Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
1 change: 0 additions & 1 deletion Lib/test/test_re.py
Original file line number Diff line number Diff line change
Expand Up @@ -66,7 +66,6 @@ def checkTemplateError(self, pattern, repl, string, errmsg, pos=None):
def test_error_is_PatternError_alias(self):
assert re.error is re.PatternError

@unittest.expectedFailure # TODO: RUSTPYTHON
def test_keep_buffer(self):
# See bug 14212
b = bytearray(b'x')
Expand Down
115 changes: 70 additions & 45 deletions crates/vm/src/stdlib/_sre.rs
Original file line number Diff line number Diff line change
Expand Up @@ -394,11 +394,16 @@ mod _sre {
}

macro_rules! with_sre_str {
($pattern:expr, $string:expr, $vm:expr, $f:expr) => {{
($pattern:expr, $string:expr, $vm:expr, $f:expr) => {
with_sre_str!($pattern, $string, $vm, $f, None::<&PyBuffer>)
};
($pattern:expr, $string:expr, $vm:expr, $f:expr, $buffer:expr) => {{
// Bind once: the branches only borrow the subject, and callers pass
// a temporary (`&x.clone()`) that would otherwise be rebuilt per arm.
let subject = $string;
if $pattern.isbytes {
if let Some(buffer) = $buffer {
buffer.contiguous_or_collect($f)
} else if $pattern.isbytes {
Pattern::with_bytes(subject, $vm, $f)
} else if Pattern::is_ascii_str(subject) {
Pattern::with_ascii_str(subject, $vm, $f)
Expand Down Expand Up @@ -647,14 +652,7 @@ mod _sre {
string_args: StringArgs,
vm: &VirtualMachine,
) -> PyResult<PyCallableIterator> {
let scanner = SreScanner {
pattern: zelf,
string: string_args.string,
start: AtomicCell::new(string_args.pos.0),
end: string_args.endpos.0,
must_advance: AtomicCell::new(false),
}
.into_ref(&vm.ctx);
let scanner = Self::scanner(zelf, string_args, vm)?;
let search = vm.get_str_method(scanner.into(), "search").unwrap()?;
let search = ArgCallable::try_from_object(vm, search)?;
let iterator = PyCallableIterator::new(search, vm.ctx.none());
Expand All @@ -666,15 +664,26 @@ mod _sre {
zelf: PyRef<Pattern>,
string_args: StringArgs,
vm: &VirtualMachine,
) -> PyRef<SreScanner> {
SreScanner {
) -> PyResult<PyRef<SreScanner>> {
let buffer = if zelf.isbytes {
Some(PyBuffer::from_object(
vm,
&string_args.string,
BufferFlags::SIMPLE,
)?)
} else {
Pattern::downcast_str(&string_args.string, vm)?;
None
};
Ok(SreScanner {
pattern: zelf,
string: string_args.string,
buffer,
start: AtomicCell::new(string_args.pos.0),
end: string_args.endpos.0,
must_advance: AtomicCell::new(false),
}
.into_ref(&vm.ctx)
.into_ref(&vm.ctx))
}

#[pymethod]
Expand Down Expand Up @@ -1142,13 +1151,17 @@ mod _sre {
}

#[pyattr]
#[pyclass(name = "SRE_Scanner")]
#[pyclass(name = "SRE_Scanner", traverse)]
#[derive(Debug, PyPayload)]
struct SreScanner {
pattern: PyRef<Pattern>,
string: PyObjectRef,
buffer: Option<PyBuffer>,
#[pytraverse(skip)]
start: AtomicCell<usize>,
#[pytraverse(skip)]
end: usize,
#[pytraverse(skip)]
must_advance: AtomicCell<bool>,
}

Expand All @@ -1165,44 +1178,56 @@ mod _sre {
#[pymethod(name = "prefixmatch")]
#[pymethod(name = "match")]
fn py_match(&self, vm: &VirtualMachine) -> PyResult<Option<PyRef<Match>>> {
with_sre_str!(self.pattern, &self.string.clone(), vm, |s| {
let mut req = s.create_request(&self.pattern, self.start.load(), self.end);
let mut state = State::default();
req.must_advance = self.must_advance.load();
let has_matched = state.py_match(&req);

self.must_advance
.store(state.cursor.position == state.start);
self.start.store(state.cursor.position);

Ok(has_matched.then(|| {
Match::new(&mut state, self.pattern.clone(), self.string.clone())
.into_ref(&vm.ctx)
}))
})
with_sre_str!(
self.pattern,
&self.string.clone(),
vm,
|s| {
let mut req = s.create_request(&self.pattern, self.start.load(), self.end);
let mut state = State::default();
req.must_advance = self.must_advance.load();
let has_matched = state.py_match(&req);

self.must_advance
.store(state.cursor.position == state.start);
self.start.store(state.cursor.position);

Ok(has_matched.then(|| {
Match::new(&mut state, self.pattern.clone(), self.string.clone())
.into_ref(&vm.ctx)
}))
},
self.buffer.as_ref()
)
}

#[pymethod]
fn search(&self, vm: &VirtualMachine) -> PyResult<Option<PyRef<Match>>> {
if self.start.load() > self.end {
return Ok(None);
}
with_sre_str!(self.pattern, &self.string.clone(), vm, |s| {
let mut req = s.create_request(&self.pattern, self.start.load(), self.end);
let mut state = State::default();
req.must_advance = self.must_advance.load();

let has_matched = state.search(req);

self.must_advance
.store(state.cursor.position == state.start);
self.start.store(state.cursor.position);

Ok(has_matched.then(|| {
Match::new(&mut state, self.pattern.clone(), self.string.clone())
.into_ref(&vm.ctx)
}))
})
with_sre_str!(
self.pattern,
&self.string.clone(),
vm,
|s| {
let mut req = s.create_request(&self.pattern, self.start.load(), self.end);
let mut state = State::default();
req.must_advance = self.must_advance.load();

let has_matched = state.search(req);

self.must_advance
.store(state.cursor.position == state.start);
self.start.store(state.cursor.position);

Ok(has_matched.then(|| {
Match::new(&mut state, self.pattern.clone(), self.string.clone())
.into_ref(&vm.ctx)
}))
},
self.buffer.as_ref()
)
}
}
}
Loading