Use Unicode properties for alnum, alpha, etc. by joshuamegnauth54 · Pull Request #7626 · RustPython/RustPython

Expand Up @@ -45,8 +45,8 @@ use rustpython_common::{ };
use icu_properties::props::{ BidiClass, BinaryProperty, CanonicalCombiningClass, EnumeratedProperty, GeneralCategory, XidContinue, XidStart, BidiClass, BinaryProperty, EnumeratedProperty, GeneralCategory, GeneralCategoryGroup, NumericType, XidContinue, XidStart, }; use unicode_casing::CharExt;
Expand Down Expand Up @@ -949,23 +949,30 @@ impl PyStr { fn isalnum(&self) -> bool { !self.data.is_empty() && self.char_all(|c| { c.is_alphanumeric() && CanonicalCombiningClass::for_char(c) == CanonicalCombiningClass::NotReordered GeneralCategoryGroup::Letter .union(GeneralCategoryGroup::Number) .contains(GeneralCategory::for_char(c)) }) }
#[pymethod] fn isnumeric(&self) -> bool { !self.data.is_empty() && self.char_all(char::is_numeric) !self.data.is_empty() && self.char_all(|c| { [ NumericType::Decimal, NumericType::Digit, NumericType::Numeric, ] .contains(&NumericType::for_char(c)) }) }
#[pymethod] fn isdigit(&self) -> bool { // python's isdigit also checks if exponents are digits, these are the unicode codepoints for exponents !self.data.is_empty() && self.char_all(|c| { c.is_ascii_digit() || matches!(c, '⁰' | '¹' | '²' | '³' | '⁴' | '⁵' | '⁶' | '⁷' | '⁸' | '⁹') [NumericType::Digit, NumericType::Decimal].contains(&NumericType::for_char(c)) }) }
Expand Down Expand Up @@ -1064,7 +1071,9 @@ impl PyStr {
#[pymethod] fn isalpha(&self) -> bool { !self.data.is_empty() && self.char_all(char::is_alphabetic) !self.data.is_empty() && self .char_all(|c| GeneralCategoryGroup::Letter.contains(GeneralCategory::for_char(c))) }
#[pymethod] Expand Down