Merge pull request github#12838 from jcogs33/jcogs33/add-class-for-callables-interesting-for-modeling

Java: add class that represents callables that are interesting for MaD models

Author: jcogs33

java/ql/lib/semmle/code/java/Member.qll

@@ -340,6 +340,47 @@ class Callable extends StmtParent, Member, @callable {
   }
 }
 
+/**
+ * Holds if the given type is public and, if it is a nested type, that all of
+ * its enclosing types are public as well.
+ */
+private predicate veryPublic(RefType t) {
+  t.isPublic() and
+  (
+    not t instanceof NestedType or
+    veryPublic(t.(NestedType).getEnclosingType())
+  )
+}
+
+/** A callable that is the same as its source declaration. */
+class SrcCallable extends Callable {
+  SrcCallable() { this.isSourceDeclaration() }
+
+  /**
+   * Holds if this callable is effectively public in the sense that it can be
+   * called from outside the codebase. This means either a `public` callable on
+   * a sufficiently public type or a `protected` callable on a sufficiently
+   * public non-`final` type.
+   */
+  predicate isEffectivelyPublic() {
+    exists(RefType t | t = this.getDeclaringType() |
+      this.isPublic() and veryPublic(t)
+      or
+      this.isProtected() and not t.isFinal() and veryPublic(t)
+    )
+    or
+    exists(SrcRefType tsub, Method m |
+      veryPublic(tsub) and
+      tsub.hasMethod(m, _) and
+      m.getSourceDeclaration() = this
+    |
+      this.isPublic()
+      or
+      this.isProtected() and not tsub.isFinal()
+    )
+  }
+}
+
 /** Gets the erasure of `t1` if it is a raw type, or `t1` itself otherwise. */
 private Type eraseRaw(Type t1) {
   if t1 instanceof RawType then result = t1.getErasure() else result = t1

java/ql/lib/semmle/code/java/dataflow/internal/ModelExclusions.qll

@@ -0,0 +1,77 @@
+/** Provides classes and predicates for exclusions related to MaD models. */
+
+import java
+
+/** Holds if the given package `p` is a test package. */
+pragma[nomagic]
+private predicate isTestPackage(Package p) {
+  p.getName()
+      .matches([
+          "org.junit%", "junit.%", "org.mockito%", "org.assertj%",
+          "com.github.tomakehurst.wiremock%", "org.hamcrest%", "org.springframework.test.%",
+          "org.springframework.mock.%", "org.springframework.boot.test.%", "reactor.test%",
+          "org.xmlunit%", "org.testcontainers.%", "org.opentest4j%", "org.mockserver%",
+          "org.powermock%", "org.skyscreamer.jsonassert%", "org.rnorth.visibleassertions",
+          "org.openqa.selenium%", "com.gargoylesoftware.htmlunit%", "org.jboss.arquillian.testng%",
+          "org.testng%"
+        ])
+}
+
+/**
+ * A test library.
+ */
+class TestLibrary extends RefType {
+  TestLibrary() { isTestPackage(this.getPackage()) }
+}
+
+/** Holds if the given file is a test file. */
+private predicate isInTestFile(File file) {
+  file.getAbsolutePath().matches(["%/test/%", "%/guava-tests/%", "%/guava-testlib/%"]) and
+  not file.getAbsolutePath().matches("%/ql/test/%") // allows our test cases to work
+}
+
+/** Holds if the given compilation unit's package is a JDK internal. */
+private predicate isJdkInternal(CompilationUnit cu) {
+  cu.getPackage().getName().matches("org.graalvm%") or
+  cu.getPackage().getName().matches("com.sun%") or
+  cu.getPackage().getName().matches("sun%") or
+  cu.getPackage().getName().matches("jdk%") or
+  cu.getPackage().getName().matches("java2d%") or
+  cu.getPackage().getName().matches("build.tools%") or
+  cu.getPackage().getName().matches("propertiesparser%") or
+  cu.getPackage().getName().matches("org.jcp%") or
+  cu.getPackage().getName().matches("org.w3c%") or
+  cu.getPackage().getName().matches("org.ietf.jgss%") or
+  cu.getPackage().getName().matches("org.xml.sax%") or
+  cu.getPackage().getName().matches("com.oracle%") or
+  cu.getPackage().getName().matches("org.omg%") or
+  cu.getPackage().getName().matches("org.relaxng%") or
+  cu.getPackage().getName() = "compileproperties" or
+  cu.getPackage().getName() = "transparentruler" or
+  cu.getPackage().getName() = "genstubs" or
+  cu.getPackage().getName() = "netscape.javascript" or
+  cu.getPackage().getName() = ""
+}
+
+/** Holds if the given callable is not worth modeling. */
+predicate isUninterestingForModels(Callable c) {
+  isInTestFile(c.getCompilationUnit().getFile()) or
+  isJdkInternal(c.getCompilationUnit()) or
+  c instanceof MainMethod or
+  c instanceof StaticInitializer or
+  exists(FunctionalExpr funcExpr | c = funcExpr.asMethod()) or
+  c.getDeclaringType() instanceof TestLibrary or
+  c.(Constructor).isParameterless()
+}
+
+/**
+ * A class that represents all callables for which we might be
+ * interested in having a MaD model.
+ */
+class ModelApi extends SrcCallable {
+  ModelApi() {
+    this.fromSource() and
+    this.isEffectivelyPublic() and
+    not isUninterestingForModels(this)
+  }
+}

java/ql/src/Metrics/Summaries/GeneratedVsManualCoverageQuery.qll

@@ -1,9 +1,9 @@
 private import semmle.code.java.dataflow.FlowSummary
-private import utils.modelgenerator.internal.CaptureModels
+private import semmle.code.java.dataflow.internal.ModelExclusions
 private import TopJdkApis
 
 /**
- * Returns the number of `DataFlowTargetApi`s with Summary MaD models
+ * Returns the number of `ModelApi`s with Summary MaD models
  * for a given package and provenance.
  */
 bindingset[package, apiSubset]
@@ -13,7 +13,7 @@ private int getNumMadModeledApis(string package, string provenance, string apiSu
     count(SummarizedCallable sc |
       callableSubset(sc.asCallable(), apiSubset) and
       package = sc.asCallable().getCompilationUnit().getPackage().getName() and
-      sc.asCallable() instanceof DataFlowTargetApi and
+      sc.asCallable() instanceof ModelApi and
       (
         // "auto-only"
         not sc.hasManualModel() and
@@ -34,12 +34,12 @@ private int getNumMadModeledApis(string package, string provenance, string apiSu
     )
 }
 
-/** Returns the total number of `DataFlowTargetApi`s for a given package. */
+/** Returns the total number of `ModelApi`s for a given package. */
 private int getNumApis(string package, string apiSubset) {
   result =
-    strictcount(DataFlowTargetApi dataFlowTargApi |
-      callableSubset(dataFlowTargApi, apiSubset) and
-      package = dataFlowTargApi.getCompilationUnit().getPackage().getName()
+    strictcount(ModelApi api |
+      callableSubset(api, apiSubset) and
+      package = api.getCompilationUnit().getPackage().getName()
     )
 }
 
@@ -70,7 +70,7 @@ predicate modelCoverageGenVsMan(
     // calculate the total generated and total manual numbers
     generated = generatedOnly + both and
     manual = manualOnly + both and
-    // count the total number of `DataFlowTargetApi`s for each package
+    // count the total number of `ModelApi`s for each package
     all = getNumApis(package, apiSubset) and
     non = all - (generatedOnly + both + manualOnly) and
     // Proportion of coverage

java/ql/src/Telemetry/ExternalApi.qll

@@ -8,27 +8,7 @@ private import semmle.code.java.dataflow.FlowSummary
 private import semmle.code.java.dataflow.internal.DataFlowPrivate
 private import semmle.code.java.dataflow.internal.FlowSummaryImpl as FlowSummaryImpl
 private import semmle.code.java.dataflow.TaintTracking
-
-pragma[nomagic]
-private predicate isTestPackage(Package p) {
-  p.getName()
-      .matches([
-          "org.junit%", "junit.%", "org.mockito%", "org.assertj%",
-          "com.github.tomakehurst.wiremock%", "org.hamcrest%", "org.springframework.test.%",
-          "org.springframework.mock.%", "org.springframework.boot.test.%", "reactor.test%",
-          "org.xmlunit%", "org.testcontainers.%", "org.opentest4j%", "org.mockserver%",
-          "org.powermock%", "org.skyscreamer.jsonassert%", "org.rnorth.visibleassertions",
-          "org.openqa.selenium%", "com.gargoylesoftware.htmlunit%", "org.jboss.arquillian.testng%",
-          "org.testng%"
-        ])
-}
-
-/**
- * A test library.
- */
-private class TestLibrary extends RefType {
-  TestLibrary() { isTestPackage(this.getPackage()) }
-}
+private import semmle.code.java.dataflow.internal.ModelExclusions
 
 private string containerAsJar(Container container) {
   if container instanceof JarFile then result = container.getBaseName() else result = "rt.jar"

java/ql/src/utils/modelgenerator/internal/CaptureModelsSpecific.qll

@@ -5,6 +5,7 @@
 private import java as J
 private import semmle.code.java.dataflow.internal.DataFlowPrivate
 private import semmle.code.java.dataflow.internal.ContainerFlow as ContainerFlow
+private import semmle.code.java.dataflow.internal.ModelExclusions
 private import semmle.code.java.dataflow.DataFlow as Df
 private import semmle.code.java.dataflow.SSA as Ssa
 private import semmle.code.java.dataflow.TaintTracking as Tt
@@ -26,33 +27,6 @@ private J::Method superImpl(J::Method m) {
   not m instanceof J::ToStringMethod
 }
 
-private predicate isInTestFile(J::File file) {
-  file.getAbsolutePath().matches(["%/test/%", "%/guava-tests/%", "%/guava-testlib/%"]) and
-  not file.getAbsolutePath().matches("%/ql/test/%") // allows our test cases to work
-}
-
-private predicate isJdkInternal(J::CompilationUnit cu) {
-  cu.getPackage().getName().matches("org.graalvm%") or
-  cu.getPackage().getName().matches("com.sun%") or
-  cu.getPackage().getName().matches("sun%") or
-  cu.getPackage().getName().matches("jdk%") or
-  cu.getPackage().getName().matches("java2d%") or
-  cu.getPackage().getName().matches("build.tools%") or
-  cu.getPackage().getName().matches("propertiesparser%") or
-  cu.getPackage().getName().matches("org.jcp%") or
-  cu.getPackage().getName().matches("org.w3c%") or
-  cu.getPackage().getName().matches("org.ietf.jgss%") or
-  cu.getPackage().getName().matches("org.xml.sax%") or
-  cu.getPackage().getName().matches("com.oracle%") or
-  cu.getPackage().getName().matches("org.omg%") or
-  cu.getPackage().getName().matches("org.relaxng%") or
-  cu.getPackage().getName() = "compileproperties" or
-  cu.getPackage().getName() = "transparentruler" or
-  cu.getPackage().getName() = "genstubs" or
-  cu.getPackage().getName() = "netscape.javascript" or
-  cu.getPackage().getName() = ""
-}
-
 private predicate isInfrequentlyUsed(J::CompilationUnit cu) {
   cu.getPackage().getName().matches("javax.swing%") or
   cu.getPackage().getName().matches("java.awt%")
@@ -62,13 +36,8 @@ private predicate isInfrequentlyUsed(J::CompilationUnit cu) {
  * Holds if it is relevant to generate models for `api`.
  */
 private predicate isRelevantForModels(J::Callable api) {
-  not isInTestFile(api.getCompilationUnit().getFile()) and
-  not isJdkInternal(api.getCompilationUnit()) and
-  not isInfrequentlyUsed(api.getCompilationUnit()) and
-  not api instanceof J::MainMethod and
-  not api instanceof J::StaticInitializer and
-  not exists(J::FunctionalExpr funcExpr | api = funcExpr.asMethod()) and
-  not api.(J::Constructor).isParameterless()
+  not isUninterestingForModels(api) and
+  not isInfrequentlyUsed(api.getCompilationUnit())
 }
 
 /**

java/ql/test/query-tests/Metrics/GeneratedVsManualCoverage/TopJdkApisTest/TopJdkApis/java/lang/AbstractStringBuilder.java

@@ -7,8 +7,8 @@ abstract class AbstractStringBuilder implements Appendable, CharSequence
 {
     public char charAt(int p0){ return '0'; } // manual summary
     public int length(){ return 0; } // manual neutral
-    public void setCharAt(int p0, char p1){} // manual neutral, Note: not currently counted by query due to exclusions in `TargetApiSpecific`
-    public void setLength(int p0){} // manual neutral, Note: not currently counted by query due to exclusions in `TargetApiSpecific`
+    public void setCharAt(int p0, char p1){} // manual neutral
+    public void setLength(int p0){} // manual neutral
 
 
     public AbstractStringBuilder append(CharSequence p0){ return null; }

java/ql/test/query-tests/Metrics/GeneratedVsManualCoverage/TopJdkApisTest/TopJdkApis/java/lang/Enum.java

@@ -7,7 +7,7 @@
 abstract public class Enum<E extends Enum<E>> implements Comparable<E>, Serializable
 {
     protected Enum() {}
-    protected Enum(String p0, int p1){} // manual neutral, Note: this will not be counted in query results since `protected` not `public`
+    protected Enum(String p0, int p1){} // manual neutral
     public String toString(){ return null; } // manual neutral
     public final String name(){ return null; } // manual neutral
     public final boolean equals(Object p0){ return false; } // manual neutral

java/ql/test/query-tests/Metrics/GeneratedVsManualCoverage/TopJdkApisTest/TopJdkApis/java/lang/StringBuffer.java

@@ -12,7 +12,4 @@ public class StringBuffer extends AbstractStringBuilder implements Serializable
     public StringBuffer append(char p0){ return null; } // manual summary
 
     public StringBuffer append(CharSequence s, int start, int end) { return null; }
-
-    public void setCharAt(int p0, char p1){}
-    public void setLength(int p0){}
 }

java/ql/test/query-tests/Metrics/GeneratedVsManualCoverage/TopJdkApisTest/TopJdkApis/java/lang/StringBuilder.java

@@ -19,7 +19,4 @@ public StringBuilder(String p0){} // manual summary
     public StringBuilder(int p0){} // manual summary
 
     public StringBuilder append(CharSequence s, int start, int end) { return null; }
-
-    public void setCharAt(int p0, char p1){}
-    public void setLength(int p0){}
 }

java/ql/test/query-tests/Metrics/GeneratedVsManualCoverage/TopJdkApisTest/TopJdkApisTest.expected

@@ -1,21 +1,23 @@
-| java.io | 0 | 0 | 22 | 9 | 31 | 0.7096774193548387 | 0.0 | 0.7096774193548387 | 0.0 | NaN | 0.2903225806451613 |
-| java.lang | 0 | 0 | 60 | 89 | 149 | 0.40268456375838924 | 0.0 | 0.40268456375838924 | 0.0 | NaN | 0.5973154362416108 |
+| java.awt | 0 | 0 | 2 | 1 | 3 | 0.6666666666666666 | 0.0 | 0.6666666666666666 | 0.0 | NaN | 0.3333333333333333 |
+| java.io | 0 | 0 | 22 | 15 | 37 | 0.5945945945945946 | 0.0 | 0.5945945945945946 | 0.0 | NaN | 0.40540540540540543 |
+| java.lang | 0 | 0 | 62 | 94 | 156 | 0.3974358974358974 | 0.0 | 0.3974358974358974 | 0.0 | NaN | 0.6025641025641025 |
 | java.lang.invoke | 0 | 0 | 0 | 1 | 1 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.lang.reflect | 0 | 0 | 0 | 4 | 4 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.math | 0 | 0 | 0 | 16 | 16 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.net | 0 | 0 | 5 | 0 | 5 | 1.0 | 0.0 | 1.0 | 0.0 | NaN | 0.0 |
 | java.nio | 0 | 0 | 2 | 3 | 5 | 0.4 | 0.0 | 0.4 | 0.0 | NaN | 0.6 |
 | java.nio.charset | 0 | 0 | 0 | 1 | 1 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
-| java.nio.file | 0 | 0 | 1 | 1 | 2 | 0.5 | 0.0 | 0.5 | 0.0 | NaN | 0.5 |
-| java.sql | 0 | 0 | 0 | 2 | 2 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
+| java.nio.file | 0 | 0 | 7 | 1 | 8 | 0.875 | 0.0 | 0.875 | 0.0 | NaN | 0.125 |
+| java.sql | 0 | 0 | 2 | 14 | 16 | 0.125 | 0.0 | 0.125 | 0.0 | NaN | 0.875 |
 | java.text | 0 | 0 | 0 | 5 | 5 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.time | 0 | 0 | 0 | 17 | 17 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.time.chrono | 0 | 0 | 0 | 1 | 1 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.time.format | 0 | 0 | 0 | 2 | 2 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
-| java.util | 0 | 0 | 56 | 45 | 101 | 0.5544554455445545 | 0.0 | 0.5544554455445545 | 0.0 | NaN | 0.44554455445544555 |
-| java.util.concurrent | 0 | 0 | 6 | 7 | 13 | 0.46153846153846156 | 0.0 | 0.46153846153846156 | 0.0 | NaN | 0.5384615384615384 |
+| java.util | 0 | 0 | 84 | 68 | 152 | 0.5526315789473685 | 0.0 | 0.5526315789473685 | 0.0 | NaN | 0.4473684210526316 |
+| java.util.concurrent | 0 | 0 | 9 | 9 | 18 | 0.5 | 0.0 | 0.5 | 0.0 | NaN | 0.5 |
 | java.util.concurrent.atomic | 0 | 0 | 2 | 11 | 13 | 0.15384615384615385 | 0.0 | 0.15384615384615385 | 0.0 | NaN | 0.8461538461538461 |
-| java.util.function | 0 | 0 | 0 | 1 | 1 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
+| java.util.concurrent.locks | 0 | 0 | 0 | 2 | 2 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
+| java.util.function | 0 | 0 | 0 | 6 | 6 | 0.0 | 0.0 | 0.0 | NaN | NaN | 1.0 |
 | java.util.logging | 0 | 0 | 1 | 1 | 2 | 0.5 | 0.0 | 0.5 | 0.0 | NaN | 0.5 |
 | java.util.regex | 0 | 0 | 3 | 1 | 4 | 0.75 | 0.0 | 0.75 | 0.0 | NaN | 0.25 |
-| java.util.stream | 0 | 0 | 4 | 5 | 9 | 0.4444444444444444 | 0.0 | 0.4444444444444444 | 0.0 | NaN | 0.5555555555555556 |
+| java.util.stream | 0 | 0 | 18 | 8 | 26 | 0.6923076923076923 | 0.0 | 0.6923076923076923 | 0.0 | NaN | 0.3076923076923077 |