@eagleoutice/flowr 2.9.6 → 2.9.8

package/README.md

@@ -24,7 +24,7 @@ It offers a wide variety of features, for example:
     
     ```shell
     $ docker run -it --rm eagleoutice/flowr # or npm run flowr 
-    flowR repl using flowR v2.9.5, R grammar v14 (tree-sitter engine)
+    flowR repl using flowR v2.9.7, R grammar v14 (tree-sitter engine)
     R> :query @linter "read.csv(\"/root/x.txt\")"
     ```
     
@@ -33,15 +33,15 @@ It offers a wide variety of features, for example:
     
     
     ```text
-    Query: linter (3 ms)
+    Query: linter (2 ms)
        ╰ Deprecated Functions (deprecated-functions):
-           ╰ Metadata: totalCalls: 0, totalFunctionDefinitions: 0, searchTimeMs: 1, processTimeMs: 0
+           ╰ Metadata: totalCalls: 0, totalFunctionDefinitions: 0, searchTimeMs: 0, processTimeMs: 0
        ╰ File Path Validity (file-path-validity):
            ╰ certain:
                ╰ Path `/root/x.txt` at 1.1-23
-           ╰ Metadata: totalReads: 1, totalUnknown: 0, totalWritesBeforeAlways: 0, totalValid: 0, searchTimeMs: 0, processTimeMs: 0
+           ╰ Metadata: totalReads: 1, totalUnknown: 0, totalWritesBeforeAlways: 0, totalValid: 0, searchTimeMs: 1, processTimeMs: 0
        ╰ Seeded Randomness (seeded-randomness):
-           ╰ Metadata: consumerCalls: 0, callsWithFunctionProducers: 0, callsWithAssignmentProducers: 0, callsWithNonConstantProducers: 0, callsWithOtherBranchProducers: 0, searchTimeMs: 1, processTimeMs: 0
+           ╰ Metadata: consumerCalls: 0, callsWithFunctionProducers: 0, callsWithAssignmentProducers: 0, callsWithNonConstantProducers: 0, callsWithOtherBranchProducers: 0, searchTimeMs: 0, processTimeMs: 0
        ╰ Absolute Paths (absolute-file-paths):
            ╰ certain:
                ╰ Path `/root/x.txt` at 1.1-23
@@ -53,12 +53,12 @@ It offers a wide variety of features, for example:
        ╰ Network Functions (network-functions):
            ╰ Metadata: totalCalls: 0, totalFunctionDefinitions: 0, searchTimeMs: 0, processTimeMs: 0
        ╰ Dataframe Access Validation (dataframe-access-validation):
-           ╰ Metadata: numOperations: 0, numAccesses: 0, totalAccessed: 0, searchTimeMs: 0, processTimeMs: 1
+           ╰ Metadata: numOperations: 0, numAccesses: 0, totalAccessed: 0, searchTimeMs: 1, processTimeMs: 0
        ╰ Dead Code (dead-code):
            ╰ Metadata: consideredNodes: 5, searchTimeMs: 0, processTimeMs: 0
        ╰ Useless Loops (useless-loop):
            ╰ Metadata: numOfUselessLoops: 0, searchTimeMs: 0, processTimeMs: 0
-    All queries together required ≈3 ms (1ms accuracy, total 3 ms)
+    All queries together required ≈2 ms (1ms accuracy, total 3 ms)
     ```
     
     
@@ -92,7 +92,7 @@ It offers a wide variety of features, for example:
        ╰ **Absolute Paths** (absolute-file-paths):\
            ╰ certain:\
                ╰ Path `/root/x.txt` at 1.1-23\
-    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>totalConsidered: 1, totalUnknown: 0, searchTimeMs: 0, processTimeMs: 0</code>\
+    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>totalConsidered: 1, totalUnknown: 0, searchTimeMs: 1, processTimeMs: 0</code>\
     &nbsp;&nbsp;&nbsp;╰ **Unused Definitions** (unused-definitions):\
     &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>totalConsidered: 0, searchTimeMs: 0, processTimeMs: 0</code>\
     &nbsp;&nbsp;&nbsp;╰ **Naming Convention** (naming-convention):\
@@ -100,16 +100,16 @@ It offers a wide variety of features, for example:
     &nbsp;&nbsp;&nbsp;╰ **Network Functions** (network-functions):\
     &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>totalCalls: 0, totalFunctionDefinitions: 0, searchTimeMs: 0, processTimeMs: 0</code>\
     &nbsp;&nbsp;&nbsp;╰ **Dataframe Access Validation** (dataframe-access-validation):\
-    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>numOperations: 0, numAccesses: 0, totalAccessed: 0, searchTimeMs: 0, processTimeMs: 0</code>\
+    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>numOperations: 0, numAccesses: 0, totalAccessed: 0, searchTimeMs: 0, processTimeMs: 1</code>\
     &nbsp;&nbsp;&nbsp;╰ **Dead Code** (dead-code):\
-    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>consideredNodes: 5, searchTimeMs: 1, processTimeMs: 0</code>\
+    &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>consideredNodes: 5, searchTimeMs: 0, processTimeMs: 0</code>\
     &nbsp;&nbsp;&nbsp;╰ **Useless Loops** (useless-loop):\
     &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;╰ _Metadata_: <code>numOfUselessLoops: 0, searchTimeMs: 0, processTimeMs: 0</code>\
     _All queries together required ≈3 ms (1ms accuracy, total 3 ms)_
     
     <details> <summary style="color:gray">Show Detailed Results as Json</summary>
     
-    The analysis required _3.2 ms_ (including parsing and normalization and the query) within the generation environment.
+    The analysis required _3.3 ms_ (including parsing and normalization and the query) within the generation environment.
     
     In general, the JSON contains the Ids of the nodes in question as they are present in the normalized AST or the dataflow graph of flowR.
     Please consult the [Interface](https://github.com/flowr-analysis/flowr/wiki/Interface) wiki page for more information on how to get those.
@@ -181,7 +181,7 @@ It offers a wide variety of features, for example:
             ".meta": {
               "totalConsidered": 1,
               "totalUnknown": 0,
-              "searchTimeMs": 0,
+              "searchTimeMs": 1,
               "processTimeMs": 0
             }
           },
@@ -218,14 +218,14 @@ It offers a wide variety of features, for example:
               "numAccesses": 0,
               "totalAccessed": 0,
               "searchTimeMs": 0,
-              "processTimeMs": 0
+              "processTimeMs": 1
             }
           },
           "dead-code": {
             "results": [],
             ".meta": {
               "consideredNodes": 5,
-              "searchTimeMs": 1,
+              "searchTimeMs": 0,
               "processTimeMs": 0
             }
           },
@@ -308,7 +308,7 @@ It offers a wide variety of features, for example:
     
     ```shell
     $ docker run -it --rm eagleoutice/flowr # or npm run flowr 
-    flowR repl using flowR v2.9.5, R grammar v14 (tree-sitter engine)
+    flowR repl using flowR v2.9.7, R grammar v14 (tree-sitter engine)
     R> :query @static-slice (11@sum) file://test/testfiles/example.R
     ```
     
@@ -356,16 +356,16 @@ It offers a wide variety of features, for example:
          
 
 * 🚀 **fast call-graph, data-, and control-flow graphs**\
-  Within just [<i><span title="This measurement is automatically fetched from the latest benchmark!">99.6 ms</span></i> (as of Feb 5, 2026)](https://flowr-analysis.github.io/flowr/wiki/stats/benchmark), 
-  _flowR_ can analyze the data- and control-flow of the average real-world R script. See the [benchmarks](https://flowr-analysis.github.io/flowr/wiki/stats/benchmark) for more information,
-  and consult the [wiki pages](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph) for more details on the dataflow graphs as well as call graphs.
+  Within just [<i><span title="This measurement is automatically fetched from the latest benchmark!">111.7 ms</span></i> (as of Feb 7, 2026)](https://flowr-analysis.github.io/flowr/wiki/stats/benchmark), 
+  _flowR_ can analyze the data- and control-flow of the average real-world R&nbsp;script. See the [benchmarks](https://flowr-analysis.github.io/flowr/wiki/stats/benchmark) for more information,
+  and consult the [wiki pages](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph) for more details on the [dataflow graphs](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph) as well as [call graphs](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph#perspectives-cg).
 
     
     <details><summary>Example: Generating a dataflow graph with flowR</summary>
     
     
     You can investigate flowR's analyses using the [REPL](https://github.com/flowr-analysis/flowr/wiki/Interface#using-the-repl).
-    Commands like <span title="Description (Repl Command, starred version): Returns the URL to mermaid.live; Base Command: Get mermaid code for the dataflow graph, start with 'file://' to indicate a file (aliases: :d*, :df*)">`:dataflow*`</span> allow you to view a dataflow graph for a given R script.
+    Commands like <span title="Description (Repl Command, starred version): Returns the URL to mermaid.live; Base Command: Get mermaid code for the dataflow graph, start with 'file://' to indicate a file (aliases: :d*, :df*)">`:dataflow*`</span> allow you to view a [dataflow graph](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph) for a given R script.
     
     Let's have a look at the following example:
     
@@ -386,13 +386,13 @@ It offers a wide variety of features, for example:
     ```
     
     
-    To get the dataflow graph for this script, you can use the following command:
+    To get the [dataflow graph](https://github.com/flowr-analysis/flowr/wiki/wiki/dataflow-graph) for this script, you can use the following command:
     
     
     
     ```shell
     $ docker run -it --rm eagleoutice/flowr # or npm run flowr 
-    flowR repl using flowR v2.9.5, R grammar v14 (tree-sitter engine)
+    flowR repl using flowR v2.9.7, R grammar v14 (tree-sitter engine)
     R> :dataflow* test/testfiles/example.R
     ```
     
@@ -697,7 +697,7 @@ It offers a wide variety of features, for example:
     ```
     
     	
-    (The analysis required _4.3 ms_ (including parse and normalize, using the [tree-sitter](https://github.com/flowr-analysis/flowr/wiki/Engines) engine) within the generation environment.)
+    (The analysis required _1.8 ms_ (including parse and normalize, using the [tree-sitter](https://github.com/flowr-analysis/flowr/wiki/Engines) engine) within the generation environment.)
     
     
     
@@ -724,8 +724,8 @@ If you are already using flowR and want to give feedback, please consider fillin
  
 ## ⭐ Getting Started
 
-To get started with _flowR_ and its features, please check out the [Overview](https://github.com/flowr-analysis/flowr/wiki/Overview) wiki page. 
-The [Setup](https://github.com/flowr-analysis/flowr/wiki/Setup) wiki page explains how you can download and setup _flowR_ on your system. 
+To get started with _flowR_ and its features, please check out the [Overview](https://github.com/flowr-analysis/flowr/wiki/wiki/overview) wiki page. 
+The [Setup](https://github.com/flowr-analysis/flowr/wiki/wiki/setup) wiki page explains how you can download and setup _flowR_ on your system. 
 With docker&nbsp;🐳️, the following line should be enough (and drop you directly into the read-eval-print loop):
 
 
@@ -753,7 +753,9 @@ If you want to use the same commands:
 ## 📜 More Information
 
 For more details on how to use _flowR_ please refer to the [wiki pages](https://github.com/flowr-analysis/flowr/wiki),
-as well as the deployed [code documentation](https://flowr-analysis.github.io/flowr/doc/).
+as well as the deployed [code documentation](https://flowr-analysis.github.io/flowr/docs).
+To cite flowR, please check out the publications below. To specifically refer to the source code, 
+please check out flowR's [Zenodo archive](https://zenodo.org/doi/10.5281/zenodo.13319290).
 
 ## 📃 Publications on flowR
 
@@ -860,7 +862,7 @@ please check out the following publications (if you find that a paper is missing
 
 ## 🚀 Contributing
 
-We welcome every contribution! Please check out the [developer onboarding](https://github.com/flowr-analysis/flowr/wiki/Onboarding) section in the wiki for all the information you will need.
+We welcome every contribution! Please check out the [developer onboarding](https://github.com/flowr-analysis/flowr/wiki/wiki/onboarding) section in the wiki for all the information you will need.
 
 ### Contributors
 

package/abstract-interpretation/absint-visitor.d.ts

@@ -1,4 +1,5 @@
-import type { CfgSimpleVertex, ControlFlowInformation } from '../control-flow/control-flow-graph';
+import type { ControlFlowInformation } from '../control-flow/control-flow-graph';
+import { CfgVertex } from '../control-flow/control-flow-graph';
 import type { SemanticCfgGuidedVisitorConfiguration } from '../control-flow/semantic-cfg-guided-visitor';
 import { SemanticCfgGuidedVisitor } from '../control-flow/semantic-cfg-guided-visitor';
 import { BuiltInProcName } from '../dataflow/environments/built-in';
@@ -98,15 +99,15 @@ export declare abstract class AbstractInterpretationVisitor<Domain extends AnyAb
      * Checks whether to continue visiting the control flow graph after a widening point.
      * By default, we only continue visiting if the widening point is visited for the first time or the abstract state at the widening point changed.
      */
-    protected shouldContinueVisiting(wideningPoint: CfgSimpleVertex, oldState: StateAbstractDomain<Domain>): boolean;
+    protected shouldContinueVisiting(wideningPoint: CfgVertex, oldState: StateAbstractDomain<Domain>): boolean;
     /**
      * Checks whether a control flow graph vertex should be skipped during visitation.
      * By default, we only process vertices of leaf nodes and exit vertices (no entry nodes of complex nodes).
      */
-    protected shouldSkipVertex(vertex: CfgSimpleVertex): boolean;
+    protected shouldSkipVertex(vertex: CfgVertex): boolean;
     /**
      * Whether widening should be performed at a widening point.
      * By default, we perform widening when the number of visitation of the widening point reaches the widening threshold of the config.
      */
-    protected shouldWiden(wideningPoint: CfgSimpleVertex): boolean;
+    protected shouldWiden(wideningPoint: CfgVertex): boolean;
 }

package/abstract-interpretation/absint-visitor.js

@@ -100,7 +100,7 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
      */
     getEndState() {
         const exitPoints = this.config.controlFlow.exitPoints.map(id => this.getCfgVertex(id)).filter(assert_1.isNotUndefined);
-        const exitNodes = exitPoints.map(vertex => (0, control_flow_graph_1.getVertexRootId)(vertex)).filter(assert_1.isNotUndefined);
+        const exitNodes = exitPoints.map(vertex => control_flow_graph_1.CfgVertex.getRootId(vertex)).filter(assert_1.isNotUndefined);
         const states = exitNodes.map(node => this.getAbstractState(node)).filter(assert_1.isNotUndefined);
         return this.config.domain.bottom().joinAll(states);
     }
@@ -121,14 +121,14 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
         if (vertex === undefined) {
             return true;
         }
-        const nodeId = (0, control_flow_graph_1.getVertexRootId)(vertex);
+        const nodeId = control_flow_graph_1.CfgVertex.getRootId(vertex);
         if (this.isWideningPoint(nodeId)) {
             // only check widening points at the entry vertex
-            if (vertex.type === control_flow_graph_1.CfgVertexType.EndMarker) {
+            if (control_flow_graph_1.CfgVertex.isMarker(vertex)) {
                 return true;
             }
             const oldState = this.getAbstractState(nodeId) ?? this.config.domain.bottom();
-            const predecessorDomains = this.getPredecessorNodes(vertex.id).map(pred => this.getAbstractState(pred)).filter(assert_1.isNotUndefined);
+            const predecessorDomains = this.getPredecessorNodes(control_flow_graph_1.CfgVertex.getId(vertex)).map(pred => this.getAbstractState(pred)).filter(assert_1.isNotUndefined);
             this.currentState = this.config.domain.bottom().joinAll(predecessorDomains);
             if (this.shouldWiden(vertex)) {
                 this.currentState = oldState.widen(this.currentState);
@@ -139,7 +139,7 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
         else if (this.shouldSkipVertex(vertex)) {
             return true;
         }
-        const predecessorDomains = this.getPredecessorNodes(vertex.id).map(pred => this.getAbstractState(pred)).filter(assert_1.isNotUndefined);
+        const predecessorDomains = this.getPredecessorNodes(control_flow_graph_1.CfgVertex.getId(vertex)).map(pred => this.getAbstractState(pred)).filter(assert_1.isNotUndefined);
         this.currentState = this.config.domain.bottom().joinAll(predecessorDomains);
         this.onVisitNode(vertexId);
         // discard the inferred abstract state when encountering functions with unknown side effects (e.g. `eval`)
@@ -211,10 +211,10 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
                 return [];
             }
             else if (this.shouldSkipVertex(vertex)) {
-                return this.getPredecessorNodes(vertex.id);
+                return this.getPredecessorNodes(control_flow_graph_1.CfgVertex.getId(vertex));
             }
             else {
-                return [(0, control_flow_graph_1.getVertexRootId)(vertex)];
+                return [control_flow_graph_1.CfgVertex.getRootId(vertex)];
             }
         })
             .toArray() ?? [];
@@ -248,8 +248,9 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
      * By default, we only continue visiting if the widening point is visited for the first time or the abstract state at the widening point changed.
      */
     shouldContinueVisiting(wideningPoint, oldState) {
-        const visitedCount = this.visited.get(wideningPoint.id) ?? 0;
-        this.visited.set(wideningPoint.id, visitedCount + 1);
+        const wid = control_flow_graph_1.CfgVertex.getId(wideningPoint);
+        const visitedCount = this.visited.get(wid) ?? 0;
+        this.visited.set(wid, visitedCount + 1);
         return visitedCount === 0 || !oldState.equals(this.currentState);
     }
     /**
@@ -257,14 +258,14 @@ class AbstractInterpretationVisitor extends semantic_cfg_guided_visitor_1.Semant
      * By default, we only process vertices of leaf nodes and exit vertices (no entry nodes of complex nodes).
      */
     shouldSkipVertex(vertex) {
-        return vertex.type !== control_flow_graph_1.CfgVertexType.EndMarker && vertex.end !== undefined;
+        return !control_flow_graph_1.CfgVertex.isMarker(vertex) && !control_flow_graph_1.CfgVertex.isBlock(vertex) && control_flow_graph_1.CfgVertex.getEnd(vertex) !== undefined;
     }
     /**
      * Whether widening should be performed at a widening point.
      * By default, we perform widening when the number of visitation of the widening point reaches the widening threshold of the config.
      */
     shouldWiden(wideningPoint) {
-        return (this.visited.get(wideningPoint.id) ?? 0) >= this.config.ctx.config.abstractInterpretation.wideningThreshold;
+        return (this.visited.get(control_flow_graph_1.CfgVertex.getId(wideningPoint)) ?? 0) >= this.config.ctx.config.abstractInterpretation.wideningThreshold;
     }
 }
 exports.AbstractInterpretationVisitor = AbstractInterpretationVisitor;

package/abstract-interpretation/data-frame/dataframe-domain.js

@@ -1,7 +1,6 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.DataFrameDomain = void 0;
-const lattice_1 = require("../domains/lattice");
 const positive_interval_domain_1 = require("../domains/positive-interval-domain");
 const product_domain_1 = require("../domains/product-domain");
 const set_range_domain_1 = require("../domains/set-range-domain");
@@ -53,7 +52,7 @@ class DataFrameDomain extends product_domain_1.ProductDomain {
     }
     static refine(value) {
         if (value.colnames.isValue() && value.cols.isValue()) {
-            if (value.colnames.value.range === lattice_1.Top && value.colnames.value.min.size >= value.cols.value[1]) {
+            if (value.colnames.value.min.size >= value.cols.value[1]) {
                 value.colnames = value.colnames.meet({ min: new Set(), range: value.colnames.value.min });
             }
             if (value.colnames.isValue()) {

package/benchmark/slicer.d.ts

@@ -57,7 +57,6 @@ export declare class BenchmarkSlicer {
     private readonly perSliceMeasurements;
     private readonly deltas;
     private readonly parserName;
-    private config;
     private context;
     private stats;
     private loadedXml;

package/benchmark/slicer.js

@@ -41,7 +41,6 @@ class BenchmarkSlicer {
     perSliceMeasurements = new Map();
     deltas = new Map();
     parserName;
-    config;
     context;
     stats;
     loadedXml;
@@ -63,7 +62,6 @@ class BenchmarkSlicer {
      */
     async init(request, config, autoSelectIf, threshold) {
         (0, assert_1.guard)(this.stats === undefined, 'cannot initialize the slicer twice');
-        this.config = config;
         // we know these are in sync so we just cast to one of them
         this.parser = await this.commonMeasurements.measure('initialize R session', async () => {
             if (this.parserName === 'r-shell') {
@@ -227,11 +225,8 @@ class BenchmarkSlicer {
         exports.benchmarkLogger.trace('try to extract the control flow graph');
         this.guardActive();
         (0, assert_1.guard)(this.normalizedAst !== undefined, 'normalizedAst should be defined for control flow extraction');
-        (0, assert_1.guard)(this.dataflow !== undefined, 'dataflow should be defined for control flow extraction');
-        (0, assert_1.guard)(this.config !== undefined, 'config should be defined for control flow extraction');
         const ast = this.normalizedAst;
-        const dfg = this.dataflow.graph;
-        this.controlFlow = this.measureSimpleStep('extract control flow graph', () => (0, extract_cfg_1.extractCfg)(ast, this.context, dfg));
+        this.controlFlow = this.measureSimpleStep('extract control flow graph', () => (0, extract_cfg_1.extractCfg)(ast, this.context, undefined, undefined, true));
     }
     /**
      * Infer the shape of data frames using abstract interpretation with {@link inferDataFrameShapes}
@@ -264,6 +259,7 @@ class BenchmarkSlicer {
         this.measureSimpleStep('infer data frame shapes', () => inference.start());
         const result = inference.getEndState();
         stats.numberOfResultConstraints = result.value.size;
+        stats.sizeOfInfo = (0, size_of_1.safeSizeOf)([inference.getAbstractTrace()]);
         for (const value of result.value.values()) {
             if (value.isTop()) {
                 stats.numberOfResultingTop++;
@@ -283,10 +279,8 @@ class BenchmarkSlicer {
                 stats.numberOfEmptyNodes++;
                 return;
             }
-            const state = inference.getAbstractState(node.info.id);
-            stats.sizeOfInfo += (0, size_of_1.safeSizeOf)([state]);
             const nodeStats = {
-                numberOfEntries: state?.value.size ?? 0
+                numberOfEntries: inference.getAbstractState(node.info.id)?.value.size ?? 0
             };
             if (operations !== undefined) {
                 nodeStats.mappedOperations = operations.map(op => op.operation);

package/benchmark/stats/print.js

@@ -268,6 +268,7 @@ Dataframe shape inference:
   Number of abstract value nodes: ${formatSummarizedMeasure(stats.dataFrameShape.numberOfValueNodes)}
   Number of entries per node:     ${formatSummarizedMeasure(stats.dataFrameShape.numberOfEntriesPerNode)}
   Number of operations:           ${formatSummarizedMeasure(stats.dataFrameShape.numberOfOperations)}
+  Number of total exact:          ${formatSummarizedMeasure(stats.dataFrameShape.numberOfTotalExact)}
   Number of total values:         ${formatSummarizedMeasure(stats.dataFrameShape.numberOfTotalValues)}
   Number of total top:            ${formatSummarizedMeasure(stats.dataFrameShape.numberOfTotalTop)}
   Inferred column names per node: ${formatSummarizedMeasure(stats.dataFrameShape.inferredColNames)}

package/benchmark/summarizer/data.d.ts

@@ -83,6 +83,8 @@ export interface UltimateSlicerStats {
 export interface SummarizedDfShapeStats<T = number> extends Omit<SlicerStatsDfShape<T>, 'perNodeStats'> {
     numberOfEntriesPerNode: SummarizedMeasurement;
     numberOfOperations: T;
+    numberOfTotalConstraints: T;
+    numberOfTotalExact: T;
     numberOfTotalValues: T;
     numberOfTotalBottom: T;
     numberOfTotalTop: T;

package/benchmark/summarizer/first-phase/process.js

@@ -265,8 +265,10 @@ function summarizeDfShapeStats({ perNodeStats, ...stats }) {
         ...stats,
         numberOfEntriesPerNode: (0, summarizer_1.summarizeMeasurement)(nodeStats.map(s => s.numberOfEntries)),
         numberOfOperations: (0, arrays_1.arraySum)(nodeStats.map(s => s.mappedOperations?.length).filter(assert_1.isNotUndefined)),
+        numberOfTotalConstraints: nodeStats.filter(s => s.inferredColNames !== undefined && s.inferredColCount !== undefined && s.inferredRowCount !== undefined).length,
+        numberOfTotalExact: nodeStats.filter(s => s.approxRangeColNames === 0 && s.approxRangeColCount === 0 && s.approxRangeRowCount === 0).length,
         numberOfTotalValues: nodeStats.filter(s => isValue(s.inferredColNames) && isValue(s.inferredColCount) && isValue(s.inferredRowCount)).length,
-        numberOfTotalBottom: nodeStats.filter(s => s.inferredColNames === 0 && isBottom(s.inferredColCount) && isBottom(s.inferredRowCount)).length,
+        numberOfTotalBottom: nodeStats.filter(s => isBottom(s.inferredColCount) && isBottom(s.inferredColCount) && isBottom(s.inferredRowCount)).length,
         numberOfTotalTop: nodeStats.filter(s => isTop(s.inferredColNames) && isTop(s.inferredColCount) && isTop(s.inferredRowCount)).length,
         inferredColNames: (0, summarizer_1.summarizeMeasurement)(nodeStats.map(s => s.inferredColNames).filter(isValue)),
         approxRangeColNames: (0, summarizer_1.summarizeMeasurement)(nodeStats.map(s => s.approxRangeColNames).filter(assert_1.isNotUndefined).filter(isFinite)),

package/benchmark/summarizer/second-phase/process.js

@@ -119,6 +119,8 @@ function summarizeAllSummarizedStats(stats) {
             sizeOfInfo: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.sizeOfInfo).filter(assert_1.isNotUndefined)),
             numberOfEntriesPerNode: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfEntriesPerNode).filter(assert_1.isNotUndefined)),
             numberOfOperations: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfOperations).filter(assert_1.isNotUndefined)),
+            numberOfTotalConstraints: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalConstraints).filter(assert_1.isNotUndefined)),
+            numberOfTotalExact: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalExact).filter(assert_1.isNotUndefined)),
             numberOfTotalValues: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalValues).filter(assert_1.isNotUndefined)),
             numberOfTotalBottom: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalBottom).filter(assert_1.isNotUndefined)),
             numberOfTotalTop: (0, summarizer_1.summarizeMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalTop).filter(assert_1.isNotUndefined)),
@@ -203,6 +205,8 @@ function summarizeAllUltimateStats(stats) {
             sizeOfInfo: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.sizeOfInfo).filter(assert_1.isNotUndefined)),
             numberOfEntriesPerNode: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfEntriesPerNode).filter(assert_1.isNotUndefined)),
             numberOfOperations: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfOperations).filter(assert_1.isNotUndefined)),
+            numberOfTotalConstraints: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalConstraints).filter(assert_1.isNotUndefined)),
+            numberOfTotalExact: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalExact).filter(assert_1.isNotUndefined)),
             numberOfTotalValues: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalValues).filter(assert_1.isNotUndefined)),
             numberOfTotalBottom: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalBottom).filter(assert_1.isNotUndefined)),
             numberOfTotalTop: (0, process_1.summarizeSummarizedMeasurement)(stats.map(s => s.dataFrameShape?.numberOfTotalTop).filter(assert_1.isNotUndefined)),

package/cli/repl/commands/repl-commands.d.ts

@@ -12,6 +12,7 @@ declare const _commands: {
     readonly parse: ReplCodeCommand;
     readonly normalize: ReplCodeCommand;
     readonly 'normalize*': ReplCodeCommand;
+    readonly 'normalize#': ReplCodeCommand;
     readonly dataflow: ReplCodeCommand;
     readonly 'dataflow*': ReplCodeCommand;
     readonly dataflowsimple: ReplCodeCommand;

package/cli/repl/commands/repl-commands.js

@@ -79,6 +79,7 @@ const _commands = {
     'parse': repl_parse_1.parseCommand,
     'normalize': repl_normalize_1.normalizeCommand,
     'normalize*': repl_normalize_1.normalizeStarCommand,
+    'normalize#': repl_normalize_1.normalizeHashCommand,
     'dataflow': repl_dataflow_1.dataflowCommand,
     'dataflow*': repl_dataflow_1.dataflowStarCommand,
     'dataflowsimple': repl_dataflow_1.dataflowSimplifiedCommand,

package/cli/repl/commands/repl-normalize.d.ts

@@ -1,3 +1,4 @@
 import type { ReplCodeCommand } from './repl-main';
 export declare const normalizeCommand: ReplCodeCommand;
 export declare const normalizeStarCommand: ReplCodeCommand;
+export declare const normalizeHashCommand: ReplCodeCommand;

package/cli/repl/commands/repl-normalize.js

@@ -33,11 +33,13 @@ var __importStar = (this && this.__importStar) || (function () {
     };
 })();
 Object.defineProperty(exports, "__esModule", { value: true });
-exports.normalizeStarCommand = exports.normalizeCommand = void 0;
+exports.normalizeHashCommand = exports.normalizeStarCommand = exports.normalizeCommand = void 0;
 const retriever_1 = require("../../../r-bridge/retriever");
 const ast_1 = require("../../../util/mermaid/ast");
 const ansi_1 = require("../../../util/text/ansi");
 const core_1 = require("../core");
+const defaultmap_1 = require("../../../util/collections/defaultmap");
+const visitor_1 = require("../../../r-bridge/lang-4.x/ast/model/processing/visitor");
 function formatInfo(out, type, meta) {
     return out.formatter.format(`Copied ${type} to clipboard (normalize: ${meta['.meta'].timing + 'ms'}).`, { color: 7 /* Colors.White */, effect: ansi_1.ColorEffect.Foreground, style: 3 /* FontStyles.Italic */ });
 }
@@ -79,4 +81,29 @@ exports.normalizeStarCommand = {
         catch { /* do nothing this is a service thing */ }
     }
 };
+exports.normalizeHashCommand = {
+    description: 'Returns summarization stats for the normalized AST',
+    isCodeCommand: true,
+    usageExample: ':normalize#',
+    aliases: ['n#'],
+    script: false,
+    argsParser: (args) => (0, core_1.handleString)(args),
+    fn: async ({ output, analyzer }) => {
+        const result = await analyzer.normalize();
+        const counts = new defaultmap_1.DefaultMap(() => 0);
+        let total = 0;
+        const files = result.ast.files.length;
+        (0, visitor_1.visitAst)(result.ast.files.map(f => f.root), n => {
+            counts.set(n.type, counts.get(n.type) + 1);
+            total++;
+        });
+        const num = (s, pad = 0) => output.formatter.format(s.toString().padStart(pad, ' '), { color: 6 /* Colors.Cyan */, effect: ansi_1.ColorEffect.Foreground });
+        output.stdout(output.formatter.format('Calculated in ' + result['.meta'].timing + 'ms', { color: 7 /* Colors.White */, effect: ansi_1.ColorEffect.Foreground, style: 3 /* FontStyles.Italic */ }));
+        output.stdout(`${num(total)} nodes total over ${num(files)} file(s)`);
+        const longestType = counts.keys().map(p => p.toString().length).reduce((a, b) => Math.max(a, b), 0);
+        for (const [type, count] of counts.entries().toArray().sort((a, b) => b[1] - a[1])) {
+            output.stdout(`  ${(type + ':').padEnd(longestType + 1, ' ')} ${num(count, 7)}`);
+        }
+    }
+};
 //# sourceMappingURL=repl-normalize.js.map

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

@@ -1,4 +1,4 @@
-import { type CfgBasicBlockVertex, type CfgEndMarkerVertex, type CfgExpressionVertex, type CfgSimpleVertex, type CfgStatementVertex, type ControlFlowInformation } from './control-flow-graph';
+import { type CfgBasicBlockVertex, type CfgMarkerVertex, type CfgExpressionVertex, CfgVertex, type CfgStatementVertex, type ControlFlowInformation } from './control-flow-graph';
 import type { NodeId } from '../r-bridge/lang-4.x/ast/model/processing/node-id';
 export interface BasicCfgGuidedVisitorConfiguration<ControlFlow extends ControlFlowInformation = ControlFlowInformation> {
     readonly controlFlow: ControlFlow;
@@ -28,10 +28,10 @@ export declare class BasicCfgGuidedVisitor<ControlFlow extends ControlFlowInform
     /**
      * Get the control flow vertex for the given node id or fail if it does not exist.
      */
-    protected getCfgVertex(id: NodeId): CfgSimpleVertex | undefined;
+    protected getCfgVertex(id: NodeId): CfgVertex | undefined;
     protected onVisitNode(node: NodeId): void;
     protected onBasicBlockNode(node: CfgBasicBlockVertex): void;
     protected onStatementNode(_node: CfgStatementVertex): void;
     protected onExpressionNode(_node: CfgExpressionVertex): void;
-    protected onEndMarkerNode(_node: CfgEndMarkerVertex): void;
+    protected onEndMarkerNode(_node: CfgMarkerVertex): void;
 }

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

@@ -3,7 +3,6 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.BasicCfgGuidedVisitor = void 0;
 const control_flow_graph_1 = require("./control-flow-graph");
 const assert_1 = require("../util/assert");
-const invert_cfg_1 = require("./invert-cfg");
 /**
  * In contrast to {@link visitCfgInOrder} and {@link visitCfgInReverseOrder}, this visitor is not a simple visitor
  * and serves as the basis for a variety of more complicated visiting orders of the control flow graph.
@@ -34,8 +33,7 @@ class BasicCfgGuidedVisitor {
         const graph = this.config.controlFlow.graph;
         let getNext;
         if (this.config.defaultVisitingOrder === 'forward') {
-            const inverseGraph = (0, invert_cfg_1.invertCfg)(graph);
-            getNext = (node) => inverseGraph.outgoingEdges(node)?.keys().toArray().reverse();
+            getNext = (node) => graph.ingoingEdges(node)?.keys().toArray().reverse();
         }
         else {
             getNext = (node) => graph.outgoingEdges(node)?.keys();
@@ -68,7 +66,7 @@ class BasicCfgGuidedVisitor {
         if (vertex === undefined) {
             return;
         }
-        const type = vertex.type;
+        const type = control_flow_graph_1.CfgVertex.getType(vertex);
         switch (type) {
             case control_flow_graph_1.CfgVertexType.Statement:
                 this.onStatementNode(vertex);
@@ -76,7 +74,7 @@ class BasicCfgGuidedVisitor {
             case control_flow_graph_1.CfgVertexType.Expression:
                 this.onExpressionNode(vertex);
                 break;
-            case control_flow_graph_1.CfgVertexType.EndMarker:
+            case control_flow_graph_1.CfgVertexType.Marker:
                 this.onEndMarkerNode(vertex);
                 break;
             case control_flow_graph_1.CfgVertexType.Block:
@@ -87,14 +85,15 @@ class BasicCfgGuidedVisitor {
         }
     }
     onBasicBlockNode(node) {
+        const elems = control_flow_graph_1.CfgVertex.getBasicBlockElements(node);
         if (this.config.defaultVisitingOrder === 'forward') {
-            for (const elem of node.elems.toReversed()) {
-                this.visitNode(elem.id);
+            for (const elem of elems.toReversed()) {
+                this.visitNode(control_flow_graph_1.CfgVertex.getId(elem));
             }
         }
         else {
-            for (const elem of node.elems) {
-                this.visitNode(elem.id);
+            for (const elem of elems) {
+                this.visitNode(control_flow_graph_1.CfgVertex.getId(elem));
             }
         }
     }