Merge pull request #21017 from github/idrissrio/cpp/overlay/xml

Author: IdrissRio

config/identical-files.json

@@ -282,6 +282,7 @@
     "java/ql/lib/semmle/code/java/internal/OverlayXml.qll",
     "go/ql/lib/semmle/go/internal/OverlayXml.qll",
     "python/ql/lib/semmle/python/internal/OverlayXml.qll",
-    "csharp/ql/lib/semmle/code/csharp/internal/OverlayXml.qll"
+    "csharp/ql/lib/semmle/code/csharp/internal/OverlayXml.qll",
+    "cpp/ql/lib/semmle/code/cpp/internal/OverlayXml.qll"
   ]
 }

cpp/ql/lib/semmle/code/cpp/internal/Overlay.qll

@@ -2,6 +2,8 @@
  * Defines entity discard predicates for C++ overlay analysis.
  */
 
+private import OverlayXml
+
 /**
  * Holds always for the overlay variant and never for the base variant.
  * This local predicate is used to define local predicates that behave

cpp/ql/lib/semmle/code/cpp/internal/OverlayXml.qll

@@ -0,0 +1,46 @@
+overlay[local]
+module;
+
+/**
+ * A local predicate that always holds for the overlay variant and never holds for the base variant.
+ * This is used to define local predicates that behave differently for the base and overlay variant.
+ */
+private predicate isOverlay() { databaseMetadata("isOverlay", "true") }
+
+private string getXmlFile(@xmllocatable locatable) {
+  exists(@location_default location, @file file | xmllocations(locatable, location) |
+    locations_default(location, file, _, _, _, _) and
+    files(file, result)
+  )
+}
+
+private string getXmlFileInBase(@xmllocatable locatable) {
+  not isOverlay() and
+  result = getXmlFile(locatable)
+}
+
+/**
+ * Holds if the given `file` was extracted as part of the overlay and was extracted by the HTML/XML
+ * extractor.
+ */
+private predicate overlayXmlExtracted(string file) {
+  isOverlay() and
+  exists(@xmllocatable locatable |
+    not files(locatable, _) and not xmlNs(locatable, _, _, _) and file = getXmlFile(locatable)
+  )
+}
+
+/**
+ * Holds if the given XML `locatable` should be discarded, because it is part of the overlay base
+ * and is in a file that was also extracted as part of the overlay database.
+ */
+overlay[discard_entity]
+private predicate discardXmlLocatable(@xmllocatable locatable) {
+  exists(string file | file = getXmlFileInBase(locatable) |
+    overlayChangedFiles(file)
+    or
+    // The HTML/XML extractor is currently not incremental and may extract more files than those
+    // included in overlayChangedFiles.
+    overlayXmlExtracted(file)
+  )
+}