Fix {str|bytes|bytearray}.{isascii,islower,isupper}

youknowone · youknowone · commit 95e938cb9b64 · 2020-04-11T11:01:05.000+09:00
diff --git a/Lib/test/string_tests.py b/Lib/test/string_tests.py
@@ -857,7 +857,6 @@ def test_zfill(self):
 
         self.checkraises(TypeError, '123', 'zfill')
 
-    @unittest.skip("TODO: RUSTPYTHON test_bytes")
     def test_islower(self):
         self.checkequal(False, '', 'islower')
         self.checkequal(True, 'a', 'islower')
@@ -868,7 +867,6 @@ def test_islower(self):
         self.checkequal(True, 'abc\n', 'islower')
         self.checkraises(TypeError, 'abc', 'islower', 42)
 
-    @unittest.skip("TODO: RUSTPYTHON test_bytes")
     def test_isupper(self):
         self.checkequal(False, '', 'isupper')
         self.checkequal(False, 'a', 'isupper')
@@ -925,7 +923,6 @@ def test_isalnum(self):
         self.checkequal(False, 'abc\n', 'isalnum')
         self.checkraises(TypeError, 'abc', 'isalnum', 42)
 
-    @unittest.skip("TODO: RUSTPYTHON test_bytes")
     def test_isascii(self):
         self.checkequal(True, '', 'isascii')
         self.checkequal(True, '\x00', 'isascii')
diff --git a/vm/src/obj/objbyteinner.rs b/vm/src/obj/objbyteinner.rs
@@ -611,33 +611,47 @@ impl PyByteInner {
     }
 
     pub fn isascii(&self) -> bool {
-        !self.elements.is_empty() && self.elements.iter().all(|x| char::from(*x).is_ascii())
+        self.elements.iter().all(|x| char::from(*x).is_ascii())
     }
 
     pub fn isdigit(&self) -> bool {
         !self.elements.is_empty() && self.elements.iter().all(|x| char::from(*x).is_digit(10))
     }
 
     pub fn islower(&self) -> bool {
-        !self.elements.is_empty()
-            && self
-                .elements
-                .iter()
-                .filter(|x| !char::from(**x).is_whitespace())
-                .all(|x| char::from(*x).is_lowercase())
+        // CPython _Py_bytes_islower
+        let mut cased = false;
+        for b in self.elements.iter() {
+            let c = *b as char;
+            if c.is_uppercase() {
+                return false;
+            } else if !cased && c.is_lowercase() {
+                cased = true
+            }
+        }
+        cased
     }
 
-    pub fn isspace(&self) -> bool {
-        !self.elements.is_empty() && self.elements.iter().all(|x| char::from(*x).is_whitespace())
+    pub fn isupper(&self) -> bool {
+        // CPython _Py_bytes_isupper
+        let mut cased = false;
+        for b in self.elements.iter() {
+            let c = *b as char;
+            if c.is_lowercase() {
+                return false;
+            } else if !cased && c.is_uppercase() {
+                cased = true
+            }
+        }
+        cased
     }
 
-    pub fn isupper(&self) -> bool {
+    pub fn isspace(&self) -> bool {
         !self.elements.is_empty()
             && self
                 .elements
                 .iter()
-                .filter(|x| !char::from(**x).is_whitespace())
-                .all(|x| char::from(*x).is_uppercase())
+                .all(|x| char::from(*x).is_ascii_whitespace())
     }
 
     pub fn istitle(&self) -> bool {
diff --git a/vm/src/obj/objstr.rs b/vm/src/obj/objstr.rs
@@ -9,7 +9,6 @@ use std::string::ToString;
 use num_traits::ToPrimitive;
 use unic::ucd::category::GeneralCategory;
 use unic::ucd::ident::{is_xid_continue, is_xid_start};
-use unic::ucd::is_cased;
 use unicode_casing::CharExt;
 
 use super::objbytes::{PyBytes, PyBytesRef};
@@ -826,29 +825,31 @@ impl PyString {
         !self.value.is_empty() && self.value.chars().all(|c| c.is_ascii_whitespace())
     }
 
-    // Return true if all cased characters in the string are uppercase and there is at least one cased character, false otherwise.
+    // Return true if all cased characters in the string are lowercase and there is at least one cased character, false otherwise.
     #[pymethod]
-    fn isupper(&self) -> bool {
+    fn islower(&self) -> bool {
+        // CPython unicode_islower_impl
         let mut cased = false;
         for c in self.value.chars() {
-            if is_cased(c) && c.is_uppercase() {
-                cased = true
-            } else if is_cased(c) && c.is_lowercase() {
+            if c.is_uppercase() {
                 return false;
+            } else if !cased && c.is_lowercase() {
+                cased = true
             }
         }
         cased
     }
 
-    // Return true if all cased characters in the string are lowercase and there is at least one cased character, false otherwise.
+    // Return true if all cased characters in the string are uppercase and there is at least one cased character, false otherwise.
     #[pymethod]
-    fn islower(&self) -> bool {
+    fn isupper(&self) -> bool {
+        // CPython unicode_isupper_impl
         let mut cased = false;
         for c in self.value.chars() {
-            if is_cased(c) && c.is_lowercase() {
-                cased = true
-            } else if is_cased(c) && c.is_uppercase() {
+            if c.is_lowercase() {
                 return false;
+            } else if !cased && c.is_uppercase() {
+                cased = true
             }
         }
         cased
