0
type Op string
Op defines the types of tracing events.
const ( // EnterOp is emitted when a new query is about to be evaluated. EnterOp Op = "Enter" // ExitOp is emitted when a query has evaluated to true. ExitOp Op = "Exit" // EvalOp is emitted when an expression is about to be evaluated. EvalOp Op = "Eval" // RedoOp is emitted when an expression, rule, or query is being re-evaluated. RedoOp Op = "Redo" // FailOp is emitted when an expression evaluates to false. FailOp Op = "Fail" )
type QueryParams ¶
type QueryParams struct { Context context.Context Compiler *ast.Compiler Store *storage.Storage Transaction storage.Transaction Input ast.Value Tracer Tracer Path ast.Ref }
QueryParams defines input parameters for the query interface.
func NewQueryParams ¶
func NewQueryParams(ctx context.Context, compiler *ast.Compiler, store *storage.Storage, txn storage.Transaction, input ast.Value, path ast.Ref) *QueryParams
NewQueryParams returns a new QueryParams.
func (*QueryParams) NewTopdown ¶ added in v0.3.0
func (q *QueryParams) NewTopdown(query ast.Body) *Topdown
NewTopdown returns a new Topdown object.
This function will not propagate optional values such as the tracer, input document, etc. Those must be set by the caller.
type QueryResult ¶ added in v0.2.1
type QueryResult struct { Result interface{} // Result contains the document referred to by the params Path. Bindings map[string]interface{} // Bindings contains values for variables in the params Input. }
QueryResult represents a single query result.
func (*QueryResult) String ¶ added in v0.2.1
func (qr *QueryResult) String() string
type QueryResultSet ¶ added in v0.2.1
type QueryResultSet []*QueryResult
QueryResultSet represents a collection of query results.
func Query ¶
func Query(params *QueryParams) (QueryResultSet, error)
Query returns the value of the document referred to by the params' path. If the params' input contains non-ground terms, there may be multiple query results.
Example ¶
package main import ( "context" "fmt" "github.com/open-policy-agent/opa/ast" "github.com/open-policy-agent/opa/storage" "github.com/open-policy-agent/opa/topdown" ) func main() { // Initialize context for the example. Normally the caller would obtain the // context from an input parameter or instantiate their own. ctx := context.Background() compiler := ast.NewCompiler() // Define a dummy module with rules that produce documents that we will query below. module, err := ast.ParseModule("my_module.rego", `package opa.example p[x] { q[x]; not r[x] } q[y] { a = [1, 2, 3]; y = a[_] } r[z] { b = [2, 4]; z = b[_] }`, ) mods := map[string]*ast.Module{ "my_module": module, } if compiler.Compile(mods); compiler.Failed() { fmt.Println(compiler.Errors) } if err != nil { // Handle error. } // Instantiate the policy engine's storage layer. store := storage.New(storage.InMemoryConfig()) // Create a new transaction. Transactions allow the policy engine to // evaluate the query over a consistent snapshot fo the storage layer. txn, err := store.NewTransaction(ctx) if err != nil { // Handle error. } defer store.Close(ctx, txn) // Prepare query parameters. In this case, there are no additional documents // required by the policy so the input is nil. var input ast.Value params := topdown.NewQueryParams(ctx, compiler, store, txn, input, ast.MustParseRef("data.opa.example.p")) // Execute the query against "p". v1, err1 := topdown.Query(params) // Inspect the result. fmt.Println("v1:", v1[0].Result) fmt.Println("err1:", err1) }
Output: v1: [1 3] err1: <nil>
func (*QueryResultSet) Add ¶ added in v0.2.1
func (qrs *QueryResultSet) Add(qr *QueryResult)
Add inserts a result into the query result set.
func (QueryResultSet) Undefined ¶ added in v0.2.1
func (qrs QueryResultSet) Undefined() bool
Undefined returns true if the query did not find any results.
type Resolver ¶ added in v0.2.1
Resolver defines the interface for resolving references to base documents to native Go values. The native Go value types map to JSON types.
type Topdown ¶ added in v0.3.0
type Topdown struct { Query ast.Body Compiler *ast.Compiler Input ast.Value Index int Previous *Topdown Store *storage.Storage Tracer Tracer Context context.Context // contains filtered or unexported fields }
Topdown stores the state of the evaluation process and contains context needed to evaluate queries.
func New ¶ added in v0.3.0
func New(ctx context.Context, query ast.Body, compiler *ast.Compiler, store *storage.Storage, txn storage.Transaction) *Topdown
New returns a new Topdown object without any bindings.
func (*Topdown) Bind ¶ added in v0.3.0
Bind updates t to include a binding from the key to the value. The return value is used to return t to the state before the binding was added.
func (*Topdown) Child ¶ added in v0.3.0
Child returns a new Topdown object to evaluate query without bindings from t.
func (*Topdown) Closure ¶ added in v0.4.1
Closure returns a new Topdown object to evaluate query with bindings from t.
func (*Topdown) Resolve ¶ added in v0.3.0
Resolve returns the native Go value referred to by the ref.
func (*Topdown) Step ¶ added in v0.3.0
Step returns a new Topdown object to evaluate the next expression.
func (*Topdown) Unbind ¶ added in v0.3.0
Unbind updates t by removing the binding represented by the undo.
func (*Topdown) WithInput ¶ added in v0.4.1
WithInput returns a new Topdown object that has the input document set.
func (*Topdown) WithTracer ¶ added in v0.4.1
WithTracer returns a new Topdown object that has a tracer set.
Source Files ¶
Directories ¶
Path | Synopsis |
---|---|
Package builtins contains utilities for implementing built-in functions.
|
Package builtins contains utilities for implementing built-in functions. |
Package explain contains utilities for post-processing traces emitted by the evaluation engine.
|
Package explain contains utilities for post-processing traces emitted by the evaluation engine. |