diff --git a/Cargo.lock b/Cargo.lock
index a71ceb53b3..28fe718260 100644
--- a/Cargo.lock
+++ b/Cargo.lock
@@ -1270,9 +1270,9 @@ checksum = "830d08ce1d1d941e6b30645f1a0eb5643013d835ce3779a5fc208261dbe10f55"
[[package]]
name = "lexical-parse-float"
-version = "0.8.5"
+version = "1.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "683b3a5ebd0130b8fb52ba0bdc718cc56815b6a097e28ae5a6997d0ad17dc05f"
+checksum = "de6f9cb01fb0b08060209a057c048fcbab8717b4c1ecd2eac66ebfe39a65b0f2"
dependencies = [
"lexical-parse-integer",
"lexical-util",
@@ -1281,9 +1281,9 @@ dependencies = [
[[package]]
name = "lexical-parse-integer"
-version = "0.8.6"
+version = "1.0.5"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "6d0994485ed0c312f6d965766754ea177d07f9c00c9b82a5ee62ed5b47945ee9"
+checksum = "72207aae22fc0a121ba7b6d479e42cbfea549af1479c3f3a4f12c70dd66df12e"
dependencies = [
"lexical-util",
"static_assertions",
@@ -1291,9 +1291,9 @@ dependencies = [
[[package]]
name = "lexical-util"
-version = "0.8.5"
+version = "1.0.6"
source = "registry+https://github.com/rust-lang/crates.io-index"
-checksum = "5255b9ff16ff898710eb9eb63cb39248ea8a5bb036bea8085b1a767ff6c4e3fc"
+checksum = "5a82e24bf537fd24c177ffbbdc6ebcc8d54732c35b50a3f28cc3f4e4c949a0b3"
dependencies = [
"static_assertions",
]
diff --git a/Lib/test/test_float.py b/Lib/test/test_float.py
index 19c17af596..30d27072fb 100644
--- a/Lib/test/test_float.py
+++ b/Lib/test/test_float.py
@@ -35,8 +35,6 @@ class OtherFloatSubclass(float):
class GeneralFloatCases(unittest.TestCase):
- # TODO: RUSTPYTHON
- @unittest.expectedFailure
def test_float(self):
self.assertEqual(float(3.14), 3.14)
self.assertEqual(float(314), 314.0)
diff --git a/common/src/str.rs b/common/src/str.rs
index e72f2efb95..ca1723e7ef 100644
--- a/common/src/str.rs
+++ b/common/src/str.rs
@@ -609,6 +609,49 @@ macro_rules! ascii {
}
pub use ascii;
+// TODO: this should probably live in a crate like unic or unicode-properties
+const UNICODE_DECIMAL_VALUES: &[char] = &[
+ '0', '1', '2', '3', '4', '5', '6', '7', '8', '9', '٠', '١', '٢', '٣', '٤', '٥', '٦', '٧', '٨',
+ '٩', '۰', '۱', '۲', '۳', '۴', '۵', '۶', '۷', '۸', '۹', '߀', '߁', '߂', '߃', '߄', '߅', '߆', '߇',
+ '߈', '߉', '०', '१', '२', '३', '४', '५', '६', '७', '८', '९', '০', '১', '২', '৩', '৪', '৫', '৬',
+ '৭', '৮', '৯', '੦', '੧', '੨', '੩', '੪', '੫', '੬', '੭', '੮', '੯', '૦', '૧', '૨', '૩', '૪', '૫',
+ '૬', '૭', '૮', '૯', '୦', '୧', '୨', '୩', '୪', '୫', '୬', '୭', '୮', '୯', '௦', '௧', '௨', '௩', '௪',
+ '௫', '௬', '௭', '௮', '௯', '౦', '౧', '౨', '౩', '౪', '౫', '౬', '౭', '౮', '౯', '೦', '೧', '೨', '೩',
+ '೪', '೫', '೬', '೭', '೮', '೯', '൦', '൧', '൨', '൩', '൪', '൫', '൬', '൭', '൮', '൯', '෦', '෧', '෨',
+ '෩', '෪', '෫', '෬', '෭', '෮', '෯', '๐', '๑', '๒', '๓', '๔', '๕', '๖', '๗', '๘', '๙', '໐', '໑',
+ '໒', '໓', '໔', '໕', '໖', '໗', '໘', '໙', '༠', '༡', '༢', '༣', '༤', '༥', '༦', '༧', '༨', '༩', '၀',
+ '၁', '၂', '၃', '၄', '၅', '၆', '၇', '၈', '၉', '႐', '႑', '႒', '႓', '႔', '႕', '႖', '႗', '႘', '႙',
+ '០', '១', '២', '៣', '៤', '៥', '៦', '៧', '៨', '៩', '᠐', '᠑', '᠒', '᠓', '᠔', '᠕', '᠖', '᠗', '᠘',
+ '᠙', '᥆', '᥇', '᥈', '᥉', '᥊', '᥋', '᥌', '᥍', '᥎', '᥏', '᧐', '᧑', '᧒', '᧓', '᧔', '᧕', '᧖', '᧗',
+ '᧘', '᧙', '᪀', '᪁', '᪂', '᪃', '᪄', '᪅', '᪆', '᪇', '᪈', '᪉', '᪐', '᪑', '᪒', '᪓', '᪔', '᪕', '᪖',
+ '᪗', '᪘', '᪙', '᭐', '᭑', '᭒', '᭓', '᭔', '᭕', '᭖', '᭗', '᭘', '᭙', '᮰', '᮱', '᮲', '᮳', '᮴', '᮵',
+ '᮶', '᮷', '᮸', '᮹', '᱀', '᱁', '᱂', '᱃', '᱄', '᱅', '᱆', '᱇', '᱈', '᱉', '᱐', '᱑', '᱒', '᱓', '᱔',
+ '᱕', '᱖', '᱗', '᱘', '᱙', '꘠', '꘡', '꘢', '꘣', '꘤', '꘥', '꘦', '꘧', '꘨', '꘩', '꣐', '꣑', '꣒', '꣓',
+ '꣔', '꣕', '꣖', '꣗', '꣘', '꣙', '꤀', '꤁', '꤂', '꤃', '꤄', '꤅', '꤆', '꤇', '꤈', '꤉', '꧐', '꧑', '꧒',
+ '꧓', '꧔', '꧕', '꧖', '꧗', '꧘', '꧙', '꧰', '꧱', '꧲', '꧳', '꧴', '꧵', '꧶', '꧷', '꧸', '꧹', '꩐', '꩑',
+ '꩒', '꩓', '꩔', '꩕', '꩖', '꩗', '꩘', '꩙', '꯰', '꯱', '꯲', '꯳', '꯴', '꯵', '꯶', '꯷', '꯸', '꯹', '０',
+ '１', '２', '３', '４', '５', '６', '７', '８', '９', '𐒠', '𐒡', '𐒢', '𐒣', '𐒤', '𐒥', '𐒦', '𐒧',
+ '𐒨', '𐒩', '𑁦', '𑁧', '𑁨', '𑁩', '𑁪', '𑁫', '𑁬', '𑁭', '𑁮', '𑁯', '𑃰', '𑃱', '𑃲', '𑃳', '𑃴', '𑃵', '𑃶',
+ '𑃷', '𑃸', '𑃹', '𑄶', '𑄷', '𑄸', '𑄹', '𑄺', '𑄻', '𑄼', '𑄽', '𑄾', '𑄿', '𑇐', '𑇑', '𑇒', '𑇓', '𑇔', '𑇕',
+ '𑇖', '𑇗', '𑇘', '𑇙', '𑋰', '𑋱', '𑋲', '𑋳', '𑋴', '𑋵', '𑋶', '𑋷', '𑋸', '𑋹', '𑑐', '𑑑', '𑑒', '𑑓', '𑑔',
+ '𑑕', '𑑖', '𑑗', '𑑘', '𑑙', '𑓐', '𑓑', '𑓒', '𑓓', '𑓔', '𑓕', '𑓖', '𑓗', '𑓘', '𑓙', '𑙐', '𑙑', '𑙒', '𑙓',
+ '𑙔', '𑙕', '𑙖', '𑙗', '𑙘', '𑙙', '𑛀', '𑛁', '𑛂', '𑛃', '𑛄', '𑛅', '𑛆', '𑛇', '𑛈', '𑛉', '𑜰', '𑜱', '𑜲',
+ '𑜳', '𑜴', '𑜵', '𑜶', '𑜷', '𑜸', '𑜹', '𑣠', '𑣡', '𑣢', '𑣣', '𑣤', '𑣥', '𑣦', '𑣧', '𑣨', '𑣩', '𑱐', '𑱑',
+ '𑱒', '𑱓', '𑱔', '𑱕', '𑱖', '𑱗', '𑱘', '𑱙', '𑵐', '𑵑', '𑵒', '𑵓', '𑵔', '𑵕', '𑵖', '𑵗', '𑵘', '𑵙', '𖩠',
+ '𖩡', '𖩢', '𖩣', '𖩤', '𖩥', '𖩦', '𖩧', '𖩨', '𖩩', '𖭐', '𖭑', '𖭒', '𖭓', '𖭔', '𖭕', '𖭖', '𖭗', '𖭘', '𖭙',
+ '𝟎', '𝟏', '𝟐', '𝟑', '𝟒', '𝟓', '𝟔', '𝟕', '𝟖', '𝟗', '𝟘', '𝟙', '𝟚', '𝟛', '𝟜', '𝟝', '𝟞', '𝟟', '𝟠',
+ '𝟡', '𝟢', '𝟣', '𝟤', '𝟥', '𝟦', '𝟧', '𝟨', '𝟩', '𝟪', '𝟫', '𝟬', '𝟭', '𝟮', '𝟯', '𝟰', '𝟱', '𝟲', '𝟳',
+ '𝟴', '𝟵', '𝟶', '𝟷', '𝟸', '𝟹', '𝟺', '𝟻', '𝟼', '𝟽', '𝟾', '𝟿', '𞥐', '𞥑', '𞥒', '𞥓', '𞥔', '𞥕', '𞥖',
+ '𞥗', '𞥘', '𞥙',
+];
+
+pub fn char_to_decimal(ch: char) -> Option<u8> {
+ UNICODE_DECIMAL_VALUES
+ .binary_search(&ch)
+ .ok()
+ .map(|i| (i % 10) as u8)
+}
+
#[cfg(test)]
mod tests {
use super::*;
diff --git a/compiler/codegen/src/unparse.rs b/compiler/codegen/src/unparse.rs
index 458ff76fc7..1ecf1f9334 100644
--- a/compiler/codegen/src/unparse.rs
+++ b/compiler/codegen/src/unparse.rs
@@ -366,7 +366,7 @@ impl<'a, 'b, 'c> Unparser<'a, 'b, 'c> {
}
}
&ruff::Number::Complex { real, imag } => self
- .p(&rustpython_literal::float::complex_to_string(real, imag)
+ .p(&rustpython_literal::complex::to_string(real, imag)
.replace("inf", inf_str))?,
}
}
diff --git a/compiler/literal/Cargo.toml b/compiler/literal/Cargo.toml
index b4fa6229f1..da55d107b3 100644
--- a/compiler/literal/Cargo.toml
+++ b/compiler/literal/Cargo.toml
@@ -13,7 +13,7 @@ rustpython-wtf8 = { workspace = true }
hexf-parse = "0.2.1"
is-macro.workspace = true
-lexical-parse-float = { version = "0.8.0", features = ["format"] }
+lexical-parse-float = { version = "1.0.4", features = ["format"] }
num-traits = { workspace = true }
unic-ucd-category = { workspace = true }
diff --git a/compiler/literal/src/complex.rs b/compiler/literal/src/complex.rs
new file mode 100644
index 0000000000..076f2807c9
--- /dev/null
+++ b/compiler/literal/src/complex.rs
@@ -0,0 +1,73 @@
+use crate::float;
+
+/// Convert a complex number to a string.
+pub fn to_string(re: f64, im: f64) -> String {
+ // integer => drop ., fractional => float_ops
+ let mut im_part = if im.fract() == 0.0 {
+ im.to_string()
+ } else {
+ float::to_string(im)
+ };
+ im_part.push('j');
+
+ // positive empty => return im_part, integer => drop ., fractional => float_ops
+ let re_part = if re == 0.0 {
+ if re.is_sign_positive() {
+ return im_part;
+ } else {
+ "-0".to_owned()
+ }
+ } else if re.fract() == 0.0 {
+ re.to_string()
+ } else {
+ float::to_string(re)
+ };
+ let mut result =
+ String::with_capacity(re_part.len() + im_part.len() + 2 + im.is_sign_positive() as usize);
+ result.push('(');
+ result.push_str(&re_part);
+ if im.is_sign_positive() || im.is_nan() {
+ result.push('+');
+ }
+ result.push_str(&im_part);
+ result.push(')');
+ result
+}
+
+/// Parse a complex number from a string.
+///
+/// Returns `Some((re, im))` on success.
+pub fn parse_str(s: &str) -> Option<(f64, f64)> {
+ let s = s.trim();
+ // Handle parentheses
+ let s = match s.strip_prefix('(') {
+ None => s,
+ Some(s) => s.strip_suffix(')')?.trim(),
+ };
+
+ let value = match s.strip_suffix(|c| c == 'j' || c == 'J') {
+ None => (float::parse_str(s)?, 0.0),
+ Some(mut s) => {
+ let mut real = 0.0;
+ // Find the central +/- operator. If it exists, parse the real part.
+ for (i, w) in s.as_bytes().windows(2).enumerate() {
+ if (w[1] == b'+' || w[1] == b'-') && !(w[0] == b'e' || w[0] == b'E') {
+ real = float::parse_str(&s[..=i])?;
+ s = &s[i + 1..];
+ break;
+ }
+ }
+
+ let imag = match s {
+ // "j", "+j"
+ "" | "+" => 1.0,
+ // "-j"
+ "-" => -1.0,
+ s => float::parse_str(s)?,
+ };
+
+ (real, imag)
+ }
+ };
+ Some(value)
+}
diff --git a/compiler/literal/src/float.rs b/compiler/literal/src/float.rs
index e05a105fd4..49771b8184 100644
--- a/compiler/literal/src/float.rs
+++ b/compiler/literal/src/float.rs
@@ -6,49 +6,8 @@ pub fn parse_str(literal: &str) -> Option<f64> {
parse_inner(literal.trim().as_bytes())
}
-fn strip_underlines(literal: &[u8]) -> Option<Vec<u8>> {
- let mut prev = b'\0';
- let mut dup = Vec::<u8>::new();
- for p in literal {
- if *p == b'_' {
- // Underscores are only allowed after digits.
- if !prev.is_ascii_digit() {
- return None;
- }
- } else {
- dup.push(*p);
- // Underscores are only allowed before digits.
- if prev == b'_' && !p.is_ascii_digit() {
- return None;
- }
- }
- prev = *p;
- }
-
- // Underscores are not allowed at the end.
- if prev == b'_' {
- return None;
- }
-
- Some(dup)
-}
-
pub fn parse_bytes(literal: &[u8]) -> Option<f64> {
- parse_inner(trim_slice(literal, |b| b.is_ascii_whitespace()))
-}
-
-fn trim_slice<T>(v: &[T], mut trim: impl FnMut(&T) -> bool) -> &[T] {
- let mut it = v.iter();
- // it.take_while_ref(&mut trim).for_each(drop);
- // hmm.. `&mut slice::Iter<_>` is not `Clone`
- // it.by_ref().rev().take_while_ref(&mut trim).for_each(drop);
- while it.clone().next().is_some_and(&mut trim) {
- it.next();
- }
- while it.clone().next_back().is_some_and(&mut trim) {
- it.next_back();
- }
- it.as_slice()
+ parse_inner(literal.trim_ascii())
}
fn parse_inner(literal: &[u8]) -> Option<f64> {
@@ -56,15 +15,11 @@ fn parse_inner(literal: &[u8]) -> Option<f64> {
FromLexicalWithOptions, NumberFormatBuilder, Options, format::PYTHON3_LITERAL,
};
- // Use custom function for underline handling for now.
- // For further information see https://github.com/Alexhuszagh/rust-lexical/issues/96.
- let stripped = strip_underlines(literal)?;
-
// lexical-core's format::PYTHON_STRING is inaccurate
const PYTHON_STRING: u128 = NumberFormatBuilder::rebuild(PYTHON3_LITERAL)
.no_special(false)
.build();
- f64::from_lexical_with_options::<PYTHON_STRING>(&stripped, &Options::new()).ok()
+ f64::from_lexical_with_options::<PYTHON_STRING>(literal, &Options::new()).ok()
}
pub fn is_integer(v: f64) -> bool {
@@ -223,39 +178,6 @@ pub fn to_string(value: f64) -> String {
}
}
-pub fn complex_to_string(re: f64, im: f64) -> String {
- // integer => drop ., fractional => float_ops
- let mut im_part = if im.fract() == 0.0 {
- im.to_string()
- } else {
- to_string(im)
- };
- im_part.push('j');
-
- // positive empty => return im_part, integer => drop ., fractional => float_ops
- let re_part = if re == 0.0 {
- if re.is_sign_positive() {
- return im_part;
- } else {
- re.to_string()
- }
- } else if re.fract() == 0.0 {
- re.to_string()
- } else {
- to_string(re)
- };
- let mut result =
- String::with_capacity(re_part.len() + im_part.len() + 2 + im.is_sign_positive() as usize);
- result.push('(');
- result.push_str(&re_part);
- if im.is_sign_positive() || im.is_nan() {
- result.push('+');
- }
- result.push_str(&im_part);
- result.push(')');
- result
-}
-
pub fn from_hex(s: &str) -> Option<f64> {
if let Ok(f) = hexf_parse::parse_hexf64(s, false) {
return Some(f);
diff --git a/compiler/literal/src/lib.rs b/compiler/literal/src/lib.rs
index 9b9620573d..2997107012 100644
--- a/compiler/literal/src/lib.rs
+++ b/compiler/literal/src/lib.rs
@@ -1,4 +1,5 @@
pub mod char;
+pub mod complex;
pub mod escape;
pub mod float;
pub mod format;
diff --git a/vm/src/builtins/complex.rs b/vm/src/builtins/complex.rs
index a3a6d4d681..d48707261c 100644
--- a/vm/src/builtins/complex.rs
+++ b/vm/src/builtins/complex.rs
@@ -179,13 +179,13 @@ impl Constructor for PyComplex {
"complex() can't take second arg if first is a string".to_owned(),
));
}
- let value = s
+ let (re, im) = s
.to_str()
- .and_then(|s| parse_str(s.trim()))
+ .and_then(rustpython_literal::complex::parse_str)
.ok_or_else(|| {
vm.new_value_error("complex() arg is a malformed string".to_owned())
})?;
- return Self::from(value)
+ return Self::from(Complex64 { re, im })
.into_ref_with_type(vm, cls)
.map(Into::into);
} else {
@@ -494,7 +494,7 @@ impl Representable for PyComplex {
// TODO: when you fix this, move it to rustpython_common::complex::repr and update
// ast/src/unparse.rs + impl Display for Constant in ast/src/constant.rs
let Complex64 { re, im } = zelf.value;
- Ok(rustpython_literal::float::complex_to_string(re, im))
+ Ok(rustpython_literal::complex::to_string(re, im))
}
}
@@ -519,40 +519,3 @@ pub struct ComplexArgs {
#[pyarg(any, optional)]
imag: OptionalArg<PyObjectRef>,
}
-
-fn parse_str(s: &str) -> Option<Complex64> {
- // Handle parentheses
- let s = match s.strip_prefix('(') {
- None => s,
- Some(s) => match s.strip_suffix(')') {
- None => return None,
- Some(s) => s.trim(),
- },
- };
-
- let value = match s.strip_suffix(|c| c == 'j' || c == 'J') {
- None => Complex64::new(crate::literal::float::parse_str(s)?, 0.0),
- Some(mut s) => {
- let mut real = 0.0;
- // Find the central +/- operator. If it exists, parse the real part.
- for (i, w) in s.as_bytes().windows(2).enumerate() {
- if (w[1] == b'+' || w[1] == b'-') && !(w[0] == b'e' || w[0] == b'E') {
- real = crate::literal::float::parse_str(&s[..=i])?;
- s = &s[i + 1..];
- break;
- }
- }
-
- let imag = match s {
- // "j", "+j"
- "" | "+" => 1.0,
- // "-j"
- "-" => -1.0,
- s => crate::literal::float::parse_str(s)?,
- };
-
- Complex64::new(real, imag)
- }
- };
- Some(value)
-}
diff --git a/vm/src/builtins/float.rs b/vm/src/builtins/float.rs
index 48ccd2c437..27f1f3273f 100644
--- a/vm/src/builtins/float.rs
+++ b/vm/src/builtins/float.rs
@@ -159,9 +159,31 @@ impl Constructor for PyFloat {
}
fn float_from_string(val: PyObjectRef, vm: &VirtualMachine) -> PyResult<f64> {
- let (bytearray, buffer, buffer_lock);
+ let (bytearray, buffer, buffer_lock, mapped_string);
let b = if let Some(s) = val.payload_if_subclass::<PyStr>(vm) {
- s.as_wtf8().trim().as_bytes()
+ use crate::common::str::PyKindStr;
+ match s.as_str_kind() {
+ PyKindStr::Ascii(s) => s.trim().as_bytes(),
+ PyKindStr::Utf8(s) => {
+ mapped_string = s
+ .trim()
+ .chars()
+ .map(|c| {
+ if let Some(n) = rustpython_common::str::char_to_decimal(c) {
+ char::from_digit(n.into(), 10).unwrap()
+ } else if c.is_whitespace() {
+ ' '
+ } else {
+ c
+ }
+ })
+ .collect::<String>();
+ mapped_string.as_bytes()
+ }
+ // if there are surrogates, it's not gonna parse anyway,
+ // so we can just choose a known bad value
+ PyKindStr::Wtf8(_) => b"",
+ }
} else if let Some(bytes) = val.payload_if_subclass::<PyBytes>(vm) {
bytes.as_bytes()
} else if let Some(buf) = val.payload_if_subclass::<PyByteArray>(vm) {