From d068a337ba4e2900140bd810383764019bacee3f Mon Sep 17 00:00:00 2001
From: 5225225 <5225225@mailbox.org>
Date: Thu, 4 May 2023 15:21:05 +0100
Subject: [PATCH] Fuzz options, and make failure print out a test

---
 fuzz/Cargo.toml                       |  2 +-
 fuzz/fuzz_targets/fuzz_regex_match.rs | 81 +++++++++++++++++++++------
 2 files changed, 66 insertions(+), 17 deletions(-)

diff --git a/fuzz/Cargo.toml b/fuzz/Cargo.toml
index c1e2776348..453b308699 100644
--- a/fuzz/Cargo.toml
+++ b/fuzz/Cargo.toml
@@ -9,7 +9,7 @@ edition = "2018"
 cargo-fuzz = true
 
 [dependencies]
-libfuzzer-sys = "0.4.1"
+libfuzzer-sys = { version = "0.4.1", features = ["arbitrary-derive"] }
 
 [dependencies.regex]
 path = ".."
diff --git a/fuzz/fuzz_targets/fuzz_regex_match.rs b/fuzz/fuzz_targets/fuzz_regex_match.rs
index bd9eefad54..7a66b16544 100644
--- a/fuzz/fuzz_targets/fuzz_regex_match.rs
+++ b/fuzz/fuzz_targets/fuzz_regex_match.rs
@@ -1,22 +1,71 @@
 #![no_main]
 use libfuzzer_sys::fuzz_target;
 
-fuzz_target!(|data: &[u8]| {
-    if data.len() < 2 {
-        return;
+use libfuzzer_sys::arbitrary;
+
+#[derive(arbitrary::Arbitrary)]
+struct FuzzCase<'a> {
+    pattern: &'a str,
+    input: &'a str,
+    case_insensitive: bool,
+    multi_line: bool,
+    dot_matches_new_line: bool,
+    swap_greed: bool,
+    ignore_whitespace: bool,
+    unicode: bool,
+    octal: bool,
+}
+
+impl std::fmt::Debug for FuzzCase<'_> {
+    fn fmt(
+        &self,
+        fmt: &mut std::fmt::Formatter,
+    ) -> Result<(), std::fmt::Error> {
+        let Self {
+            pattern,
+            case_insensitive,
+            multi_line,
+            dot_matches_new_line,
+            swap_greed,
+            ignore_whitespace,
+            unicode,
+            octal,
+            input,
+        } = self;
+
+        write!(
+            fmt,
+            r#"
+let r = regex::RegexBuilder::new({pattern:?})
+    .case_insensitive({case_insensitive:?})
+    .multi_line({multi_line:?})
+    .dot_matches_new_line({dot_matches_new_line:?})
+    .swap_greed({swap_greed:?})
+    .ignore_whitespace({ignore_whitespace:?})
+    .unicode({unicode:?})
+    .octal({octal:?})
+    .build();
+
+if let Ok(re) = r {{
+    re.is_match({input:?});
+}}
+        "#
+        )
     }
-    let split_point = data[0] as usize;
-    if let Ok(data) = std::str::from_utf8(&data[1..]) {
-        use std::cmp::max;
-        // split data into regular expression and actual input to search through
-        let len = data.chars().count();
-        let split_off_point = max(split_point, 1) % len as usize;
-        let char_index = data.char_indices().nth(split_off_point);
-        if let Some((char_index, _)) = char_index {
-            let (pattern, input) = data.split_at(char_index);
-            if let Ok(re) = regex::Regex::new(pattern) {
-                re.is_match(input);
-            }
-        }
+}
+
+fuzz_target!(|case: FuzzCase| {
+    let r = regex::RegexBuilder::new(case.pattern)
+        .case_insensitive(case.case_insensitive)
+        .multi_line(case.multi_line)
+        .dot_matches_new_line(case.dot_matches_new_line)
+        .swap_greed(case.swap_greed)
+        .ignore_whitespace(case.ignore_whitespace)
+        .unicode(case.unicode)
+        .octal(case.octal)
+        .build();
+
+    if let Ok(re) = r {
+        re.is_match(case.input);
     }
 });