From 2f01e133b9064cf35c6d0a2664996a017be4f4f3 Mon Sep 17 00:00:00 2001
From: "copilot-swe-agent[bot]" <198982749+Copilot@users.noreply.github.com>
Date: Tue, 21 Jul 2026 12:27:34 +0000
Subject: [PATCH] Commit 8: Complete location-test coverage for u8 and u8R
 forms

Add u8"..." and u8R"(...)" cases to test_locations in test.cpp, guarded
by #if __cplusplus >= 201703L. Add an options file forcing
-std=c++17 so the u8 char8_t-era rows are actually extracted.

Plain "..." rows are unchanged. New u8/u8R rows in locations.expected
show correct columns computed by regexpContentOffset: offset 3 for
`u8"` and offset 5 for `u8R"(`. No change to regexpContentOffset was
needed.

Bundle: github/codeql-action codeql-bundle-v2.26.1 (CodeQL CLI 2.26.1).
---
 .../library-tests/regex/locations.expected    |  8 +++++++
 cpp/ql/test/library-tests/regex/options       |  1 +
 .../test/library-tests/regex/parse.expected   | 22 +++++++++++++++++++
 .../test/library-tests/regex/regexp.expected  | 12 ++++++++++
 cpp/ql/test/library-tests/regex/test.cpp      |  8 +++++++
 5 files changed, 51 insertions(+)
 create mode 100644 cpp/ql/test/library-tests/regex/options

diff --git a/cpp/ql/test/library-tests/regex/locations.expected b/cpp/ql/test/library-tests/regex/locations.expected
index a70822fa3a6..8b3a01df017 100644
--- a/cpp/ql/test/library-tests/regex/locations.expected
+++ b/cpp/ql/test/library-tests/regex/locations.expected
@@ -94,3 +94,11 @@
 | test.cpp:196:22:196:25 | a\\.b | 21 | 0 | 4 | 22 | 25 |
 | test.cpp:196:23:196:24 | \\. | 21 | 1 | 3 | 23 | 24 |
 | test.cpp:196:25:196:25 | b | 21 | 3 | 4 | 25 | 25 |
+| test.cpp:200:22:200:22 | a | 19 | 0 | 1 | 22 | 22 |
+| test.cpp:200:22:200:23 | a+ | 19 | 0 | 2 | 22 | 23 |
+| test.cpp:200:22:200:24 | a+b | 19 | 0 | 3 | 22 | 24 |
+| test.cpp:200:24:200:24 | b | 19 | 2 | 3 | 24 | 24 |
+| test.cpp:203:28:203:28 | a | 23 | 0 | 1 | 28 | 28 |
+| test.cpp:203:28:203:29 | a+ | 23 | 0 | 2 | 28 | 29 |
+| test.cpp:203:28:203:30 | a+b | 23 | 0 | 3 | 28 | 30 |
+| test.cpp:203:30:203:30 | b | 23 | 2 | 3 | 30 | 30 |
diff --git a/cpp/ql/test/library-tests/regex/options b/cpp/ql/test/library-tests/regex/options
new file mode 100644
index 00000000000..2ebedec9a4c
--- /dev/null
+++ b/cpp/ql/test/library-tests/regex/options
@@ -0,0 +1 @@
+semmle-extractor-options: -std=c++17
diff --git a/cpp/ql/test/library-tests/regex/parse.expected b/cpp/ql/test/library-tests/regex/parse.expected
index b1ee43d1748..f7361ec71c4 100644
--- a/cpp/ql/test/library-tests/regex/parse.expected
+++ b/cpp/ql/test/library-tests/regex/parse.expected
@@ -861,3 +861,25 @@ test.cpp:
 #  196| [RegExpConstant, RegExpEscape] \.
 
 #  196| [RegExpConstant, RegExpNormalChar] b
+
+#  200| [RegExpConstant, RegExpNormalChar] a
+
+#  200| [RegExpPlus] a+
+#-----| 0 -> [RegExpConstant, RegExpNormalChar] a
+
+#  200| [RegExpSequence] a+b
+#-----| 0 -> [RegExpPlus] a+
+#-----| 1 -> [RegExpConstant, RegExpNormalChar] b
+
+#  200| [RegExpConstant, RegExpNormalChar] b
+
+#  203| [RegExpConstant, RegExpNormalChar] a
+
+#  203| [RegExpPlus] a+
+#-----| 0 -> [RegExpConstant, RegExpNormalChar] a
+
+#  203| [RegExpSequence] a+b
+#-----| 0 -> [RegExpPlus] a+
+#-----| 1 -> [RegExpConstant, RegExpNormalChar] b
+
+#  203| [RegExpConstant, RegExpNormalChar] b
diff --git a/cpp/ql/test/library-tests/regex/regexp.expected b/cpp/ql/test/library-tests/regex/regexp.expected
index cdb750c90fb..cb83e11287f 100644
--- a/cpp/ql/test/library-tests/regex/regexp.expected
+++ b/cpp/ql/test/library-tests/regex/regexp.expected
@@ -323,6 +323,14 @@ term
 | test.cpp:196:22:196:25 | a\\.b | RegExpSequence |
 | test.cpp:196:23:196:24 | \\. | RegExpConstant,RegExpEscape |
 | test.cpp:196:25:196:25 | b | RegExpConstant,RegExpNormalChar |
+| test.cpp:200:22:200:22 | a | RegExpConstant,RegExpNormalChar |
+| test.cpp:200:22:200:23 | a+ | RegExpPlus |
+| test.cpp:200:22:200:24 | a+b | RegExpSequence |
+| test.cpp:200:24:200:24 | b | RegExpConstant,RegExpNormalChar |
+| test.cpp:203:28:203:28 | a | RegExpConstant,RegExpNormalChar |
+| test.cpp:203:28:203:29 | a+ | RegExpPlus |
+| test.cpp:203:28:203:30 | a+b | RegExpSequence |
+| test.cpp:203:30:203:30 | b | RegExpConstant,RegExpNormalChar |
 regExpNormalCharValue
 | test.cpp:33:21:33:23 | abc | abc |
 | test.cpp:36:22:36:22 | a | a |
@@ -461,3 +469,7 @@ regExpNormalCharValue
 | test.cpp:196:22:196:22 | a | a |
 | test.cpp:196:23:196:24 | \\. | . |
 | test.cpp:196:25:196:25 | b | b |
+| test.cpp:200:22:200:22 | a | a |
+| test.cpp:200:24:200:24 | b | b |
+| test.cpp:203:28:203:28 | a | a |
+| test.cpp:203:30:203:30 | b | b |
diff --git a/cpp/ql/test/library-tests/regex/test.cpp b/cpp/ql/test/library-tests/regex/test.cpp
index 37cb61d4a63..93aace2760c 100644
--- a/cpp/ql/test/library-tests/regex/test.cpp
+++ b/cpp/ql/test/library-tests/regex/test.cpp
@@ -194,4 +194,12 @@ void test_locations() {
 
   // Escape with dot — "a\\.b" value is a\.b; offset 1 correct
   std::regex r_esc2("a\\.b");
+
+  // u8 encoding prefix — content offset 3 (u8" = 3 chars). Requires C++17+.
+#if __cplusplus >= 201703L
+  std::regex r_u8(u8"a+b");
+
+  // u8R raw — content offset 5 (u8R"( = 5 chars).
+  std::regex r_u8_raw(u8R"(a+b)");
+#endif
 }