From d059265fa171b3b835b706b3b45e7cd596143bad Mon Sep 17 00:00:00 2001
From: Copilot <198982749+Copilot@users.noreply.github.com>
Date: Sun, 12 Oct 2025 15:36:56 +0000
Subject: [PATCH] Add explicit Unicode mode to regex patterns for proper
 Unicode handling (#40)

---
 src/lib.rs                         | 4 ++--
 src/run/commands.rs                | 9 ++++++++-
 src/run/environment.rs             | 4 +++-
 testdata/unicode_dot_regex.txt     | 3 +++
 testdata/unicode_emoji.txt         | 3 +++
 testdata/unicode_grep.txt          | 6 ++++++
 testdata/unicode_multiple.txt      | 3 +++
 testdata/unicode_regex.txt         | 3 +++
 testdata/unicode_special_chars.txt | 3 +++
 testdata/unicode_stderr.txt        | 3 +++
 testdata/unicode_stdout.txt        | 3 +++
 11 files changed, 40 insertions(+), 4 deletions(-)
 create mode 100644 testdata/unicode_dot_regex.txt
 create mode 100644 testdata/unicode_emoji.txt
 create mode 100644 testdata/unicode_grep.txt
 create mode 100644 testdata/unicode_multiple.txt
 create mode 100644 testdata/unicode_regex.txt
 create mode 100644 testdata/unicode_special_chars.txt
 create mode 100644 testdata/unicode_stderr.txt
 create mode 100644 testdata/unicode_stdout.txt

diff --git a/src/lib.rs b/src/lib.rs
index 62c0a70..21e6ab6 100644
--- a/src/lib.rs
+++ b/src/lib.rs
@@ -71,9 +71,9 @@ fn run(params: &mut RunParams, test_data_glob: &str) -> Result<()> {
             (".", test_data_glob)
         };
 
-        // Convert glob pattern to a simple matcher
+        // Convert glob pattern to a simple matcher with Unicode support
         let pattern_regex = pattern.replace("*", ".*");
-        let regex = regex::Regex::new(&format!("^{}$", pattern_regex))?;
+        let regex = regex::Regex::new(&format!("(?u)^{}$", pattern_regex))?;
 
         // Walk the directory and collect matching files
         for entry in WalkDir::new(base_dir).min_depth(1).max_depth(1) {
diff --git a/src/run/commands.rs b/src/run/commands.rs
index 0e4c8ed..64c7337 100644
--- a/src/run/commands.rs
+++ b/src/run/commands.rs
@@ -218,7 +218,14 @@ impl TestEnvironment {
 
     /// Search for a pattern in files (like grep)
     pub fn grep_files(&mut self, pattern: &str, files: &[String]) -> Result<()> {
-        let regex = Regex::new(pattern)
+        // Enable Unicode mode for proper Unicode character matching in grep patterns
+        let pattern_with_flags = if pattern.starts_with("(?") {
+            // Pattern already has flags, don't add more
+            pattern.to_string()
+        } else {
+            format!("(?u){}", pattern)
+        };
+        let regex = Regex::new(&pattern_with_flags)
             .map_err(|e| Error::command_error("grep", format!("Invalid regex: {}", e)))?;
 
         let mut _found_matches = false;
diff --git a/src/run/environment.rs b/src/run/environment.rs
index f8352f7..b7ef0ad 100644
--- a/src/run/environment.rs
+++ b/src/run/environment.rs
@@ -236,7 +236,9 @@ impl TestEnvironment {
             || expected_substituted.contains('*')
             || expected_substituted.contains('.')
         {
-            let regex_pattern = format!("(?s){}", expected_substituted); // (?s) enables DOTALL mode
+            // Enable DOTALL mode (?s) for . to match newlines
+            // Enable Unicode mode (?u) for proper Unicode character matching
+            let regex_pattern = format!("(?su){}", expected_substituted);
             let regex = Regex::new(&regex_pattern)
                 .map_err(|e| Error::command_error(output_type, format!("Invalid regex: {}", e)))?;
 
diff --git a/testdata/unicode_dot_regex.txt b/testdata/unicode_dot_regex.txt
new file mode 100644
index 0000000..bd4acb5
--- /dev/null
+++ b/testdata/unicode_dot_regex.txt
@@ -0,0 +1,3 @@
+# Test if . in regex matches Unicode characters
+exec echo "A✓B"
+stdout "A.B"
diff --git a/testdata/unicode_emoji.txt b/testdata/unicode_emoji.txt
new file mode 100644
index 0000000..7582e74
--- /dev/null
+++ b/testdata/unicode_emoji.txt
@@ -0,0 +1,3 @@
+# Test emoji in output
+exec echo "Status: 🎉 Success!"
+stdout "Status: 🎉 Success!"
diff --git a/testdata/unicode_grep.txt b/testdata/unicode_grep.txt
new file mode 100644
index 0000000..20d2ef0
--- /dev/null
+++ b/testdata/unicode_grep.txt
@@ -0,0 +1,6 @@
+# Test grep command with Unicode content
+grep "Success.*✓" test.txt
+stdout "test.txt:1: Success ✓"
+
+-- test.txt --
+Success ✓
diff --git a/testdata/unicode_multiple.txt b/testdata/unicode_multiple.txt
new file mode 100644
index 0000000..cb77be2
--- /dev/null
+++ b/testdata/unicode_multiple.txt
@@ -0,0 +1,3 @@
+# Test multiple Unicode characters
+exec echo "Tests: ✓ ✗ ⚠ ℹ"
+stdout "Tests: ✓ ✗ ⚠ ℹ"
diff --git a/testdata/unicode_regex.txt b/testdata/unicode_regex.txt
new file mode 100644
index 0000000..e7cb824
--- /dev/null
+++ b/testdata/unicode_regex.txt
@@ -0,0 +1,3 @@
+# Test Unicode in regex pattern
+exec echo "Success ✓ completed"
+stdout "Success .* completed"
diff --git a/testdata/unicode_special_chars.txt b/testdata/unicode_special_chars.txt
new file mode 100644
index 0000000..0d6c776
--- /dev/null
+++ b/testdata/unicode_special_chars.txt
@@ -0,0 +1,3 @@
+# Test Unicode with regex special characters
+exec echo "Result: ✓ (success)"
+stdout "Result: ✓ \(success\)"
diff --git a/testdata/unicode_stderr.txt b/testdata/unicode_stderr.txt
new file mode 100644
index 0000000..fbfda19
--- /dev/null
+++ b/testdata/unicode_stderr.txt
@@ -0,0 +1,3 @@
+# Test Unicode checkmark in stderr exact match
+exec sh -c 'echo "Error ✗" >&2'
+stderr "Error ✗"
diff --git a/testdata/unicode_stdout.txt b/testdata/unicode_stdout.txt
new file mode 100644
index 0000000..99d297e
--- /dev/null
+++ b/testdata/unicode_stdout.txt
@@ -0,0 +1,3 @@
+# Test Unicode checkmark in stdout exact match
+exec echo "Success ✓"
+stdout "Success ✓"