@eagleoutice/flowr 2.2.14 → 2.2.16

package/control-flow/semantic-cfg-guided-visitor.js

@@ -6,6 +6,7 @@ const dfg_get_origin_1 = require("../dataflow/origin/dfg-get-origin");
 const type_1 = require("../r-bridge/lang-4.x/ast/model/type");
 const edge_1 = require("../dataflow/graph/edge");
 const assert_1 = require("../util/assert");
+const r_function_call_1 = require("../r-bridge/lang-4.x/ast/model/nodes/r-function-call");
 /**
  * This visitor extends on the {@link DataflowAwareCfgGuidedVisitor} by dispatching visitors for separate function calls as well,
  * providing more information!
@@ -51,8 +52,15 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
             case type_1.RType.Number: return this.onNumberConstant({ vertex: val, node: astNode });
             case type_1.RType.Logical: return this.onLogicalConstant({ vertex: val, node: astNode });
             case type_1.RType.Symbol:
-                (0, assert_1.guard)(astNode.lexeme === 'NULL', `Expected NULL constant, got ${astNode.lexeme}`);
-                return this.onNullConstant({ vertex: val, node: astNode });
+                if (astNode.lexeme === 'NULL') {
+                    return this.onNullConstant({
+                        vertex: val,
+                        node: astNode
+                    });
+                }
+                else {
+                    return this.onSymbolConstant({ vertex: val, node: astNode });
+                }
         }
         (0, assert_1.guard)(false, `Unexpected value type ${astNode.type} for value ${astNode.lexeme}`);
     }
@@ -90,7 +98,13 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
      */
     visitFunctionDefinition(vertex) {
         super.visitFunctionDefinition(vertex);
-        this.onFunctionDefinition({ vertex });
+        const ast = this.getNormalizedAst(vertex.id);
+        if (ast?.type === type_1.RType.FunctionDefinition) {
+            this.onFunctionDefinition({ vertex, parameters: ast.parameters.map(p => p.info.id) });
+        }
+        else {
+            this.onFunctionDefinition({ vertex });
+        }
     }
     /**
      * See {@link DataflowAwareCfgGuidedVisitor#visitFunctionCall} for the base implementation.
@@ -161,8 +175,26 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
                 return this.onSourceCall({ call });
             case 'builtin:access':
                 return this.onAccessCall({ call });
-            case 'builtin:if-then-else':
-                return this.onIfThenElseCall({ call, condition: call.args[0], then: call.args[1], else: call.args[2] });
+            case 'builtin:if-then-else': {
+                // recover dead arguments from ast
+                const ast = this.getNormalizedAst(call.id);
+                if (!ast || ast.type !== type_1.RType.IfThenElse) {
+                    return this.onIfThenElseCall({
+                        call,
+                        condition: call.args[0] === r_function_call_1.EmptyArgument ? undefined : call.args[0].nodeId,
+                        then: call.args[1] === r_function_call_1.EmptyArgument ? undefined : call.args[1].nodeId,
+                        else: call.args[2] === r_function_call_1.EmptyArgument ? undefined : call.args[2].nodeId
+                    });
+                }
+                else {
+                    return this.onIfThenElseCall({
+                        call,
+                        condition: ast.condition.info.id,
+                        then: ast.then.info.id,
+                        else: ast.otherwise?.info.id
+                    });
+                }
+            }
             case 'builtin:get':
                 return this.onGetCall({ call });
             case 'builtin:rm':
@@ -173,11 +205,11 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
                 return this.onVectorCall({ call });
             case 'table:assign':
             case 'builtin:assignment': {
-                const outgoing = this.config.dataflow.graph.outgoingEdges(call.id);
+                const outgoing = this.config.dfg.outgoingEdges(call.id);
                 if (outgoing) {
                     const target = [...outgoing.entries()].filter(([, e]) => (0, edge_1.edgeIncludesType)(e.types, edge_1.EdgeType.Returns));
                     if (target.length === 1) {
-                        const targetOut = this.config.dataflow.graph.outgoingEdges(target[0][0]);
+                        const targetOut = this.config.dfg.outgoingEdges(target[0][0]);
                         if (targetOut) {
                             const source = [...targetOut.entries()].filter(([t, e]) => (0, edge_1.edgeIncludesType)(e.types, edge_1.EdgeType.DefinedBy) && t !== call.id);
                             if (source.length === 1) {
@@ -189,9 +221,15 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
                 return this.onAssignmentCall({ call, target: undefined, source: undefined });
             }
             case 'builtin:special-bin-op':
-                return this.onSpecialBinaryOpCall({ call });
+                if (call.args.length !== 2) {
+                    return this.onSpecialBinaryOpCall({ call });
+                }
+                return this.onSpecialBinaryOpCall({ call, lhs: call.args[0], rhs: call.args[1] });
             case 'builtin:pipe':
-                return this.onPipeCall({ call });
+                if (call.args.length !== 2) {
+                    return this.onPipeCall({ call });
+                }
+                return this.onPipeCall({ call, lhs: call.args[0], rhs: call.args[1] });
             case 'builtin:quote':
                 return this.onQuoteCall({ call });
             case 'builtin:for-loop':
@@ -201,11 +239,11 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
             case 'builtin:while-loop':
                 return this.onWhileLoopCall({ call, condition: call.args[0], body: call.args[1] });
             case 'builtin:replacement': {
-                const outgoing = this.config.dataflow.graph.outgoingEdges(call.id);
+                const outgoing = this.config.dfg.outgoingEdges(call.id);
                 if (outgoing) {
                     const target = [...outgoing.entries()].filter(([, e]) => (0, edge_1.edgeIncludesType)(e.types, edge_1.EdgeType.Returns));
                     if (target.length === 1) {
-                        const targetOut = this.config.dataflow.graph.outgoingEdges(target[0][0]);
+                        const targetOut = this.config.dfg.outgoingEdges(target[0][0]);
                         if (targetOut) {
                             const source = [...targetOut.entries()].filter(([t, e]) => (0, edge_1.edgeIncludesType)(e.types, edge_1.EdgeType.DefinedBy) && t !== call.id);
                             if (source.length === 1) {
@@ -223,15 +261,23 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
                 return this.onDefaultFunctionCall({ call });
         }
     }
-    onProgram(_data) {
-    }
+    /**
+     * This event is called for the root program node, i.e., the program that is being analyzed.
+     *
+     * @protected
+     */
+    onProgram(_data) { }
     /**
      * A helper function to request the {@link getOriginInDfg|origins} of the given node.
      */
     getOrigins(id) {
-        return (0, dfg_get_origin_1.getOriginInDfg)(this.config.dataflow.graph, id);
+        return (0, dfg_get_origin_1.getOriginInDfg)(this.config.dfg, id);
     }
-    /** Called for every occurrence of a `NULL` in the program. */
+    /**
+     * Called for every occurrence of a `NULL` in the program.
+     *
+     * For other symbols that are not referenced as a variable, see {@link SemanticCfgGuidedVisitor#onSymbolConstant|`onSymbolConstant`}.
+     */
     onNullConstant(_data) { }
     /**
      * Called for every constant string value in the program.
@@ -251,6 +297,15 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
      * For example, `TRUE` in `if(TRUE) { ... }`.
      */
     onLogicalConstant(_data) { }
+    /**
+     * Called for every constant symbol value in the program.
+     *
+     * For example, `foo` in `library(foo)` or `a` in `l$a`. This most likely happens as part of non-standard-evaluation, i.e., the symbol is not evaluated to a value,
+     * but used as a symbol in and of itself.
+     *
+     * Please note, that due to its special behaviors, `NULL` is handled in {@link SemanticCfgGuidedVisitor#onNullConstant|`onNullConstant`} and not here.
+     */
+    onSymbolConstant(_data) { }
     /**
      * Called for every variable that is read within the program.
      * You can use {@link getOrigins} to get the origins of the variable.
@@ -300,7 +355,7 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
      *
      * This explicitly will not trigger for scenarios in which the function has no name (i.e., if it is anonymous).
      * For such cases, you may rely on the {@link SemanticCfgGuidedVisitor#onUnnamedCall|`onUnnamedCall`} event.
-     * The main reason for this separation is part of flowR's handling of these functions, as anonmyous calls cannot be resolved using the active environment.
+     * The main reason for this separation is part of flowR's handling of these functions, as anonymous calls cannot be resolved using the active environment.
      *
      * @protected
      */
@@ -351,15 +406,13 @@ class SemanticCfgGuidedVisitor extends dfg_cfg_guided_visitor_1.DataflowAwareCfg
      *
      * @protected
      */
-    onAccessCall(_data) {
-    }
+    onAccessCall(_data) { }
     /**
      * This event triggers for every call to the `if` function, which is used to implement the `if-then-else` control flow.
      *
      * @protected
      */
-    onIfThenElseCall(_data) {
-    }
+    onIfThenElseCall(_data) { }
     /**
      * This event triggers for every call to the `get` function, which is used to access variables in the global environment.
      *

package/control-flow/simple-visitor.d.ts

@@ -23,3 +23,7 @@ export declare function visitCfgInReverseOrder(graph: ControlFlowGraph, startNod
  * @see {@link visitCfgInReverseOrder} for a traversal in reversed order
  */
 export declare function visitCfgInOrder(graph: ControlFlowGraph, startNodes: readonly NodeId[], visitor: (node: NodeId) => boolean | void): void;
+/**
+ * Check if a node can reach another node in the control flow graph.
+ */
+export declare function canReach(graph: ControlFlowGraph, from: NodeId[], to: NodeId): boolean;

package/control-flow/simple-visitor.js

@@ -2,6 +2,7 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.visitCfgInReverseOrder = visitCfgInReverseOrder;
 exports.visitCfgInOrder = visitCfgInOrder;
+exports.canReach = canReach;
 const control_flow_graph_1 = require("./control-flow-graph");
 /**
  * Visit all nodes reachable from the start node in the control flow graph, traversing the dependencies but ignoring cycles.
@@ -77,4 +78,17 @@ visitor) {
         }
     }
 }
+/**
+ * Check if a node can reach another node in the control flow graph.
+ */
+function canReach(graph, from, to) {
+    let reached = false;
+    visitCfgInOrder(graph, from, node => {
+        if (node === to) {
+            reached = true;
+            return true;
+        }
+    });
+    return reached;
+}
 //# sourceMappingURL=simple-visitor.js.map

package/control-flow/syntax-cfg-guided-visitor.d.ts

@@ -25,7 +25,7 @@ import type { RNext } from '../r-bridge/lang-4.x/ast/model/nodes/r-next';
 import type { RNumber } from '../r-bridge/lang-4.x/ast/model/nodes/r-number';
 import type { RSymbol } from '../r-bridge/lang-4.x/ast/model/nodes/r-symbol';
 import type { NoInfo, RNode } from '../r-bridge/lang-4.x/ast/model/model';
-export interface SyntaxCfgGuidedVisitorConfiguration<OtherInfo = NoInfo, Cfg extends ControlFlowInformation = ControlFlowInformation, Ast extends NormalizedAst<OtherInfo> = NormalizedAst<OtherInfo>> extends BasicCfgGuidedVisitorConfiguration<Cfg> {
+export interface SyntaxCfgGuidedVisitorConfiguration<OtherInfo = NoInfo, ControlFlow extends ControlFlowInformation = ControlFlowInformation, Ast extends NormalizedAst<OtherInfo> = NormalizedAst<OtherInfo>> extends BasicCfgGuidedVisitorConfiguration<ControlFlow> {
     readonly normalizedAst: Ast;
 }
 /**
@@ -33,7 +33,7 @@ export interface SyntaxCfgGuidedVisitorConfiguration<OtherInfo = NoInfo, Cfg ext
  *
  * Use {@link BasicCfgGuidedVisitor#start} to start the traversal.
  */
-export declare class SyntaxAwareCfgGuidedVisitor<OtherInfo = NoInfo, Cfg extends ControlFlowInformation = ControlFlowInformation, Ast extends NormalizedAst<OtherInfo> = NormalizedAst<OtherInfo>, Config extends SyntaxCfgGuidedVisitorConfiguration<OtherInfo, Cfg, Ast> = SyntaxCfgGuidedVisitorConfiguration<OtherInfo, Cfg, Ast>> extends BasicCfgGuidedVisitor<Cfg, Config> {
+export declare class SyntaxAwareCfgGuidedVisitor<OtherInfo = NoInfo, ControlFlow extends ControlFlowInformation = ControlFlowInformation, Ast extends NormalizedAst<OtherInfo> = NormalizedAst<OtherInfo>, Config extends SyntaxCfgGuidedVisitorConfiguration<OtherInfo, ControlFlow, Ast> = SyntaxCfgGuidedVisitorConfiguration<OtherInfo, ControlFlow, Ast>> extends BasicCfgGuidedVisitor<ControlFlow, Config> {
     /**
      * Get the normalized AST node for the given id or fail if it does not exist.
      */

package/core/pipeline-executor.d.ts

@@ -1,6 +1,7 @@
 import type { PipelineStepName } from './steps/pipeline-step';
 import { PipelineStepStage } from './steps/pipeline-step';
 import type { Pipeline, PipelineInput, PipelineOutput, PipelinePerRequestInput, PipelineStepOutputWithName } from './steps/pipeline/pipeline';
+import type { FlowrConfigOptions } from '../config';
 /**
  * The pipeline executor allows to execute arbitrary {@link Pipeline|pipelines} in a step-by-step fashion.
  * If you are not yet in the possession of a {@link Pipeline|pipeline}, you can use the {@link createPipeline} function
@@ -91,6 +92,7 @@ export declare class PipelineExecutor<P extends Pipeline> {
     private output;
     private currentExecutionStage;
     private stepCounter;
+    private readonly flowrConfig;
     /**
      * Construct a new pipeline executor.
      * The required additional input is specified by the {@link IPipelineStep#requiredInput|required input configuration} of each step in the `pipeline`.
@@ -99,8 +101,9 @@ export declare class PipelineExecutor<P extends Pipeline> {
      *
      * @param pipeline - The {@link Pipeline} to execute, probably created with {@link createPipeline}.
      * @param input    - External {@link PipelineInput|configuration and input} required to execute the given pipeline.
+     * @param flowrConfig   - The flowr config containing the built-in definitions
      */
-    constructor(pipeline: P, input: PipelineInput<P>);
+    constructor(pipeline: P, input: PipelineInput<P>, flowrConfig: FlowrConfigOptions);
     /**
      * Retrieve the {@link Pipeline|pipeline} that is currently being.
      */

package/core/pipeline-executor.js

@@ -2,7 +2,6 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.PipelineExecutor = void 0;
 const assert_1 = require("../util/assert");
-const config_1 = require("../config");
 const built_in_1 = require("../dataflow/environments/built-in");
 const built_in_config_1 = require("../dataflow/environments/built-in-config");
 /**
@@ -95,6 +94,7 @@ class PipelineExecutor {
     output = {};
     currentExecutionStage = 0 /* PipelineStepStage.OncePerFile */;
     stepCounter = 0;
+    flowrConfig;
     /**
      * Construct a new pipeline executor.
      * The required additional input is specified by the {@link IPipelineStep#requiredInput|required input configuration} of each step in the `pipeline`.
@@ -103,13 +103,14 @@ class PipelineExecutor {
      *
      * @param pipeline - The {@link Pipeline} to execute, probably created with {@link createPipeline}.
      * @param input    - External {@link PipelineInput|configuration and input} required to execute the given pipeline.
+     * @param flowrConfig   - The flowr config containing the built-in definitions
      */
-    constructor(pipeline, input) {
+    constructor(pipeline, input, flowrConfig) {
         this.pipeline = pipeline;
         this.length = pipeline.order.length;
         this.input = input;
-        const config = (0, config_1.getConfig)();
-        const builtIns = config.semantics.environment.overwriteBuiltIns;
+        this.flowrConfig = flowrConfig;
+        const builtIns = flowrConfig.semantics.environment.overwriteBuiltIns;
         if (!builtIns.loadDefaults) {
             built_in_1.BuiltInMemory.clear();
             built_in_1.EmptyBuiltInMemory.clear();
@@ -190,7 +191,7 @@ class PipelineExecutor {
         if (expectedStepName !== undefined) {
             (0, assert_1.guard)(step.name === expectedStepName, () => `Cannot execute next step, expected step ${JSON.stringify(expectedStepName)} but got ${step.name}.`);
         }
-        return [step.name, step.processor(this.output, this.input)];
+        return [step.name, step.processor(this.output, this.input, this.flowrConfig)];
     }
     /**
      * This only makes sense if you have already run a request and want to re-use the per-file results for a new one.

package/core/steps/all/core/10-normalize.d.ts

@@ -9,6 +9,8 @@ export interface NormalizeRequiredInput {
     /** This id generator is only necessary if you want to retrieve a dataflow from the parsed R AST, it determines the id generator to use and by default uses the {@link deterministicCountingIdGenerator}*/
     readonly getId?: IdGenerator<NoInfo>;
     readonly request: RParseRequests;
+    /** assume the input stems from a file and use the given path as the file path for the AST */
+    readonly overwriteFilePath?: string;
 }
 export declare function getCurrentRequestFile(request: RParseRequests | undefined): string | undefined;
 declare function processor(results: {

package/core/steps/all/core/10-normalize.js

@@ -17,7 +17,7 @@ function getCurrentRequestFile(request) {
     }
 }
 function processor(results, input) {
-    return (0, parser_1.normalize)(results.parse, input.getId, getCurrentRequestFile(input.request));
+    return (0, parser_1.normalize)(results.parse, input.getId, input.overwriteFilePath ?? getCurrentRequestFile(input.request));
 }
 exports.NORMALIZE = {
     name: 'normalize',

package/core/steps/all/core/11-normalize-tree-sitter.d.ts

@@ -3,9 +3,10 @@ import { normalizedAstToJson, normalizedAstToQuads, printNormalizedAstToMermaid,
 import { PipelineStepStage } from '../../pipeline-step';
 import type { NormalizeRequiredInput } from './10-normalize';
 import type { ParseStepOutputTS } from './01-parse-tree-sitter';
+import type { FlowrConfigOptions } from '../../../../config';
 declare function processor(results: {
     'parse'?: ParseStepOutputTS;
-}, input: Partial<NormalizeRequiredInput>): import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").NormalizedAst<import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").ParentInformation, import("../../../../r-bridge/lang-4.x/ast/model/model").RNode<import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").ParentInformation>>;
+}, input: Partial<NormalizeRequiredInput>, config: FlowrConfigOptions): import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").NormalizedAst<import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").ParentInformation, import("../../../../r-bridge/lang-4.x/ast/model/model").RNode<import("../../../../r-bridge/lang-4.x/ast/model/processing/decorate").ParentInformation>>;
 export declare const NORMALIZE_TREE_SITTER: {
     readonly name: "normalize";
     readonly humanReadableName: "normalize tree-sitter tree";

package/core/steps/all/core/11-normalize-tree-sitter.js

@@ -5,8 +5,8 @@ const print_1 = require("../../../print/print");
 const normalize_printer_1 = require("../../../print/normalize-printer");
 const parser_1 = require("../../../../r-bridge/lang-4.x/ast/parser/json/parser");
 const _10_normalize_1 = require("./10-normalize");
-function processor(results, input) {
-    return (0, parser_1.normalizeTreeSitter)(results['parse'], input.getId, (0, _10_normalize_1.getCurrentRequestFile)(input.request));
+function processor(results, input, config) {
+    return (0, parser_1.normalizeTreeSitter)(results['parse'], input.getId, config, input.overwriteFilePath ?? (0, _10_normalize_1.getCurrentRequestFile)(input.request));
 }
 exports.NORMALIZE_TREE_SITTER = {
     name: 'normalize',

package/core/steps/all/core/20-dataflow.d.ts

@@ -4,12 +4,13 @@ import { dataflowGraphToJson, dataflowGraphToMermaid, dataflowGraphToMermaidUrl,
 import type { NormalizedAst } from '../../../../r-bridge/lang-4.x/ast/model/processing/decorate';
 import type { RParseRequests } from '../../../../r-bridge/retriever';
 import type { KnownParserType, Parser } from '../../../../r-bridge/parser';
+import type { FlowrConfigOptions } from '../../../../config';
 declare function processor(results: {
     normalize?: NormalizedAst;
 }, input: {
     request?: RParseRequests;
     parser?: Parser<KnownParserType>;
-}): import("../../../../dataflow/info").DataflowInformation;
+}, config: FlowrConfigOptions): import("../../../../dataflow/info").DataflowInformation;
 export declare const STATIC_DATAFLOW: {
     readonly humanReadableName: "dataflow";
     readonly processor: typeof processor;

package/core/steps/all/core/20-dataflow.js

@@ -17,8 +17,8 @@ const staticDataflowCommon = {
     },
     dependencies: ['normalize'],
 };
-function processor(results, input) {
-    return (0, extractor_1.produceDataFlowGraph)(input.parser, input.request, results.normalize);
+function processor(results, input, config) {
+    return (0, extractor_1.produceDataFlowGraph)(input.parser, input.request, results.normalize, config);
 }
 exports.STATIC_DATAFLOW = {
     ...staticDataflowCommon,

package/core/steps/all/static-slicing/00-slice.d.ts

@@ -3,6 +3,7 @@ import { PipelineStepStage } from '../../pipeline-step';
 import type { DataflowInformation } from '../../../../dataflow/info';
 import type { SlicingCriteria } from '../../../../slicing/criterion/parse';
 import type { NormalizedAst } from '../../../../r-bridge/lang-4.x/ast/model/processing/decorate';
+import type { FlowrConfigOptions } from '../../../../config';
 export interface SliceRequiredInput {
     /** The slicing criterion is only of interest if you actually want to slice the R code */
     readonly criterion: SlicingCriteria;
@@ -12,7 +13,7 @@ export interface SliceRequiredInput {
 declare function processor(results: {
     dataflow?: DataflowInformation;
     normalize?: NormalizedAst;
-}, input: Partial<SliceRequiredInput>): Readonly<import("../../../../slicing/static/slicer-types").SliceResult>;
+}, input: Partial<SliceRequiredInput>, config: FlowrConfigOptions): Readonly<import("../../../../slicing/static/slicer-types").SliceResult>;
 export declare const STATIC_SLICE: {
     readonly name: "slice";
     readonly humanReadableName: "static slice";

package/core/steps/all/static-slicing/00-slice.js

@@ -3,8 +3,8 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.STATIC_SLICE = void 0;
 const print_1 = require("../../../print/print");
 const static_slicer_1 = require("../../../../slicing/static/static-slicer");
-function processor(results, input) {
-    return (0, static_slicer_1.staticSlicing)(results.dataflow.graph, results.normalize, input.criterion, input.threshold);
+function processor(results, input, config) {
+    return (0, static_slicer_1.staticSlicing)(results.dataflow.graph, results.normalize, input.criterion, input.threshold ?? config.solver.slicer?.threshold);
 }
 exports.STATIC_SLICE = {
     name: 'slice',