From 7439050d85f674a1886c4dc497fade20a9d794a7 Mon Sep 17 00:00:00 2001
From: Daniel Hofstetter <daniel.hofstetter@42dh.com>
Date: Wed, 28 May 2025 15:53:54 +0200
Subject: [PATCH 1/2] csplit: only allow ASCII digits for repeat pattern

---
 src/uu/csplit/src/patterns.rs | 2 +-
 tests/by-util/test_csplit.rs  | 9 +++++++++
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/src/uu/csplit/src/patterns.rs b/src/uu/csplit/src/patterns.rs
index bdf15b511..b49d1c606 100644
--- a/src/uu/csplit/src/patterns.rs
+++ b/src/uu/csplit/src/patterns.rs
@@ -107,7 +107,7 @@ fn extract_patterns(args: &[String]) -> Result<Vec<Pattern>, CsplitError> {
     let mut patterns = Vec::with_capacity(args.len());
     let to_match_reg =
         Regex::new(r"^(/(?P<UPTO>.+)/|%(?P<SKIPTO>.+)%)(?P<OFFSET>[\+-]?\d+)?$").unwrap();
-    let execute_ntimes_reg = Regex::new(r"^\{(?P<TIMES>\d+)|\*\}$").unwrap();
+    let execute_ntimes_reg = Regex::new(r"^\{(?P<TIMES>[0-9]+)|\*\}$").unwrap();
     let mut iter = args.iter().peekable();
 
     while let Some(arg) = iter.next() {
diff --git a/tests/by-util/test_csplit.rs b/tests/by-util/test_csplit.rs
index a7a802b92..2a94c10bf 100644
--- a/tests/by-util/test_csplit.rs
+++ b/tests/by-util/test_csplit.rs
@@ -83,6 +83,15 @@ fn test_up_to_line_sequence() {
     assert_eq!(at.read("xx02"), generate(25, 51));
 }
 
+#[test]
+fn test_up_to_line_with_non_ascii_repeat() {
+    // we use a different error message than GNU
+    new_ucmd!()
+        .args(&["numbers50.txt", "10", "{𝟚}"])
+        .fails()
+        .stderr_contains("invalid pattern");
+}
+
 #[test]
 fn test_up_to_match() {
     let (at, mut ucmd) = at_and_ucmd!();

From d5b6af5216fd9cdefa9004703ecaa76684181213 Mon Sep 17 00:00:00 2001
From: Daniel Hofstetter <daniel.hofstetter@42dh.com>
Date: Wed, 28 May 2025 16:07:45 +0200
Subject: [PATCH 2/2] csplit: only allow ASCII digits as offset

---
 src/uu/csplit/src/patterns.rs | 2 +-
 tests/by-util/test_csplit.rs  | 9 +++++++++
 2 files changed, 10 insertions(+), 1 deletion(-)

diff --git a/src/uu/csplit/src/patterns.rs b/src/uu/csplit/src/patterns.rs
index b49d1c606..9326ea379 100644
--- a/src/uu/csplit/src/patterns.rs
+++ b/src/uu/csplit/src/patterns.rs
@@ -106,7 +106,7 @@ pub fn get_patterns(args: &[String]) -> Result<Vec<Pattern>, CsplitError> {
 fn extract_patterns(args: &[String]) -> Result<Vec<Pattern>, CsplitError> {
     let mut patterns = Vec::with_capacity(args.len());
     let to_match_reg =
-        Regex::new(r"^(/(?P<UPTO>.+)/|%(?P<SKIPTO>.+)%)(?P<OFFSET>[\+-]?\d+)?$").unwrap();
+        Regex::new(r"^(/(?P<UPTO>.+)/|%(?P<SKIPTO>.+)%)(?P<OFFSET>[\+-]?[0-9]+)?$").unwrap();
     let execute_ntimes_reg = Regex::new(r"^\{(?P<TIMES>[0-9]+)|\*\}$").unwrap();
     let mut iter = args.iter().peekable();
 
diff --git a/tests/by-util/test_csplit.rs b/tests/by-util/test_csplit.rs
index 2a94c10bf..96322e444 100644
--- a/tests/by-util/test_csplit.rs
+++ b/tests/by-util/test_csplit.rs
@@ -176,6 +176,15 @@ fn test_up_to_match_offset_repeat_twice() {
     assert_eq!(at.read("xx03"), generate(32, 51));
 }
 
+#[test]
+fn test_up_to_match_non_ascii_offset() {
+    // we use a different error message than GNU
+    new_ucmd!()
+        .args(&["numbers50.txt", "/9$/𝟚"])
+        .fails()
+        .stderr_contains("invalid pattern");
+}
+
 #[test]
 fn test_up_to_match_negative_offset() {
     let (at, mut ucmd) = at_and_ucmd!();