Skip to content

Improve Unique[By]/Duplicates[By] debug messages #1121

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

Open
FHTMitchell wants to merge 1 commit into
base: master
Choose a base branch
from
Open

Improve Unique[By]/Duplicates[By] debug messages #1121

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
33 changes: 27 additions & 6 deletions src/duplicates_impl.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,21 +19,42 @@ mod private {
pub(crate) meta: Meta<Key, F, S>,
}

impl<I, V, F, S> fmt::Debug for DuplicatesBy<I, V, F, S>
impl<I, V, S> fmt::Debug for DuplicatesBy<I, V, ById, S>
where
I: Iterator + fmt::Debug,
V: fmt::Debug + Hash + Eq,
S: BuildHasher,
{
debug_fmt_fields!(DuplicatesBy, iter, meta.used);
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("Duplicates")
.field("iter", &self.iter)
.field("seen", &self.meta.seen)
.field("pending", &self.meta.pending)
.finish()
}
}

impl<I, V, G, S> fmt::Debug for DuplicatesBy<I, V, ByFn<G>, S>
where
I: Iterator + fmt::Debug,
V: fmt::Debug + Hash + Eq,
S: BuildHasher,
{
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("DuplicatesBy")
.field("iter", &self.iter)
.field("seen", &self.meta.seen)
.field("pending", &self.meta.pending)
.finish_non_exhaustive()
}
}

impl<I: Iterator, Key: Eq + Hash, F, S: BuildHasher> DuplicatesBy<I, Key, F, S> {
pub(crate) fn new(iter: I, key_method: F, hash_builder: S) -> Self {
Self {
iter,
meta: Meta {
used: HashMap::with_hasher(hash_builder),
seen: HashMap::with_hasher(hash_builder),
pending: 0,
key_method,
},
Expand All @@ -43,7 +64,7 @@ mod private {

#[derive(Clone)]
pub struct Meta<Key, F, S> {
used: HashMap<Key, bool, S>,
seen: HashMap<Key, bool, S>,
pending: usize,
key_method: F,
}
Expand All @@ -61,9 +82,9 @@ mod private {
F: KeyMethod<Key, I>,
{
let kv = self.key_method.make(item);
match self.used.get_mut(kv.key_ref()) {
match self.seen.get_mut(kv.key_ref()) {
None => {
self.used.insert(kv.key(), false);
self.seen.insert(kv.key(), false);
self.pending += 1;
None
}
Expand Down
63 changes: 41 additions & 22 deletions src/unique_impl.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,17 +19,31 @@ where
// Use a Hashmap for the Entry API in order to prevent hashing twice.
// This can maybe be replaced with a HashSet once `get_or_insert_with`
// or a proper Entry API for Hashset is stable and meets this msrv
used: HashMap<V, (), S>,
seen: HashMap<V, (), S>,
f: F,
}

/// Formats the keys of a `HashMap` used as a set, hiding the `()` values.
struct KeySet<'a, V, S>(&'a HashMap<V, (), S>);

impl<V: fmt::Debug, S> fmt::Debug for KeySet<'_, V, S> {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_set().entries(self.0.keys()).finish()
}
}

impl<I, V, F, S> fmt::Debug for UniqueBy<I, V, F, S>
where
I: Iterator + fmt::Debug,
V: fmt::Debug + Hash + Eq,
S: BuildHasher,
{
debug_fmt_fields!(UniqueBy, iter, used);
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("UniqueBy")
.field("iter", &self.iter)
.field("seen", &KeySet(&self.seen))
.finish_non_exhaustive()
}
}

/// Create a new `UniqueBy` iterator.
Expand All @@ -42,22 +56,22 @@ where
{
UniqueBy {
iter,
used: HashMap::with_hasher(hash_builder),
seen: HashMap::with_hasher(hash_builder),
f,
}
}

// count the number of new unique keys in iterable (`used` is the set already seen)
fn count_new_keys<I, K, S>(mut used: HashMap<K, (), S>, iterable: I) -> usize
// count the number of new unique keys in iterable (`seen` is the set already seen)
fn count_new_keys<I, K, S>(mut seen: HashMap<K, (), S>, iterable: I) -> usize
where
I: IntoIterator<Item = K>,
K: Hash + Eq,
S: BuildHasher,
{
let iter = iterable.into_iter();
let current_used = used.len();
used.extend(iter.map(|key| (key, ())));
used.len() - current_used
let current_seen = seen.len();
seen.extend(iter.map(|key| (key, ())));
seen.len() - current_seen
}

impl<I, V, F, S> Iterator for UniqueBy<I, V, F, S>
Expand All @@ -70,19 +84,19 @@ where
type Item = I::Item;

fn next(&mut self) -> Option<Self::Item> {
let Self { iter, used, f } = self;
iter.find(|v| used.insert(f(v), ()).is_none())
let Self { iter, seen, f } = self;
iter.find(|v| seen.insert(f(v), ()).is_none())
}

#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let (low, hi) = self.iter.size_hint();
(usize::from(low > 0 && self.used.is_empty()), hi)
(usize::from(low > 0 && self.seen.is_empty()), hi)
}

fn count(self) -> usize {
let mut key_f = self.f;
count_new_keys(self.used, self.iter.map(move |elt| key_f(&elt)))
count_new_keys(self.seen, self.iter.map(move |elt| key_f(&elt)))
}
}

Expand All @@ -94,8 +108,8 @@ where
S: BuildHasher,
{
fn next_back(&mut self) -> Option<Self::Item> {
let Self { iter, used, f } = self;
iter.rfind(|v| used.insert(f(v), ()).is_none())
let Self { iter, seen, f } = self;
iter.rfind(|v| seen.insert(f(v), ()).is_none())
}
}

Expand All @@ -117,9 +131,9 @@ where
type Item = I::Item;

fn next(&mut self) -> Option<Self::Item> {
let UniqueBy { iter, used, .. } = &mut self.iter;
let UniqueBy { iter, seen, .. } = &mut self.iter;
iter.find_map(|v| {
if let Entry::Vacant(entry) = used.entry(v) {
if let Entry::Vacant(entry) = seen.entry(v) {
let elt = entry.key().clone();
entry.insert(());
return Some(elt);
Expand All @@ -131,11 +145,11 @@ where
#[inline]
fn size_hint(&self) -> (usize, Option<usize>) {
let (low, hi) = self.iter.iter.size_hint();
(usize::from(low > 0 && self.iter.used.is_empty()), hi)
(usize::from(low > 0 && self.iter.seen.is_empty()), hi)
}

fn count(self) -> usize {
count_new_keys(self.iter.used, self.iter.iter)
count_new_keys(self.iter.seen, self.iter.iter)
}
}

Expand All @@ -146,9 +160,9 @@ where
S: BuildHasher,
{
fn next_back(&mut self) -> Option<Self::Item> {
let UniqueBy { iter, used, .. } = &mut self.iter;
let UniqueBy { iter, seen, .. } = &mut self.iter;
iter.rev().find_map(|v| {
if let Entry::Vacant(entry) = used.entry(v) {
if let Entry::Vacant(entry) = seen.entry(v) {
let elt = entry.key().clone();
entry.insert(());
return Some(elt);
Expand Down Expand Up @@ -186,7 +200,12 @@ where
I::Item: Hash + Eq + fmt::Debug + Clone,
S: BuildHasher,
{
debug_fmt_fields!(Unique, iter);
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
f.debug_struct("Unique")
.field("iter", &self.iter.iter)
.field("seen", &KeySet(&self.iter.seen))
.finish()
}
}

pub fn unique_with_hasher<I, S>(iter: I, hash_builder: S) -> Unique<I, S>
Expand All @@ -198,7 +217,7 @@ where
Unique {
iter: UniqueBy {
iter,
used: HashMap::with_hasher(hash_builder),
seen: HashMap::with_hasher(hash_builder),
f: (),
},
}
Expand Down
Loading