@eagleoutice/flowr 1.3.14 → 1.4.2

package/statistics/features/supported/used-functions/post-process.d.ts

@@ -1,5 +1,5 @@
-import { FeatureStatisticsWithMeta } from '../../feature';
-import { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
+import type { FeatureStatisticsWithMeta } from '../../feature';
+import type { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
 /**
  * Note: the summary does not contain a 0 for each function that is _not_ called by a file. Hence, the minimum can not be 0 (division for mean etc. will still be performed on total file count)
  */

package/statistics/features/supported/used-functions/post-process.js

@@ -9,7 +9,7 @@ const common_syntax_probability_1 = require("../../common-syntax-probability");
 const process_1 = require("../../../../util/summarizer/benchmark/first-phase/process");
 const files_1 = require("../../../../util/files");
 const path_1 = __importDefault(require("path"));
-const node_fs_1 = __importDefault(require("node:fs"));
+const fs_1 = __importDefault(require("fs"));
 const time_1 = require("../../../../util/time");
 const used_functions_1 = require("./used-functions");
 const numbers_1 = require("../../../../util/numbers");
@@ -33,7 +33,7 @@ function retrieveFunctionCallInformation(featureRoot, info, config, outputPath)
         'assign',
         'get',
         'setGeneric',
-        'R6Class'].map(name => [name, node_fs_1.default.createWriteStream(path_1.default.join(outputPath, `${name}.csv`))]));
+        'R6Class'].map(name => [name, fs_1.default.createWriteStream(path_1.default.join(outputPath, `${name}.csv`))]));
     for (const [, value] of importantFunctions) {
         value.write('filepath,location,namespace,inspected by,classification,notes\n');
     }
@@ -44,7 +44,7 @@ function retrieveFunctionCallInformation(featureRoot, info, config, outputPath)
     }
     importantFunctions.clear();
     console.log(`    [${(0, time_1.date2string)(new Date())}] Used functions process completed, start to write out function info`);
-    const fnOutStream = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'function-calls.csv'));
+    const fnOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'function-calls.csv'));
     const prefixes = ['total', 'args', 'line-frac'];
     const others = prefixes.flatMap(data_1.summarizedMeasurement2CsvHeader).join(',');
     fnOutStream.write(`function,unique-projects,unique-files,${others}\n`);
@@ -58,7 +58,7 @@ function retrieveFunctionCallInformation(featureRoot, info, config, outputPath)
     fnOutStream.close();
 }
 function writeFunctionCallsMetaInformationToCsv(outputPath, data) {
-    const out = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'function-calls-meta.csv'));
+    const out = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'function-calls-meta.csv'));
     out.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     out.write(`average-call,${(0, data_1.summarizedMeasurement2Csv)((0, process_1.summarizeMeasurement)(data.averageCall.flat()))}\n`);
     out.write(`nested-calls,${(0, data_1.summarizedMeasurement2Csv)((0, process_1.summarizeMeasurement)(data.nestedCalls.flat()))}\n`);
@@ -102,7 +102,7 @@ function retrieveFunctionCallMetaInformation(info, outputPath) {
             // we treat the first/0-argument entry separate for legacy reasons
             continue;
         }
-        const out = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, `function-calls-arg-${index}.csv`));
+        const out = fs_1.default.createWriteStream(path_1.default.join(outputPath, `function-calls-arg-${index}.csv`));
         out.write(`kind,name,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
         for (const [name, vals] of Object.entries(arg)) {
             if (Array.isArray(vals)) {

package/statistics/features/supported/used-functions/used-functions.d.ts

@@ -1,6 +1,6 @@
-import { Feature } from '../../feature';
-import { Writable } from 'ts-essentials';
-import { CommonSyntaxTypeCounts } from '../../common-syntax-probability';
+import type { Feature } from '../../feature';
+import type { Writable } from 'ts-essentials';
+import type { CommonSyntaxTypeCounts } from '../../common-syntax-probability';
 declare const initialFunctionUsageInfo: {
     allFunctionCalls: number;
     args: Record<number, bigint | CommonSyntaxTypeCounts<bigint>>;

package/statistics/features/supported/used-packages/post-process.d.ts

@@ -1,3 +1,3 @@
-import { FeatureStatisticsWithMeta } from '../../feature';
-import { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
+import type { FeatureStatisticsWithMeta } from '../../feature';
+import type { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
 export declare function postProcess(featureRoot: string, info: Map<string, FeatureStatisticsWithMeta>, outputPath: string, config: StatisticsSummarizerConfiguration): void;

package/statistics/features/supported/used-packages/post-process.js

@@ -6,7 +6,7 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.postProcess = void 0;
 const post_processing_1 = require("../../post-processing");
 const used_packages_1 = require("./used-packages");
-const node_fs_1 = __importDefault(require("node:fs"));
+const fs_1 = __importDefault(require("fs"));
 const path_1 = __importDefault(require("path"));
 const data_1 = require("../../../../util/summarizer/benchmark/data");
 const process_1 = require("../../../../util/summarizer/benchmark/first-phase/process");
@@ -29,7 +29,7 @@ function postProcess(featureRoot, info, outputPath, config) {
             }
         }
     }
-    const variablesOutStream = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'loading-functions.csv'));
+    const variablesOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'loading-functions.csv'));
     variablesOutStream.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of Object.entries(collected)) {
         const data = val;
@@ -58,7 +58,7 @@ function postProcess(featureRoot, info, outputPath, config) {
         }
     }
     const output = path_1.default.join(outputPath, 'all-operators.csv');
-    const out = node_fs_1.default.createWriteStream(output);
+    const out = fs_1.default.createWriteStream(output);
     out.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of loadedLibraries.entries()) {
         const { count, uniqueProjects, uniqueFiles } = val;
@@ -71,7 +71,7 @@ exports.postProcess = postProcess;
 // directly writes the results to the output path
 function retrieveDataForLoad(operator, readFromPath, outputPath, config) {
     const input = path_1.default.join(readFromPath, `${operator}.txt`);
-    if (!node_fs_1.default.existsSync(input)) {
+    if (!fs_1.default.existsSync(input)) {
         // if there is nothing with this, just return :)
         return new Map();
     }
@@ -108,7 +108,7 @@ function retrieveDataForLoad(operator, readFromPath, outputPath, config) {
         }
     });
     const output = path_1.default.join(outputPath, `${operator}.csv`);
-    const out = node_fs_1.default.createWriteStream(output);
+    const out = fs_1.default.createWriteStream(output);
     out.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of collected.entries()) {
         const { count, uniqueProjects, uniqueFiles } = val;

package/statistics/features/supported/used-packages/used-packages.d.ts

@@ -1,5 +1,5 @@
-import { Feature } from '../../feature';
-import { Writable } from 'ts-essentials';
+import type { Feature } from '../../feature';
+import type { Writable } from 'ts-essentials';
 export declare const initialUsedPackageInfos: {
     library: number;
     require: number;

package/statistics/features/supported/values/post-process.d.ts

@@ -1,3 +1,3 @@
-import { FeatureStatisticsWithMeta } from '../../feature';
-import { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
+import type { FeatureStatisticsWithMeta } from '../../feature';
+import type { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
 export declare function postProcess(featureRoot: string, info: Map<string, FeatureStatisticsWithMeta>, outputPath: string, config: StatisticsSummarizerConfiguration): void;

package/statistics/features/supported/values/post-process.js

@@ -5,7 +5,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.postProcess = void 0;
 const data_1 = require("../../../../util/summarizer/benchmark/data");
-const node_fs_1 = __importDefault(require("node:fs"));
+const fs_1 = __importDefault(require("fs"));
 const path_1 = __importDefault(require("path"));
 const process_1 = require("../../../../util/summarizer/benchmark/first-phase/process");
 const files_1 = require("../../../../util/files");
@@ -28,7 +28,7 @@ function postProcess(featureRoot, info, outputPath, config) {
             }
         }
     }
-    const valuesOutStream = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'values.csv'));
+    const valuesOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'values.csv'));
     valuesOutStream.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of Object.entries(collected)) {
         const data = val;
@@ -60,7 +60,7 @@ function postProcess(featureRoot, info, outputPath, config) {
             }
         }
     });
-    const magicNumbersOutStream = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'magic-numbers.csv'));
+    const magicNumbersOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'magic-numbers.csv'));
     magicNumbersOutStream.write(`num,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of valueMap.entries()) {
         const sum = (0, process_1.summarizeMeasurement)(val.count);

package/statistics/features/supported/values/values.d.ts

@@ -1,5 +1,5 @@
-import { Feature } from '../../feature';
-import { Writable } from 'ts-essentials';
+import type { Feature } from '../../feature';
+import type { Writable } from 'ts-essentials';
 declare const initialValueInfo: {
     allNumerics: number;
     imaginaryNumbers: number;

package/statistics/features/supported/variables/post-process.d.ts

@@ -1,6 +1,6 @@
-import { FeatureStatisticsWithMeta } from '../../feature';
-import { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
-import { SummarizedWithProject } from '../../post-processing';
+import type { FeatureStatisticsWithMeta } from '../../feature';
+import type { StatisticsSummarizerConfiguration } from '../../../../util/summarizer/statistics/summarizer';
+import type { SummarizedWithProject } from '../../post-processing';
 export declare function postProcess(featureRoot: string, info: Map<string, FeatureStatisticsWithMeta>, outputPath: string, config: StatisticsSummarizerConfiguration): void;
 type VariableInfoMap = Map<string, SummarizedWithProject & {
     linePercentageInFile: number[][];

package/statistics/features/supported/variables/post-process.js

@@ -5,7 +5,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.writeVariableInfoToCsv = exports.postProcess = void 0;
 const post_processing_1 = require("../../post-processing");
-const node_fs_1 = __importDefault(require("node:fs"));
+const fs_1 = __importDefault(require("fs"));
 const path_1 = __importDefault(require("path"));
 const data_1 = require("../../../../util/summarizer/benchmark/data");
 const process_1 = require("../../../../util/summarizer/benchmark/first-phase/process");
@@ -27,7 +27,7 @@ function collectRedefinedVariables(featureRoot, info, config, outputPath) {
     redefined.clear();
 }
 function writeVariableCountsToCsv(outputPath, collected) {
-    const variablesOutStream = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, 'variable-counts.csv'));
+    const variablesOutStream = fs_1.default.createWriteStream(path_1.default.join(outputPath, 'variable-counts.csv'));
     variablesOutStream.write(`kind,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)()}\n`);
     for (const [key, val] of Object.entries(collected)) {
         if (key === 'unknownVariables') {
@@ -67,7 +67,7 @@ function postProcess(featureRoot, info, outputPath, config) {
 }
 exports.postProcess = postProcess;
 function writeVariableInfoToCsv(outputPath, filename, data) {
-    const out = node_fs_1.default.createWriteStream(path_1.default.join(outputPath, filename));
+    const out = fs_1.default.createWriteStream(path_1.default.join(outputPath, filename));
     out.write(`variable,unique-projects,unique-files,${(0, data_1.summarizedMeasurement2CsvHeader)('count')},${(0, data_1.summarizedMeasurement2CsvHeader)('line-frac')}\n`);
     for (const [key, val] of data.entries()) {
         const { count, uniqueProjects, uniqueFiles, linePercentageInFile } = val;

package/statistics/features/supported/variables/variables.d.ts

@@ -1,5 +1,5 @@
-import { Feature } from '../../feature';
-import { Writable } from 'ts-essentials';
+import type { Feature } from '../../feature';
+import type { Writable } from 'ts-essentials';
 declare const initialVariableInfo: {
     numberOfVariableUses: number;
     numberOfDefinitions: number;

package/statistics/meta-statistics.d.ts

@@ -1,4 +1,4 @@
-import { RParseRequestFromFile, RParseRequestFromText } from '../r-bridge';
+import type { RParseRequestFromFile, RParseRequestFromText } from '../r-bridge';
 /**
  * Statistics on skipped files, the time required, and more.
  *

package/statistics/output/ansi.js

@@ -17,7 +17,7 @@ exports.voidFormatter = new class {
     reset() {
         return '';
     }
-};
+}();
 /**
  * This does not work if the {@link setFormatter | formatter} is void. Tries to format the text with a bold font weight.
  */

package/statistics/output/print-stats.d.ts

@@ -1,5 +1,5 @@
-import { FeatureKey, FeatureStatistics } from '../features';
-import { MetaStatistics } from '../meta-statistics';
+import type { FeatureKey, FeatureStatistics } from '../features';
+import type { MetaStatistics } from '../meta-statistics';
 interface MinMaxAvgMedian {
     sum: number;
     min: number;

package/statistics/output/statistics-file.d.ts

@@ -1,4 +1,4 @@
-import { DummyAppendMemoryMap, StatisticAppendProvider } from './file-provider';
+import type { DummyAppendMemoryMap, StatisticAppendProvider } from './file-provider';
 /**
  * Requires source information to be attached on parsing!
  * <p>

package/statistics/statistics.d.ts

@@ -1,7 +1,7 @@
-import { RParseRequestFromFile, RParseRequestFromText, RShell } from '../r-bridge';
-import { FeatureSelection, FeatureStatistics } from './features';
-import { MetaStatistics } from './meta-statistics';
-import { StepResults } from '../core';
+import type { RParseRequestFromFile, RParseRequestFromText, RShell } from '../r-bridge';
+import type { FeatureSelection, FeatureStatistics } from './features';
+import type { MetaStatistics } from './meta-statistics';
+import type { StepResults } from '../core';
 /**
  * By default, {@link extractUsageStatistics} requires a generator, but sometimes you already know all the files
  * that you want to process. This function simply reps your requests as a generator.

package/statistics/statistics.js

@@ -4,6 +4,7 @@ var __importDefault = (this && this.__importDefault) || function (mod) {
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.extractUsageStatistics = exports.staticRequests = void 0;
+const r_bridge_1 = require("../r-bridge");
 const features_1 = require("./features");
 const xmldom_1 = require("@xmldom/xmldom");
 const fs_1 = __importDefault(require("fs"));
@@ -37,7 +38,6 @@ exports.staticRequests = staticRequests;
 async function extractUsageStatistics(shell, onRequest, features, requests, rootPath) {
     let result = initializeFeatureStatistics();
     const meta = (0, meta_statistics_1.initialMetaStatistics)();
-    let first = true;
     const outputs = new Map();
     for await (const request of requests) {
         onRequest(request);
@@ -45,14 +45,10 @@ async function extractUsageStatistics(shell, onRequest, features, requests, root
         const suffix = request.request === 'file' ? request.content.replace(new RegExp('^' + (rootPath ?? '')), '') : undefined;
         try {
             let output;
-            ({ stats: result, output } = await extractSingle(result, shell, {
-                ...request,
-                ensurePackageInstalled: first
-            }, features, suffix));
+            ({ stats: result, output } = await extractSingle(result, shell, request, features, suffix));
             outputs.set(request, output);
             processMetaOnSuccessful(meta, request);
             meta.numberOfNormalizedNodes.push(output.normalize.idMap.size);
-            first = false;
         }
         catch (e) {
             log_1.log.error('for request: ', request, e);
@@ -89,8 +85,11 @@ async function extractSingle(result, shell, request, features, suffixFilePath) {
         stepOfInterest: 'dataflow',
         request, shell
     }).allRemainingSteps();
-    // await retrieveXmlFromRCode(from, shell)
-    const doc = parser.parseFromString(slicerOutput.parse, 'text/xml');
+    // retrieve parsed xml through (legacy) xmlparsedata
+    const suffix = request.request === 'file' ? ', encoding="utf-8"' : '';
+    shell.sendCommands(`try(flowr_parsed<-parse(${request.request}=${JSON.stringify(request.content)},keep.source=TRUE${suffix}),silent=FALSE)`, 'try(flowr_output<-xmlparsedata::xml_parse_data(flowr_parsed,includeText=TRUE,pretty=FALSE),silent=FALSE)');
+    const parsed = (await shell.sendCommandWithOutput(`cat(flowr_output,${(0, r_bridge_1.ts2r)(shell.options.eol)})`)).join(shell.options.eol);
+    const doc = parser.parseFromString(parsed, 'text/xml');
     // eslint-disable-next-line @typescript-eslint/no-explicit-any
     for (const [key, feature] of Object.entries(features_1.ALL_FEATURES)) {
         if (features !== 'all' && !features.has(key)) {

package/util/args.d.ts

@@ -4,11 +4,15 @@
  * @module
  */
 /**
- * This splits an input string on the given split string (e.g., ` `) but checks if the string is quoted or escaped.
+ * This splits an input string on the given split string (e.g., ` `), but checks if the string is quoted or escaped.
  *
- * Given an input string like `a "b c" d` with a space character as split this splits the arguments similar to common shell interpreters (i.e., `a`, `b c`, and `d`).
+ * Given an input string like `a "b c" d`, with a space character as split, and escapeQuote set to true,
+ * this splits the arguments similar to common shell interpreters (i.e., `a`, `b c`, and `d`).
+ *
+ * When escapeQuote is set to false instead, we keep quotation marks in the result (i.e., `a`, `"b c"`, and `d`.).
  *
  * @param inputString - The string to split
- * @param split       - The **single** character to split on (can not be the backslash or quote)
+ * @param escapeQuote - Keep quotes in args
+ * @param split       - The **single** character to split on (can not be backslash or quote)
  */
-export declare function splitAtEscapeSensitive(inputString: string, split?: string): string[];
+export declare function splitAtEscapeSensitive(inputString: string, escapeQuote?: boolean, split?: string): string[];

package/util/args.js

@@ -7,14 +7,18 @@
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.splitAtEscapeSensitive = void 0;
 /**
- * This splits an input string on the given split string (e.g., ` `) but checks if the string is quoted or escaped.
+ * This splits an input string on the given split string (e.g., ` `), but checks if the string is quoted or escaped.
  *
- * Given an input string like `a "b c" d` with a space character as split this splits the arguments similar to common shell interpreters (i.e., `a`, `b c`, and `d`).
+ * Given an input string like `a "b c" d`, with a space character as split, and escapeQuote set to true,
+ * this splits the arguments similar to common shell interpreters (i.e., `a`, `b c`, and `d`).
+ *
+ * When escapeQuote is set to false instead, we keep quotation marks in the result (i.e., `a`, `"b c"`, and `d`.).
  *
  * @param inputString - The string to split
- * @param split       - The **single** character to split on (can not be the backslash or quote)
+ * @param escapeQuote - Keep quotes in args
+ * @param split       - The **single** character to split on (can not be backslash or quote)
  */
-function splitAtEscapeSensitive(inputString, split = ' ') {
+function splitAtEscapeSensitive(inputString, escapeQuote = true, split = ' ') {
     const args = [];
     let current = '';
     let inQuotes = false;
@@ -50,6 +54,9 @@ function splitAtEscapeSensitive(inputString, split = ' ') {
         }
         else if (c === '"' || c === "'") {
             inQuotes = !inQuotes;
+            if (!escapeQuote) {
+                current += c;
+            }
         }
         else if (c === '\\') {
             escaped = true;

package/util/{cfg.d.ts → cfg/cfg.d.ts}

@@ -1,8 +1,20 @@
-import { NodeId, NormalizedAst, ParentInformation, RFalse, RTrue } from '../r-bridge';
-import { MergeableRecord } from './objects';
-import { QuadSerializationConfiguration } from './quads';
+import type { NodeId, NormalizedAst, ParentInformation } from '../../r-bridge';
+import { RFalse, RTrue } from '../../r-bridge';
+import type { MergeableRecord } from '../objects';
+import type { QuadSerializationConfiguration } from '../quads';
+export declare const enum CfgVertexType {
+    /** Marks a break point in a construct (e.g., between the name and the value of an argument, or the formals and the body of a function)  */
+    MidMarker = "mid-marker",
+    /** The explicit exit-nodes to ensure the hammock property */
+    EndMarker = "end-marker",
+    /** something like an if, assignment, ... even though in the classical sense of R they are still expressions */
+    Statement = "statement",
+    /** something like an addition, ... */
+    Expression = "expression"
+}
 export interface CfgVertex {
     id: NodeId;
+    type: CfgVertexType;
     name: string;
     /** in case of a function definition */
     children?: NodeId[];
@@ -14,7 +26,7 @@ interface CfgControlDependencyEdge extends MergeableRecord {
     label: 'CD';
     when: typeof RTrue | typeof RFalse;
 }
-export type CFGEdge = CfgFlowDependencyEdge | CfgControlDependencyEdge;
+export type CfgEdge = CfgFlowDependencyEdge | CfgControlDependencyEdge;
 /**
  * This class represents the control flow graph of an R program.
  * The control flow may be hierarchical when confronted with function definitions (see {@link CfgVertex} and {@link CFG#rootVertexIds|rootVertexIds()}).
@@ -24,10 +36,10 @@ export declare class ControlFlowGraph {
     private vertexInformation;
     private edgeInformation;
     addVertex(vertex: CfgVertex, rootVertex?: boolean): this;
-    addEdge(from: NodeId, to: NodeId, edge: CFGEdge): this;
+    addEdge(from: NodeId, to: NodeId, edge: CfgEdge): this;
     rootVertexIds(): ReadonlySet<NodeId>;
     vertices(): ReadonlyMap<NodeId, CfgVertex>;
-    edges(): ReadonlyMap<NodeId, ReadonlyMap<NodeId, CFGEdge>>;
+    edges(): ReadonlyMap<NodeId, ReadonlyMap<NodeId, CfgEdge>>;
     merge(other: ControlFlowGraph, forceNested?: boolean): this;
 }
 export interface ControlFlowInformation extends MergeableRecord {

package/util/{cfg.js → cfg/cfg.js}

@@ -1,11 +1,11 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.cfg2quads = exports.equalCfg = exports.extractCFG = exports.emptyControlFlowInformation = exports.ControlFlowGraph = void 0;
-const r_bridge_1 = require("../r-bridge");
-const set_1 = require("./set");
-const quads_1 = require("./quads");
-const log_1 = require("./log");
-const json_1 = require("./json");
+const r_bridge_1 = require("../../r-bridge");
+const set_1 = require("../set");
+const quads_1 = require("../quads");
+const log_1 = require("../log");
+const json_1 = require("../json");
 /**
  * This class represents the control flow graph of an R program.
  * The control flow may be hierarchical when confronted with function definitions (see {@link CfgVertex} and {@link CFG#rootVertexIds|rootVertexIds()}).
@@ -65,10 +65,10 @@ function emptyControlFlowInformation() {
 }
 exports.emptyControlFlowInformation = emptyControlFlowInformation;
 const cfgFolds = {
-    foldNumber: cfgLeaf,
-    foldString: cfgLeaf,
-    foldLogical: cfgLeaf,
-    foldSymbol: cfgLeaf,
+    foldNumber: cfgLeaf("expression" /* CfgVertexType.Expression */),
+    foldString: cfgLeaf("expression" /* CfgVertexType.Expression */),
+    foldLogical: cfgLeaf("expression" /* CfgVertexType.Expression */),
+    foldSymbol: cfgLeaf("expression" /* CfgVertexType.Expression */),
     foldAccess: cfgAccess,
     binaryOp: {
         foldLogicalOp: cfgBinaryOp,
@@ -107,24 +107,29 @@ function extractCFG(ast) {
     return (0, r_bridge_1.foldAst)(ast.ast, cfgFolds);
 }
 exports.extractCFG = extractCFG;
-function cfgLeaf(leaf) {
-    const graph = new ControlFlowGraph();
-    graph.addVertex({ id: leaf.info.id, name: leaf.type });
-    return { graph, breaks: [], nexts: [], returns: [], exitPoints: [leaf.info.id], entryPoints: [leaf.info.id] };
+function cfgLeaf(type) {
+    return (leaf) => {
+        const graph = new ControlFlowGraph();
+        graph.addVertex({ id: leaf.info.id, name: leaf.type, type });
+        return { graph, breaks: [], nexts: [], returns: [], exitPoints: [leaf.info.id], entryPoints: [leaf.info.id] };
+    };
 }
 function cfgBreak(leaf) {
-    return { ...cfgLeaf(leaf), breaks: [leaf.info.id] };
+    return { ...cfgLeaf("statement" /* CfgVertexType.Statement */)(leaf), breaks: [leaf.info.id] };
 }
 function cfgNext(leaf) {
-    return { ...cfgLeaf(leaf), nexts: [leaf.info.id] };
+    return { ...cfgLeaf("statement" /* CfgVertexType.Statement */)(leaf), nexts: [leaf.info.id] };
 }
 function cfgIgnore(_leaf) {
     return { graph: new ControlFlowGraph(), breaks: [], nexts: [], returns: [], exitPoints: [], entryPoints: [] };
 }
+function identifyMayStatementType(node) {
+    return node.info.role === "expr-list-child" /* RoleInParent.ExpressionListChild */ ? "statement" /* CfgVertexType.Statement */ : "expression" /* CfgVertexType.Expression */;
+}
 function cfgIfThenElse(ifNode, condition, then, otherwise) {
     const graph = new ControlFlowGraph();
-    graph.addVertex({ id: ifNode.info.id, name: ifNode.type });
-    graph.addVertex({ id: ifNode.info.id + '-exit', name: 'if-exit' });
+    graph.addVertex({ id: ifNode.info.id, name: ifNode.type, type: identifyMayStatementType(ifNode) });
+    graph.addVertex({ id: ifNode.info.id + '-exit', name: 'if-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     graph.merge(condition.graph);
     graph.merge(then.graph);
     if (otherwise) {
@@ -160,8 +165,8 @@ function cfgIfThenElse(ifNode, condition, then, otherwise) {
 }
 function cfgRepeat(repeat, body) {
     const graph = body.graph;
-    graph.addVertex({ id: repeat.info.id, name: repeat.type });
-    graph.addVertex({ id: repeat.info.id + '-exit', name: 'repeat-exit' });
+    graph.addVertex({ id: repeat.info.id, name: repeat.type, type: identifyMayStatementType(repeat) });
+    graph.addVertex({ id: repeat.info.id + '-exit', name: 'repeat-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     for (const entryPoint of body.entryPoints) {
         graph.addEdge(repeat.info.id, entryPoint, { label: 'FD' });
     }
@@ -176,8 +181,8 @@ function cfgRepeat(repeat, body) {
 }
 function cfgWhile(whileLoop, condition, body) {
     const graph = condition.graph;
-    graph.addVertex({ id: whileLoop.info.id, name: whileLoop.type });
-    graph.addVertex({ id: whileLoop.info.id + '-exit', name: 'while-exit' });
+    graph.addVertex({ id: whileLoop.info.id, name: whileLoop.type, type: identifyMayStatementType(whileLoop) });
+    graph.addVertex({ id: whileLoop.info.id + '-exit', name: 'while-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     graph.merge(body.graph);
     for (const entry of condition.entryPoints) {
         graph.addEdge(entry, whileLoop.info.id, { label: 'FD' });
@@ -204,8 +209,8 @@ function cfgWhile(whileLoop, condition, body) {
 }
 function cfgFor(forLoop, variable, vector, body) {
     const graph = variable.graph;
-    graph.addVertex({ id: forLoop.info.id, name: forLoop.type });
-    graph.addVertex({ id: forLoop.info.id + '-exit', name: 'for-exit' });
+    graph.addVertex({ id: forLoop.info.id, name: forLoop.type, type: identifyMayStatementType(forLoop) });
+    graph.addVertex({ id: forLoop.info.id + '-exit', name: 'for-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     graph.merge(vector.graph);
     graph.merge(body.graph);
     for (const entry of vector.entryPoints) {
@@ -236,9 +241,9 @@ function cfgFor(forLoop, variable, vector, body) {
 function cfgFunctionDefinition(fn, params, body) {
     const graph = new ControlFlowGraph();
     const children = [fn.info.id + '-params', fn.info.id + '-exit'];
-    graph.addVertex({ id: fn.info.id + '-params', name: 'function-parameters' }, false);
-    graph.addVertex({ id: fn.info.id + '-exit', name: 'function-exit' }, false);
-    graph.addVertex({ id: fn.info.id, name: fn.type, children });
+    graph.addVertex({ id: fn.info.id + '-params', name: 'function-parameters', type: "mid-marker" /* CfgVertexType.MidMarker */ }, false);
+    graph.addVertex({ id: fn.info.id + '-exit', name: 'function-exit', type: "end-marker" /* CfgVertexType.EndMarker */ }, false);
+    graph.addVertex({ id: fn.info.id, name: fn.type, children, type: identifyMayStatementType(fn) });
     graph.merge(body.graph, true);
     children.push(...body.graph.rootVertexIds());
     for (const param of params) {
@@ -263,15 +268,15 @@ function cfgFunctionDefinition(fn, params, body) {
 function cfgFunctionCall(call, name, args) {
     const graph = name.graph;
     const info = { graph, breaks: [...name.breaks], nexts: [...name.nexts], returns: [...name.returns], exitPoints: [call.info.id + '-exit'], entryPoints: [call.info.id] };
-    graph.addVertex({ id: call.info.id, name: call.type });
+    graph.addVertex({ id: call.info.id, name: call.type, type: identifyMayStatementType(call) });
     for (const entryPoint of name.entryPoints) {
         graph.addEdge(entryPoint, call.info.id, { label: 'FD' });
     }
-    graph.addVertex({ id: call.info.id + '-name', name: 'call-name' });
+    graph.addVertex({ id: call.info.id + '-name', name: 'call-name', type: "mid-marker" /* CfgVertexType.MidMarker */ });
     for (const exitPoint of name.exitPoints) {
         graph.addEdge(call.info.id + '-name', exitPoint, { label: 'FD' });
     }
-    graph.addVertex({ id: call.info.id + '-exit', name: 'call-exit' });
+    graph.addVertex({ id: call.info.id + '-exit', name: 'call-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     let lastArgExits = [call.info.id + '-name'];
     for (const arg of args) {
         if (arg === undefined) {
@@ -297,7 +302,7 @@ function cfgFunctionCall(call, name, args) {
 function cfgArgumentOrParameter(node, name, value) {
     const graph = new ControlFlowGraph();
     const info = { graph, breaks: [], nexts: [], returns: [], exitPoints: [node.info.id + '-exit'], entryPoints: [node.info.id] };
-    graph.addVertex({ id: node.info.id, name: node.type });
+    graph.addVertex({ id: node.info.id, name: node.type, type: "expression" /* CfgVertexType.Expression */ });
     let currentExitPoint = [node.info.id];
     if (name) {
         graph.merge(name.graph);
@@ -309,7 +314,7 @@ function cfgArgumentOrParameter(node, name, value) {
         }
         currentExitPoint = name.exitPoints;
     }
-    graph.addVertex({ id: node.info.id + '-before-value', name: 'before-value' });
+    graph.addVertex({ id: node.info.id + '-before-value', name: 'before-value', type: "mid-marker" /* CfgVertexType.MidMarker */ });
     for (const exitPoints of currentExitPoint) {
         graph.addEdge(node.info.id + '-before-value', exitPoints, { label: 'FD' });
     }
@@ -326,7 +331,7 @@ function cfgArgumentOrParameter(node, name, value) {
         }
         currentExitPoint = value.exitPoints;
     }
-    graph.addVertex({ id: node.info.id + '-exit', name: 'exit' });
+    graph.addVertex({ id: node.info.id + '-exit', name: 'exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     for (const exit of currentExitPoint) {
         graph.addEdge(node.info.id + '-exit', exit, { label: 'FD' });
     }
@@ -335,8 +340,8 @@ function cfgArgumentOrParameter(node, name, value) {
 function cfgBinaryOp(binOp, lhs, rhs) {
     const graph = new ControlFlowGraph().merge(lhs.graph).merge(rhs.graph);
     const result = { graph, breaks: [...lhs.breaks, ...rhs.breaks], nexts: [...lhs.nexts, ...rhs.nexts], returns: [...lhs.returns, ...rhs.returns], entryPoints: [binOp.info.id], exitPoints: [binOp.info.id + '-exit'] };
-    graph.addVertex({ id: binOp.info.id, name: binOp.type });
-    graph.addVertex({ id: binOp.info.id + '-exit', name: 'binOp-exit' });
+    graph.addVertex({ id: binOp.info.id, name: binOp.type, type: binOp.flavor === 'assignment' ? "statement" /* CfgVertexType.Statement */ : "expression" /* CfgVertexType.Expression */ });
+    graph.addVertex({ id: binOp.info.id + '-exit', name: 'binOp-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     for (const exitPoint of lhs.exitPoints) {
         for (const entryPoint of rhs.entryPoints) {
             result.graph.addEdge(entryPoint, exitPoint, { label: 'FD' });
@@ -353,8 +358,8 @@ function cfgBinaryOp(binOp, lhs, rhs) {
 function cfgAccess(access, name, accessors) {
     const result = name;
     const graph = result.graph;
-    graph.addVertex({ id: access.info.id, name: access.type });
-    graph.addVertex({ id: access.info.id + '-exit', name: 'access-exit' });
+    graph.addVertex({ id: access.info.id, name: access.type, type: "expression" /* CfgVertexType.Expression */ });
+    graph.addVertex({ id: access.info.id + '-exit', name: 'access-exit', type: "end-marker" /* CfgVertexType.EndMarker */ });
     for (const entry of name.entryPoints) {
         graph.addEdge(entry, access.info.id, { label: 'FD' });
     }
@@ -383,7 +388,7 @@ function cfgAccess(access, name, accessors) {
 function cfgUnaryOp(unary, operand) {
     const graph = operand.graph;
     const result = { ...operand, graph, exitPoints: [unary.info.id] };
-    graph.addVertex({ id: unary.info.id, name: unary.type });
+    graph.addVertex({ id: unary.info.id, name: unary.type, type: "end-marker" /* CfgVertexType.EndMarker */ });
     return result;
 }
 function cfgExprList(_node, expressions) {

package/util/cfg/visitor.d.ts

@@ -0,0 +1,14 @@
+import type { NodeId } from '../../r-bridge';
+import type { CfgEdge, CfgVertex, ControlFlowInformation } from './cfg';
+export interface NodeVisitingContext {
+    parent: {
+        vertex: NodeId;
+        edge: CfgEdge;
+    } | 'root';
+    cfg: ControlFlowInformation;
+    visited: Set<NodeId>;
+    /** contains the current vertex as well */
+    siblings: NodeId[];
+}
+export type OnEnterVisitNode = (node: CfgVertex, context: NodeVisitingContext) => void;
+export declare function visitCfg(cfg: ControlFlowInformation, onVisit: OnEnterVisitNode): void;