C++: Make regex library compile

cpp/ql/lib/qlpack.yml

@@ -10,6 +10,7 @@ dependencies:
   codeql/mad: ${workspace}
   codeql/quantum: ${workspace}
   codeql/rangeanalysis: ${workspace}
+  codeql/regex: ${workspace}
   codeql/ssa: ${workspace}
   codeql/typeflow: ${workspace}
   codeql/tutorial: ${workspace}

cpp/ql/lib/semmle/code/cpp/regex/RegexTreeView.qll

@@ -1,17 +1,17 @@
-/** Provides a class hierarchy corresponding to a parse tree of regular expressions. */
+/**
+ * Provides a class hierarchy corresponding to a parse tree of C++ regular expressions.
+ */
 
 private import internal.ParseRegExp
 private import codeql.util.Numbers
-private import codeql.ruby.ast.Literal as Ast
-private import codeql.Locations
-private import codeql.regex.nfa.NfaUtils as NfaUtils
+private import semmle.code.cpp.exprs.Literal
 private import codeql.regex.RegexTreeView
 // exporting as RegexTreeView, and in the top-level scope.
 import Impl as RegexTreeView
 import Impl
 
 /** Gets the parse tree resulting from parsing `re`, if such has been constructed. */
-RegExpTerm getParsedRegExp(Ast::RegExpLiteral re) {
+RegExpTerm getParsedRegExp(StringLiteral re) {
   result.getRegExp() = re and result.isRootTerm()
 }
 
@@ -55,7 +55,7 @@ private newtype TRegExpParent =
     re.namedCharacterProperty(start, end, _)
   }
 
-/** An implementation that statisfies the RegexTreeView signature. */
+/** An implementation that satisfies the RegexTreeView signature. */
 private module Impl implements RegexTreeViewSig {
   /**
    * An element containing a regular expression term, that is, either
@@ -211,24 +211,14 @@ private module Impl implements RegexTreeViewSig {
      */
     Location getLocation() { result = re.getLocation() }
 
-    pragma[noinline]
-    private predicate componentHasLocationInfo(
-      int i, string filepath, int startline, int startcolumn, int endline, int endcolumn
-    ) {
-      re.getComponent(i)
-          .getLocation()
-          .hasLocationInfo(filepath, startline, startcolumn, endline, endcolumn)
-    }
-
     /** Holds if this term is found at the specified location offsets. */
     predicate hasLocationInfo(
       string filepath, int startline, int startcolumn, int endline, int endcolumn
     ) {
       exists(int re_start |
-        this.componentHasLocationInfo(0, filepath, startline, re_start, _, _) and
-        this.componentHasLocationInfo(re.getNumberOfComponents() - 1, filepath, _, _, endline, _) and
-        startcolumn = re_start + start and
-        endcolumn = re_start + end - 1
+        re.getLocation().hasLocationInfo(filepath, startline, re_start, endline, _) and
+        startcolumn = re_start + 1 + start and
+        endcolumn = re_start + 1 + end - 1
       )
     }
 
@@ -312,7 +302,7 @@ private module Impl implements RegexTreeViewSig {
       result.getEnd() = part_end
     }
 
-    /** Hols if this term may match an unlimited number of times. */
+    /** Hodls if this term may match an unlimited number of times. */
     predicate mayRepeatForever() { may_repeat_forever = true }
 
     /** Gets the qualifier for this term. That is e.g "?" for "a?". */
@@ -660,9 +650,9 @@ private module Impl implements RegexTreeViewSig {
    *
    * Examples:
    *
-   * ```rb
-   * /[a-fA-F0-9]/
-   * /[^abc]/
+   * ```cpp
+   * "[a-fA-F0-9]"
+   * "[^abc]"
    * ```
    */
   class RegExpCharacterClass extends RegExpTerm, TRegExpCharacterClass {
@@ -1195,9 +1185,7 @@ private module Impl implements RegexTreeViewSig {
   /**
    * Holds if the regular expression should not be considered.
    */
-  predicate isExcluded(RegExpParent parent) {
-    parent.(RegExpTerm).getRegExp().(Ast::RegExpLiteral).hasFreeSpacingFlag() // exclude free-spacing mode regexes
-  }
+  predicate isExcluded(RegExpParent parent) { none() }
 
   /**
    * Holds if `term` is a possessive quantifier.
@@ -1207,13 +1195,13 @@ private module Impl implements RegexTreeViewSig {
 
   /**
    * Holds if the regex that `term` is part of is used in a way that ignores any leading prefix of the input it's matched against.
-   * Not yet implemented for Ruby.
+   * Not yet implemented for C++.
    */
   predicate matchesAnyPrefix(RegExpTerm term) { any() }
 
   /**
    * Holds if the regex that `term` is part of is used in a way that ignores any trailing suffix of the input it's matched against.
-   * Not yet implemented for Ruby.
+   * Not yet implemented for C++.
    */
   predicate matchesAnySuffix(RegExpTerm term) { any() }
 

cpp/ql/lib/semmle/code/cpp/regex/internal/ParseRegExp.qll

@@ -1,19 +1,18 @@
 /**
- * Library for parsing for Ruby regular expressions.
+ * Library for parsing for C++ regular expressions.
  *
  * N.B. does not yet handle stripping whitespace and comments in regexes with
  * the `x` (free-spacing) flag.
  */
 
-private import codeql.ruby.AST as Ast
-private import codeql.Locations
+private import semmle.code.cpp.exprs.Literal
 
 /**
- * A `StringlikeLiteral` containing a regular expression term, that is, either
+ * A C++ string literal containing a regular expression term, that is, either
  * a regular expression literal, or a string literal used in a context where
- * it is parsed as regular expression.
+ * it is parsed as a regular expression.
  */
-abstract class RegExp extends Ast::StringlikeLiteral {
+abstract class RegExp extends StringLiteral {
   /**
    * Holds if this `RegExp` has the `s` flag for multi-line matching.
    */
@@ -254,11 +253,7 @@ abstract class RegExp extends Ast::StringlikeLiteral {
   }
 
   /** Gets the text of this regex */
-  string getText() {
-    exists(Ast::ConstantValue c | c = this.getConstantValue() |
-      result = [this.getConstantValue().getString(), this.getConstantValue().getRegExp()]
-    )
-  }
+  string getText() { result = this.getValue() }
 
   /** Gets the `i`th character of this regex */
   string getChar(int i) { result = this.getText().charAt(i) }

cpp/ql/test/library-tests/regex/parse.ql

@@ -2,10 +2,9 @@
  * @kind graph
  */
 
-import codeql.Locations
-import codeql.ruby.Regexp as RE
+import semmle.code.cpp.regex.RegexTreeView
 
-query predicate nodes(RE::RegExpTerm n, string attr, string val) {
+query predicate nodes(RegExpTerm n, string attr, string val) {
   attr = "semmle.label" and
   val = "[" + concat(n.getAPrimaryQlClass(), ", ") + "] " + n.toString()
   or
@@ -13,7 +12,7 @@ query predicate nodes(RE::RegExpTerm n, string attr, string val) {
   val =
     any(int i |
       n =
-        rank[i](RE::RegExpTerm t, string fp, int sl, int sc, int el, int ec |
+        rank[i](RegExpTerm t, string fp, int sl, int sc, int el, int ec |
           t.hasLocationInfo(fp, sl, sc, el, ec)
         |
           t order by fp, sl, sc, el, ec, t.toString()
@@ -21,7 +20,7 @@ query predicate nodes(RE::RegExpTerm n, string attr, string val) {
     ).toString()
 }
 
-query predicate edges(RE::RegExpTerm pred, RE::RegExpTerm succ, string attr, string val) {
+query predicate edges(RegExpTerm pred, RegExpTerm succ, string attr, string val) {
   attr in ["semmle.label", "semmle.order"] and
   val = any(int i | succ = pred.getChild(i)).toString()
 }

cpp/ql/test/library-tests/regex/regexp.ql

@@ -1,4 +1,4 @@
-import codeql.ruby.Regexp
+import semmle.code.cpp.regex.RegexTreeView
 
 query predicate groupName(RegExpGroup g, string name) { name = g.getName() }