Merge pull request #493 from rvermeulen/rvermeulen/fix-issue-#311

A2-3-1: Exclude wide string literals and utf8 string literals

Author: knewbury01

change_notes/2024-01-17-fix-reported-fp-for-a2-3-1.md

@@ -0,0 +1,2 @@
+`A2-3-1`: ` cpp/autosar/invalid-character-in-string-literal`
+    - Fixes #311. Exclude wide string literals and utf8 string literal.

cpp/autosar/src/rules/A2-3-1/InvalidCharacterInStringLiteral.ql

@@ -17,6 +17,7 @@
 
 import cpp
 import codingstandards.cpp.autosar
+import codingstandards.cpp.Literals
 
 bindingset[s]
 string getCharOutsideBasicSourceCharSet(string s) {
@@ -27,6 +28,9 @@ string getCharOutsideBasicSourceCharSet(string s) {
 from StringLiteral s, string ch
 where
   not isExcluded(s, NamingPackage::invalidCharacterInStringLiteralQuery()) and
-  ch = getCharOutsideBasicSourceCharSet(s.getValueText())
+  ch = getCharOutsideBasicSourceCharSet(s.getValueText()) and
+  // wide string and utf8 string literals are exempted.
+  not s instanceof WideStringLiteral and
+  not s instanceof Utf8StringLiteral
 select s,
   "String literal uses the character '" + ch + "' that is outside the language basic character set."

cpp/autosar/test/rules/A2-3-1/InvalidCharacterInComment.expected

@@ -1,2 +1,2 @@
 | test.cpp:3:1:3:37 | // Invalid character \u00ce\u00b1 NON_COMPLIANT | Comment uses the character '\u00ce\u00b1' that is outside the language basic character set. |
-| test.cpp:10:1:12:2 | /*\nInvalid character \u00e2\u0086\u00a6 NON_COMPLIANT\n*/ | Comment uses the character '\u00e2\u0086\u00a6' that is outside the language basic character set. |
+| test.cpp:12:1:14:2 | /*\nInvalid character \u00e2\u0086\u00a6 NON_COMPLIANT\n*/ | Comment uses the character '\u00e2\u0086\u00a6' that is outside the language basic character set. |

cpp/autosar/test/rules/A2-3-1/InvalidCharacterInStringLiteral.expected

@@ -1 +1 @@
-| test.cpp:7:20:7:22 | \u00ce\u00b1 | String literal uses the character '\u03b1' that is outside the language basic character set. |
+| test.cpp:7:21:7:23 | \u00ce\u00b1 | String literal uses the character '\u03b1' that is outside the language basic character set. |

cpp/autosar/test/rules/A2-3-1/test.cpp

@@ -4,9 +4,15 @@
 double α = 2.;                   // NON_COMPLIANT; U+03b1
 void *to_𐆅_and_beyond = nullptr; // NON_COMPLIANT; U+10185
 int l1_\u00A8;                   // COMPLIANT[FALSE_POSITIVE]
-const char *euro = "α";          // NON_COMPLIANT
+const char *euro1 = "α";         // NON_COMPLIANT
+const wchar_t *euro2 = L"α";     // COMPLIANT
+const char *euro3 = u8"α";       // COMPLIANT
 
 int valid;
 /*
 Invalid character ↦ NON_COMPLIANT
+*/
+
+/*
+Valid character @ in comments COMPLIANT
 */

cpp/common/src/codingstandards/cpp/Literals.qll

@@ -12,3 +12,19 @@ string getTruncatedLiteralText(Literal l) {
     else result = text
   )
 }
+
+class WideStringLiteral extends StringLiteral {
+  WideStringLiteral() { this.getValueText().regexpMatch("(?s)\\s*L\".*") }
+}
+
+class Utf8StringLiteral extends StringLiteral {
+  Utf8StringLiteral() { this.getValueText().regexpMatch("(?s)\\s*u8\".*") }
+}
+
+class Utf16StringLiteral extends StringLiteral {
+  Utf16StringLiteral() { this.getValueText().regexpMatch("(?s)\\s*u\".*") }
+}
+
+class Utf32StringLiteral extends StringLiteral {
+  Utf32StringLiteral() { this.getValueText().regexpMatch("(?s)\\s*U\".*") }
+}