Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

implement range based iteration #297

Merged
merged 1 commit into from
Nov 27, 2024
Merged

implement range based iteration #297

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
20 changes: 10 additions & 10 deletions roaring/src/bitmap/inherent.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -94,8 +94,8 @@ impl RoaringBitmap {
R: RangeBounds<u32>,
{
let (start, end) = match util::convert_range_to_inclusive(range) {
Some(range) => (*range.start(), *range.end()),
None => return 0,
Ok(range) => (*range.start(), *range.end()),
Err(_) => return 0,
};

let (start_container_key, start_index) = util::split(start);
Expand Down Expand Up @@ -239,8 +239,8 @@ impl RoaringBitmap {
R: RangeBounds<u32>,
{
let (start, end) = match util::convert_range_to_inclusive(range) {
Some(range) => (*range.start(), *range.end()),
None => return 0,
Ok(range) => (*range.start(), *range.end()),
Err(_) => return 0,
};

let (start_container_key, start_index) = util::split(start);
Expand Down Expand Up @@ -309,9 +309,9 @@ impl RoaringBitmap {
R: RangeBounds<u32>,
{
let (start, end) = match util::convert_range_to_inclusive(range) {
Some(range) => (*range.start(), *range.end()),
// Empty ranges are always contained
None => return true,
Ok(range) => (*range.start(), *range.end()),
// Empty/Invalid ranges are always contained
Err(_) => return true,
};
let (start_high, start_low) = util::split(start);
let (end_high, end_low) = util::split(end);
Expand Down Expand Up @@ -369,9 +369,9 @@ impl RoaringBitmap {
R: RangeBounds<u32>,
{
let (start, end) = match util::convert_range_to_inclusive(range) {
Some(range) => (*range.start(), *range.end()),
// Empty ranges have 0 bits set in them
None => return 0,
Ok(range) => (*range.start(), *range.end()),
// Empty/invalid ranges have 0 bits set in them
Err(_) => return 0,
};

let (start_key, start_low) = util::split(start);
Expand Down
120 changes: 120 additions & 0 deletions roaring/src/bitmap/iter.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
use alloc::vec;
use core::iter::FusedIterator;
use core::ops::RangeBounds;
use core::slice;

use super::container::Container;
Expand Down Expand Up @@ -155,6 +156,10 @@ impl Iter<'_> {
Iter { front: None, containers: containers.iter(), back: None }
}

fn empty() -> Self {
Self::new(&[])
}

/// Advance the iterator to the first position where the item has a value >= `n`
///
/// # Examples
Expand Down Expand Up @@ -199,6 +204,10 @@ impl IntoIter {
IntoIter { front: None, containers: containers.into_iter(), back: None }
}

fn empty() -> Self {
Self::new(Vec::new())
}

/// Advance the iterator to the first position where the item has a value >= `n`
///
/// # Examples
Expand Down Expand Up @@ -552,6 +561,117 @@ impl RoaringBitmap {
pub fn iter(&self) -> Iter {
Iter::new(&self.containers)
}

/// Iterator over values within a range stored in the RoaringBitmap.
///
/// # Examples
///
/// ```rust
/// use core::ops::Bound;
/// use roaring::RoaringBitmap;
///
/// let bitmap = RoaringBitmap::from([0, 1, 2, 3, 4, 5, 10, 11, 12, 20, 21, u32::MAX]);
/// let mut iter = bitmap.range(10..20);
///
/// assert_eq!(iter.next(), Some(10));
/// assert_eq!(iter.next(), Some(11));
/// assert_eq!(iter.next(), Some(12));
/// assert_eq!(iter.next(), None);
///
/// let mut iter = bitmap.range(100..);
/// assert_eq!(iter.next(), Some(u32::MAX));
/// assert_eq!(iter.next(), None);
///
/// let mut iter = bitmap.range((Bound::Excluded(0), Bound::Included(10)));
/// assert_eq!(iter.next(), Some(1));
/// assert_eq!(iter.next(), Some(2));
/// assert_eq!(iter.next(), Some(3));
/// assert_eq!(iter.next(), Some(4));
/// assert_eq!(iter.next(), Some(5));
/// assert_eq!(iter.next(), Some(10));
/// assert_eq!(iter.next(), None);
/// ```
pub fn range<R>(&self, range: R) -> Iter<'_>
where
R: RangeBounds<u32>,
{
let range = match util::convert_range_to_inclusive(range) {
Ok(range) => range,
Err(util::ConvertRangeError::Empty) => return Iter::empty(),
Err(util::ConvertRangeError::StartGreaterThanEnd) => {
panic!("range start is greater than range end")
}
Err(util::ConvertRangeError::StartAndEndEqualExcluded) => {
panic!("range start and end are equal and excluded")
}
};
let (start, end) = (*range.start(), *range.end());
let mut iter = self.iter();
if start != 0 {
iter.advance_to(start);
}
if end != u32::MAX {
iter.advance_back_to(end);
}
iter
}

/// Iterator over values within a range stored in the RoaringBitmap.
///
/// # Examples
///
/// ```rust
/// use core::ops::Bound;
/// use roaring::RoaringBitmap;
///
/// fn bitmap() -> RoaringBitmap {
/// RoaringBitmap::from([0, 1, 2, 3, 4, 5, 10, 11, 12, 20, 21, u32::MAX])
/// }
///
/// let mut iter = bitmap().into_range(10..20);
///
/// assert_eq!(iter.next(), Some(10));
/// assert_eq!(iter.next(), Some(11));
/// assert_eq!(iter.next(), Some(12));
/// assert_eq!(iter.next(), None);
///
/// let mut iter = bitmap().into_range(100..);
/// assert_eq!(iter.next(), Some(u32::MAX));
/// assert_eq!(iter.next(), None);
///
/// let mut iter = bitmap().into_range((Bound::Excluded(0), Bound::Included(10)));
/// assert_eq!(iter.next(), Some(1));
/// assert_eq!(iter.next(), Some(2));
/// assert_eq!(iter.next(), Some(3));
/// assert_eq!(iter.next(), Some(4));
/// assert_eq!(iter.next(), Some(5));
/// assert_eq!(iter.next(), Some(10));
/// assert_eq!(iter.next(), None);
/// ```
pub fn into_range<R>(self, range: R) -> IntoIter
where
R: RangeBounds<u32>,
{
let range = match util::convert_range_to_inclusive(range) {
Ok(range) => range,
Err(util::ConvertRangeError::Empty) => return IntoIter::empty(),
Err(util::ConvertRangeError::StartGreaterThanEnd) => {
panic!("range start is greater than range end")
}
Err(util::ConvertRangeError::StartAndEndEqualExcluded) => {
panic!("range start and end are equal and excluded")
}
};
let (start, end) = (*range.start(), *range.end());
let mut iter = self.into_iter();
if start != 0 {
iter.advance_to(start);
}
if end != u32::MAX {
iter.advance_back_to(end);
}
iter
}
}

impl<'a> IntoIterator for &'a RoaringBitmap {
Expand Down
85 changes: 57 additions & 28 deletions roaring/src/bitmap/util.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -14,30 +14,56 @@ pub fn join(high: u16, low: u16) -> u32 {
(u32::from(high) << 16) + u32::from(low)
}

#[derive(Debug, Copy, Clone, PartialEq, Eq)]
pub enum ConvertRangeError {
Empty,
StartGreaterThanEnd,
StartAndEndEqualExcluded,
}

/// Convert a `RangeBounds<u32>` object to `RangeInclusive<u32>`,
pub fn convert_range_to_inclusive<R>(range: R) -> Option<RangeInclusive<u32>>
pub fn convert_range_to_inclusive<R>(range: R) -> Result<RangeInclusive<u32>, ConvertRangeError>
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Indeed the code is not pretty but it is very useful, thank you 😊

where
R: RangeBounds<u32>,
{
let start: u32 = match range.start_bound() {
Bound::Included(&i) => i,
Bound::Excluded(&i) => i.checked_add(1)?,
Bound::Unbounded => 0,
};
let end: u32 = match range.end_bound() {
Bound::Included(&i) => i,
Bound::Excluded(&i) => i.checked_sub(1)?,
Bound::Unbounded => u32::MAX,
};
if end < start {
return None;
let start_bound = range.start_bound().cloned();
let end_bound = range.end_bound().cloned();
match (start_bound, end_bound) {
(Bound::Excluded(s), Bound::Excluded(e)) if s == e => {
Err(ConvertRangeError::StartAndEndEqualExcluded)
}
(Bound::Included(s) | Bound::Excluded(s), Bound::Included(e) | Bound::Excluded(e))
if s > e =>
{
Err(ConvertRangeError::StartGreaterThanEnd)
}
_ => {
let start = match start_bound {
Bound::Included(s) => s,
Bound::Excluded(s) => s.checked_add(1).ok_or(ConvertRangeError::Empty)?,
Bound::Unbounded => 0,
};

let end = match end_bound {
Bound::Included(e) => e,
Bound::Excluded(e) => e.checked_sub(1).ok_or(ConvertRangeError::Empty)?,
Bound::Unbounded => u32::MAX,
};

if start > end {
// This handles e.g. `x..x`: we've ruled out `start > end` overall, so a value must
// have been changed via exclusion.
Err(ConvertRangeError::Empty)
} else {
Ok(start..=end)
}
}
}
Some(start..=end)
}

#[cfg(test)]
mod test {
use super::{convert_range_to_inclusive, join, split};
use super::{convert_range_to_inclusive, join, split, ConvertRangeError};
use core::ops::Bound;

#[test]
Expand Down Expand Up @@ -67,27 +93,30 @@ mod test {
#[test]
#[allow(clippy::reversed_empty_ranges)]
fn test_convert_range_to_inclusive() {
assert_eq!(Some(1..=5), convert_range_to_inclusive(1..6));
assert_eq!(Some(1..=u32::MAX), convert_range_to_inclusive(1..));
assert_eq!(Some(0..=u32::MAX), convert_range_to_inclusive(..));
assert_eq!(Some(16..=16), convert_range_to_inclusive(16..=16));
assert_eq!(Ok(1..=5), convert_range_to_inclusive(1..6));
assert_eq!(Ok(1..=u32::MAX), convert_range_to_inclusive(1..));
assert_eq!(Ok(0..=u32::MAX), convert_range_to_inclusive(..));
assert_eq!(Ok(16..=16), convert_range_to_inclusive(16..=16));
assert_eq!(
Some(11..=19),
Ok(11..=19),
convert_range_to_inclusive((Bound::Excluded(10), Bound::Excluded(20)))
);

assert_eq!(None, convert_range_to_inclusive(0..0));
assert_eq!(None, convert_range_to_inclusive(5..5));
assert_eq!(None, convert_range_to_inclusive(1..0));
assert_eq!(None, convert_range_to_inclusive(10..5));
assert_eq!(Err(ConvertRangeError::Empty), convert_range_to_inclusive(0..0));
assert_eq!(Err(ConvertRangeError::Empty), convert_range_to_inclusive(5..5));
assert_eq!(Err(ConvertRangeError::StartGreaterThanEnd), convert_range_to_inclusive(1..0));
assert_eq!(Err(ConvertRangeError::StartGreaterThanEnd), convert_range_to_inclusive(10..5));
assert_eq!(
None,
Err(ConvertRangeError::Empty),
convert_range_to_inclusive((Bound::Excluded(u32::MAX), Bound::Included(u32::MAX)))
);
assert_eq!(
None,
convert_range_to_inclusive((Bound::Excluded(u32::MAX), Bound::Included(u32::MAX)))
Err(ConvertRangeError::StartAndEndEqualExcluded),
convert_range_to_inclusive((Bound::Excluded(u32::MAX), Bound::Excluded(u32::MAX)))
);
assert_eq!(
Err(ConvertRangeError::Empty),
convert_range_to_inclusive((Bound::Excluded(0), Bound::Included(0)))
);
assert_eq!(None, convert_range_to_inclusive((Bound::Excluded(0), Bound::Included(0))));
}
}
Loading
Loading