Commit 7: Implement POSIX collating symbol and equivalence class support

Add posixCollatingSymbol([.x.]) and posixEquivalenceClass([=x=]) predicates
to ParseRegExp.qll, enabling correct tokenization of all three POSIX bracket
atom types inside character classes:
  - [:name:]  — already handled, unchanged
  - [.x.]     — new: posixCollatingSymbol predicate
  - [=x=]     — new: posixEquivalenceClass predicate

Refactored:
- charSetDelimiter: use inAnyPosixBracket helper (covers all three types)
- charSet closing: use inAnyPosixBracket helper
- inPosixBracket: updated to cover all three types
- simpleCharacter: updated to exclude all three types
- namedCharacterProperty: includes posixCollatingSymbol and posixEquivalenceClass
- inAnyPosixBracket: new private helper; uses pos in [x..y-1] for QL binding

Before this commit (commit 6):
  [[.a.]] → RegExpCharacterClass([[.a.]) + stray ]   (WRONG)
  [[=a=]] → RegExpCharacterClass([[=a=]) + stray ]   (WRONG)

After this commit:
  [[.a.]] → RegExpCharacterClass containing RegExpNamedCharacterProperty [.a.]
  [[=a=]] → RegExpCharacterClass containing RegExpNamedCharacterProperty [=a=]

Regenerated parse.expected and regexp.expected via codeql test run --learn
(CodeQL 2.26.1); all 2 tests pass.

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

@@ -102,11 +102,9 @@ class RegExp extends StringLiteral {
     pos =
       rank[index](int p |
         (this.nonEscapedCharAt(p) = "[" or this.nonEscapedCharAt(p) = "]") and
-        // Brackets that art part of POSIX expressions should not count as
+        // Brackets that are part of POSIX expressions should not count as
         // char-set delimiters.
-        not exists(int x, int y |
-          this.posixStyleNamedCharacterProperty(x, y, _) and pos >= x and pos < y
-        )
+        not this.inAnyPosixBracket(p)
       ) and
     (
       this.nonEscapedCharAt(pos) = "[" and result = true
@@ -136,9 +134,7 @@ class RegExp extends StringLiteral {
         min(int e |
           e > innerStart and
           this.nonEscapedCharAt(e) = "]" and
-          not exists(int x, int y |
-            this.posixStyleNamedCharacterProperty(x, y, _) and e >= x and e < y
-          )
+          not this.inAnyPosixBracket(e)
         |
           e
         )
@@ -289,7 +285,9 @@ class RegExp extends StringLiteral {
   /** Matches named character properties such as `\p{Word}` and `[[:digit:]]` */
   predicate namedCharacterProperty(int start, int end, string name) {
     this.pStyleNamedCharacterProperty(start, end, name) or
-    this.posixStyleNamedCharacterProperty(start, end, name)
+    this.posixStyleNamedCharacterProperty(start, end, name) or
+    this.posixCollatingSymbol(start, end, name) or
+    this.posixEquivalenceClass(start, end, name)
   }
 
   /** Gets the name of the character property in start,end */
@@ -297,6 +295,21 @@ class RegExp extends StringLiteral {
     this.namedCharacterProperty(start, end, result)
   }
 
+  /**
+   * Holds if the position `pos` falls inside any POSIX bracket atom
+   * (`[:name:]`, `[.x.]`, or `[=x=]`).
+   * Used to prevent the inner brackets from being treated as charSet delimiters.
+   */
+  private predicate inAnyPosixBracket(int pos) {
+    exists(int x, int y |
+      this.posixStyleNamedCharacterProperty(x, y, _) or
+      this.posixCollatingSymbol(x, y, _) or
+      this.posixEquivalenceClass(x, y, _)
+    |
+      pos in [x .. y - 1]
+    )
+  }
+
   /** Matches a POSIX bracket expression such as `[:alnum:]` within a character class. */
   private predicate posixStyleNamedCharacterProperty(int start, int end, string name) {
     this.getChar(start) = "[" and
@@ -318,6 +331,42 @@ class RegExp extends StringLiteral {
     )
   }
 
+  /**
+   * Matches a POSIX collating symbol such as `[.a.]` within a character class.
+   * Example: `[[.a.]]` — the `[.a.]` atom nested inside the outer bracket.
+   */
+  private predicate posixCollatingSymbol(int start, int end, string name) {
+    this.getChar(start) = "[" and
+    this.getChar(start + 1) = "." and
+    end =
+      min(int e |
+        e > start and
+        this.getChar(e - 2) = "." and
+        this.getChar(e - 1) = "]"
+      |
+        e
+      ) and
+    name = this.getText().substring(start + 2, end - 2)
+  }
+
+  /**
+   * Matches a POSIX equivalence class such as `[=a=]` within a character class.
+   * Example: `[[=a=]]` — the `[=a=]` atom nested inside the outer bracket.
+   */
+  private predicate posixEquivalenceClass(int start, int end, string name) {
+    this.getChar(start) = "[" and
+    this.getChar(start + 1) = "=" and
+    end =
+      min(int e |
+        e > start and
+        this.getChar(e - 2) = "=" and
+        this.getChar(e - 1) = "]"
+      |
+        e
+      ) and
+    name = this.getText().substring(start + 2, end - 2)
+  }
+
   /**
    * Matches named character properties. For example:
    * - `\p{Space}`
@@ -398,11 +447,17 @@ class RegExp extends StringLiteral {
   }
 
   /**
-   * Holds if the character at `index` is inside a posix bracket.
+   * Holds if the character at `index` is inside a posix bracket
+   * (`[:name:]`, `[.x.]`, or `[=x=]`).
    */
   predicate inPosixBracket(int index) {
     exists(int x, int y |
-      this.posixStyleNamedCharacterProperty(x, y, _) and index in [x + 1 .. y - 2]
+      (
+        this.posixStyleNamedCharacterProperty(x, y, _) or
+        this.posixCollatingSymbol(x, y, _) or
+        this.posixEquivalenceClass(x, y, _)
+      ) and
+      index in [x + 1 .. y - 2]
     )
   }
 
@@ -413,11 +468,7 @@ class RegExp extends StringLiteral {
     end = start + 1 and
     not this.charSet(start, _) and
     not this.charSet(_, start + 1) and
-    not exists(int x, int y |
-      this.posixStyleNamedCharacterProperty(x, y, _) and
-      start >= x and
-      end <= y
-    ) and
+    not this.inAnyPosixBracket(start) and
     exists(string c | c = this.getChar(start) |
       exists(int x, int y, int z |
         this.charSet(x, z) and

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

@@ -245,45 +245,15 @@ test.cpp:
 
 #   76| [RegExpNamedCharacterProperty] [:space:]
 
-#   79| [RegExpCharacterClass] [[.a.]
-#-----| 0 -> [RegExpConstant, RegExpNormalChar] [
-#-----| 1 -> [RegExpConstant, RegExpNormalChar] .
-#-----| 2 -> [RegExpConstant, RegExpNormalChar] a
-#-----| 3 -> [RegExpConstant, RegExpNormalChar] .
+#   79| [RegExpCharacterClass] [[.a.]]
+#-----| 0 -> [RegExpNamedCharacterProperty] [.a.]
 
-#   79| [RegExpSequence] [[.a.]]
-#-----| 0 -> [RegExpCharacterClass] [[.a.]
-#-----| 1 -> [RegExpConstant, RegExpNormalChar] ]
+#   79| [RegExpNamedCharacterProperty] [.a.]
 
-#   79| [RegExpConstant, RegExpNormalChar] [
+#   82| [RegExpCharacterClass] [[=a=]]
+#-----| 0 -> [RegExpNamedCharacterProperty] [=a=]
 
-#   79| [RegExpConstant, RegExpNormalChar] .
-
-#   79| [RegExpConstant, RegExpNormalChar] a
-
-#   79| [RegExpConstant, RegExpNormalChar] .
-
-#   79| [RegExpConstant, RegExpNormalChar] ]
-
-#   82| [RegExpCharacterClass] [[=a=]
-#-----| 0 -> [RegExpConstant, RegExpNormalChar] [
-#-----| 1 -> [RegExpConstant, RegExpNormalChar] =
-#-----| 2 -> [RegExpConstant, RegExpNormalChar] a
-#-----| 3 -> [RegExpConstant, RegExpNormalChar] =
-
-#   82| [RegExpSequence] [[=a=]]
-#-----| 0 -> [RegExpCharacterClass] [[=a=]
-#-----| 1 -> [RegExpConstant, RegExpNormalChar] ]
-
-#   82| [RegExpConstant, RegExpNormalChar] [
-
-#   82| [RegExpConstant, RegExpNormalChar] =
-
-#   82| [RegExpConstant, RegExpNormalChar] a
-
-#   82| [RegExpConstant, RegExpNormalChar] =
-
-#   82| [RegExpConstant, RegExpNormalChar] ]
+#   82| [RegExpNamedCharacterProperty] [=a=]
 
 #   85| [RegExpCharacterClass] [[:alpha:]0-9]
 #-----| 0 -> [RegExpNamedCharacterProperty] [:alpha:]

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

@@ -100,20 +100,10 @@ term
 | test.cpp:76:29:76:40 | [a[:space:]] | RegExpCharacterClass |
 | test.cpp:76:30:76:30 | a | RegExpConstant,RegExpNormalChar |
 | test.cpp:76:31:76:39 | [:space:] | RegExpNamedCharacterProperty |
-| test.cpp:79:28:79:33 | [[.a.] | RegExpCharacterClass |
-| test.cpp:79:28:79:34 | [[.a.]] | RegExpSequence |
-| test.cpp:79:29:79:29 | [ | RegExpConstant,RegExpNormalChar |
-| test.cpp:79:30:79:30 | . | RegExpConstant,RegExpNormalChar |
-| test.cpp:79:31:79:31 | a | RegExpConstant,RegExpNormalChar |
-| test.cpp:79:32:79:32 | . | RegExpConstant,RegExpNormalChar |
-| test.cpp:79:34:79:34 | ] | RegExpConstant,RegExpNormalChar |
-| test.cpp:82:29:82:34 | [[=a=] | RegExpCharacterClass |
-| test.cpp:82:29:82:35 | [[=a=]] | RegExpSequence |
-| test.cpp:82:30:82:30 | [ | RegExpConstant,RegExpNormalChar |
-| test.cpp:82:31:82:31 | = | RegExpConstant,RegExpNormalChar |
-| test.cpp:82:32:82:32 | a | RegExpConstant,RegExpNormalChar |
-| test.cpp:82:33:82:33 | = | RegExpConstant,RegExpNormalChar |
-| test.cpp:82:35:82:35 | ] | RegExpConstant,RegExpNormalChar |
+| test.cpp:79:28:79:34 | [[.a.]] | RegExpCharacterClass |
+| test.cpp:79:29:79:33 | [.a.] | RegExpNamedCharacterProperty |
+| test.cpp:82:29:82:35 | [[=a=]] | RegExpCharacterClass |
+| test.cpp:82:30:82:34 | [=a=] | RegExpNamedCharacterProperty |
 | test.cpp:85:29:85:42 | [[:alpha:]0-9] | RegExpCharacterClass |
 | test.cpp:85:30:85:38 | [:alpha:] | RegExpNamedCharacterProperty |
 | test.cpp:85:39:85:39 | 0 | RegExpConstant,RegExpNormalChar |
@@ -252,16 +242,6 @@ regExpNormalCharValue
 | test.cpp:57:28:57:28 | f | f |
 | test.cpp:57:30:57:33 | A-F] | A-F] |
 | test.cpp:76:30:76:30 | a | a |
-| test.cpp:79:29:79:29 | [ | [ |
-| test.cpp:79:30:79:30 | . | . |
-| test.cpp:79:31:79:31 | a | a |
-| test.cpp:79:32:79:32 | . | . |
-| test.cpp:79:34:79:34 | ] | ] |
-| test.cpp:82:30:82:30 | [ | [ |
-| test.cpp:82:31:82:31 | = | = |
-| test.cpp:82:32:82:32 | a | a |
-| test.cpp:82:33:82:33 | = | = |
-| test.cpp:82:35:82:35 | ] | ] |
 | test.cpp:85:39:85:39 | 0 | 0 |
 | test.cpp:85:41:85:41 | 9 | 9 |
 | test.cpp:89:23:89:24 | \\w | w |