more wrapping arithmetic in ascii.rs

hsivonen · Aug 24, 2020 · 9e92a43 · 9e92a43
1 parent e53ff4a
commit 9e92a43
Show file tree

Hide file tree

Showing 9 changed files with 53 additions and 24 deletions.
diff --git a/src/ascii.rs b/src/ascii.rs
@@ -291,7 +291,7 @@ macro_rules! latin1_alu {
                         let src_until_alignment = (ALU_ALIGNMENT
                             - ((src as usize) & ALU_ALIGNMENT_MASK))
                             & ALU_ALIGNMENT_MASK;
-                        if (dst.add(src_until_alignment) as usize) & ALU_ALIGNMENT_MASK != 0 {
+                        if (dst.wrapping_add(src_until_alignment) as usize) & ALU_ALIGNMENT_MASK != 0 {
                             break;
                         }
                         src_until_alignment
@@ -300,7 +300,7 @@ macro_rules! latin1_alu {
                         let dst_until_alignment = (ALU_ALIGNMENT
                             - ((dst as usize) & ALU_ALIGNMENT_MASK))
                             & ALU_ALIGNMENT_MASK;
-                        if (src.add(dst_until_alignment) as usize) & ALU_ALIGNMENT_MASK != 0 {
+                        if (src.wrapping_add(dst_until_alignment) as usize) & ALU_ALIGNMENT_MASK != 0 {
                             break;
                         }
                         dst_until_alignment

diff --git a/src/big5.rs b/src/big5.rs
@@ -365,18 +365,20 @@ mod tests {
         // ASCII
         encode_big5("\u{0061}\u{0062}", b"\x61\x62");
 
-        // Edge cases
-        encode_big5("\u{9EA6}\u{0061}", b"&#40614;\x61");
-        encode_big5("\u{2626B}\u{0061}", b"&#156267;\x61");
-        encode_big5("\u{3000}", b"\xA1\x40");
-        encode_big5("\u{20AC}", b"\xA3\xE1");
-        encode_big5("\u{4E00}", b"\xA4\x40");
-        encode_big5("\u{27607}", b"\xC8\xA4");
-        encode_big5("\u{FFE2}", b"\xC8\xCD");
-        encode_big5("\u{79D4}", b"\xFE\xFE");
-
-        // Not in index
-        encode_big5("\u{2603}\u{0061}", b"&#9731;\x61");
+        if !cfg!(miri) { // Miri is too slow
+            // Edge cases
+            encode_big5("\u{9EA6}\u{0061}", b"&#40614;\x61");
+            encode_big5("\u{2626B}\u{0061}", b"&#156267;\x61");
+            encode_big5("\u{3000}", b"\xA1\x40");
+            encode_big5("\u{20AC}", b"\xA3\xE1");
+            encode_big5("\u{4E00}", b"\xA4\x40");
+            encode_big5("\u{27607}", b"\xC8\xA4");
+            encode_big5("\u{FFE2}", b"\xC8\xCD");
+            encode_big5("\u{79D4}", b"\xFE\xFE");
+
+            // Not in index
+            encode_big5("\u{2603}\u{0061}", b"&#9731;\x61");
+        }
 
         // duplicate low bits
         encode_big5("\u{203B5}", b"\xFD\x6A");
@@ -387,6 +389,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_big5_decode_all() {
         let input = include_bytes!("test_data/big5_in.txt");
         let expectation = include_str!("test_data/big5_in_ref.txt");
@@ -396,6 +399,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_big5_encode_all() {
         let input = include_str!("test_data/big5_out.txt");
         let expectation = include_bytes!("test_data/big5_out_ref.txt");
@@ -406,6 +410,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_big5_encode_from_two_low_surrogates() {
         let expectation = b"&#65533;&#65533;";
         let mut output = [0u8; 40];

diff --git a/src/euc_jp.rs b/src/euc_jp.rs
@@ -437,6 +437,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_jis0208_decode_all() {
         let input = include_bytes!("test_data/jis0208_in.txt");
         let expectation = include_str!("test_data/jis0208_in_ref.txt");
@@ -446,6 +447,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_jis0208_encode_all() {
         let input = include_str!("test_data/jis0208_out.txt");
         let expectation = include_bytes!("test_data/jis0208_out_ref.txt");
@@ -456,6 +458,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_jis0212_decode_all() {
         let input = include_bytes!("test_data/jis0212_in.txt");
         let expectation = include_str!("test_data/jis0212_in_ref.txt");

diff --git a/src/euc_kr.rs b/src/euc_kr.rs
@@ -406,6 +406,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_euc_kr_decode_all() {
         let input = include_bytes!("test_data/euc_kr_in.txt");
         let expectation = include_str!("test_data/euc_kr_in_ref.txt");
@@ -415,6 +416,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_euc_kr_encode_all() {
         let input = include_str!("test_data/euc_kr_out.txt");
         let expectation = include_bytes!("test_data/euc_kr_out_ref.txt");

diff --git a/src/gb18030.rs b/src/gb18030.rs
@@ -712,6 +712,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_gb18030_decode_all() {
         let input = include_bytes!("test_data/gb18030_in.txt");
         let expectation = include_str!("test_data/gb18030_in_ref.txt");
@@ -721,6 +722,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_gb18030_encode_all() {
         let input = include_str!("test_data/gb18030_out.txt");
         let expectation = include_bytes!("test_data/gb18030_out_ref.txt");

diff --git a/src/iso_2022_jp.rs b/src/iso_2022_jp.rs
@@ -967,6 +967,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_iso_2022_jp_decode_all() {
         let input = include_bytes!("test_data/iso_2022_jp_in.txt");
         let expectation = include_str!("test_data/iso_2022_jp_in_ref.txt");
@@ -976,6 +977,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_iso_2022_jp_encode_all() {
         let input = include_str!("test_data/iso_2022_jp_out.txt");
         let expectation = include_bytes!("test_data/iso_2022_jp_out_ref.txt");

diff --git a/src/mem.rs b/src/mem.rs
@@ -2231,8 +2231,9 @@ mod tests {
 
     #[test]
     fn test_is_utf16_latin1_fail() {
-        let mut src: Vec<u16> = Vec::with_capacity(256);
-        src.resize(256, 0);
+        let len = if cfg!(miri) { 64 } else { 256 }; // Miri is too slow
+        let mut src: Vec<u16> = Vec::with_capacity(len);
+        src.resize(len, 0);
         for i in 0..src.len() {
             src[i] = i as u16;
         }
@@ -2262,8 +2263,9 @@ mod tests {
 
     #[test]
     fn test_is_str_latin1_fail() {
-        let mut src: Vec<u16> = Vec::with_capacity(256);
-        src.resize(256, 0);
+        let len = if cfg!(miri) { 32 } else { 256 }; // Miri is too slow
+        let mut src: Vec<u16> = Vec::with_capacity(len);
+        src.resize(len, 0);
         for i in 0..src.len() {
             src[i] = i as u16;
         }
@@ -2280,8 +2282,9 @@ mod tests {
 
     #[test]
     fn test_is_utf8_latin1_success() {
-        let mut src: Vec<u16> = Vec::with_capacity(256);
-        src.resize(256, 0);
+        let len = if cfg!(miri) { 64 } else { 256 }; // Miri is too slow
+        let mut src: Vec<u16> = Vec::with_capacity(len);
+        src.resize(len, 0);
         for i in 0..src.len() {
             src[i] = i as u16;
         }
@@ -2297,8 +2300,9 @@ mod tests {
 
     #[test]
     fn test_is_utf8_latin1_fail() {
-        let mut src: Vec<u16> = Vec::with_capacity(256);
-        src.resize(256, 0);
+        let len = if cfg!(miri) { 32 } else { 256 }; // Miri is too slow
+        let mut src: Vec<u16> = Vec::with_capacity(len);
+        src.resize(len, 0);
         for i in 0..src.len() {
             src[i] = i as u16;
         }
@@ -3144,6 +3148,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_is_char_bidi_thoroughly() {
         for i in 0..0xD800u32 {
             let c: char = ::std::char::from_u32(i).unwrap();
@@ -3156,6 +3161,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_is_utf16_code_unit_bidi_thoroughly() {
         for i in 0..0x10000u32 {
             let u = i as u16;
@@ -3167,6 +3173,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_is_str_bidi_thoroughly() {
         let mut buf = [0; 4];
         for i in 0..0xD800u32 {
@@ -3186,6 +3193,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_is_utf8_bidi_thoroughly() {
         let mut buf = [0; 8];
         for i in 0..0xD800u32 {
@@ -3227,6 +3235,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_is_utf16_bidi_thoroughly() {
         let mut buf = [0; 32];
         for i in 0..0x10000u32 {

diff --git a/src/shift_jis.rs b/src/shift_jis.rs
@@ -385,6 +385,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_shift_jis_decode_all() {
         let input = include_bytes!("test_data/shift_jis_in.txt");
         let expectation = include_str!("test_data/shift_jis_in_ref.txt");
@@ -394,6 +395,7 @@ mod tests {
     }
 
     #[test]
+    #[cfg_attr(miri, ignore)] // Miri is too slow
     fn test_shift_jis_encode_all() {
         let input = include_str!("test_data/shift_jis_out.txt");
         let expectation = include_bytes!("test_data/shift_jis_out_ref.txt");

diff --git a/src/testing.rs b/src/testing.rs
@@ -12,7 +12,9 @@ use super::*;
 pub fn decode(encoding: &'static Encoding, bytes: &[u8], expect: &str) {
     let mut vec = Vec::with_capacity(bytes.len() + 32);
     let mut string = String::with_capacity(expect.len() + 32);
-    for i in 0usize..32usize {
+    let range = if cfg!(miri) { 0usize..4usize } else { 0usize..32usize };
+    for i in range {
+        eprintln!("GREPME decode: loop {}", i);
         vec.clear();
         string.clear();
         for j in 0usize..i {
@@ -44,7 +46,9 @@ fn decode_without_padding_impl(
 pub fn encode(encoding: &'static Encoding, str: &str, expect: &[u8]) {
     let mut vec = Vec::with_capacity(expect.len() + 32);
     let mut string = String::with_capacity(str.len() + 32);
-    for i in 0usize..32usize {
+    let range = if cfg!(miri) { 0usize..4usize } else { 0usize..32usize };
+    for i in range {
+        eprintln!("GREPME encode: loop {}", i);
         vec.clear();
         string.clear();
         for j in 0usize..i {