pub struct CountingReader<R> { /* private fields */ }Expand description
Wraps a reader and tallies how many ranges were requested and how many bytes
were returned — the metric that matters for a range-streamed format.
Atomically counted, so it stays Sync (a lazily-faulting remote index holds
its reader behind a shared loader).
Implementations§
Source§impl<R> CountingReader<R>where
R: RangeReader,
impl<R> CountingReader<R>where
R: RangeReader,
pub fn new(inner: R) -> CountingReader<R>
Sourcepub fn bytes_read(&self) -> u64
pub fn bytes_read(&self) -> u64
Total bytes returned so far.
Trait Implementations§
Source§impl<R> RangeReader for CountingReader<R>where
R: RangeReader,
impl<R> RangeReader for CountingReader<R>where
R: RangeReader,
Source§fn read_many(&self, ranges: &[(u64, u64)]) -> Result<Vec<Vec<u8>>, Error>
fn read_many(&self, ranges: &[(u64, u64)]) -> Result<Vec<Vec<u8>>, Error>
Delegate to the inner reader (preserving its parallelism) and tally each returned range as one request — the count reflects the coalesced spans actually fetched, however the inner reader issues them.
Source§fn is_local(&self) -> bool
fn is_local(&self) -> bool
Is a read cheap and local: bytes already in memory, or a local file?
False (the default) for anything where a read is a round trip. Purely
advisory: an engine may then read from several threads at once, each in
small pieces, where over a network it keeps one scan and batches its
reads. A wrapper reports what it wraps.
Source§fn read_at(&self, offset: u64, len: u64) -> Result<Vec<u8>, Error>
fn read_at(&self, offset: u64, len: u64) -> Result<Vec<u8>, Error>
Read
len bytes starting at offset. Implementations should error on
an out-of-bounds range rather than truncating.Source§fn read_many_scattered(
&self,
ranges: &[(u64, u64)],
) -> Result<Vec<Vec<u8>>, Error>
fn read_many_scattered( &self, ranges: &[(u64, u64)], ) -> Result<Vec<Vec<u8>>, Error>
read_many for ranges the caller knows lie far apart
(the tiles of a window of probes): a reader that rounds reads out to
blocks, or merges ranges across wide gaps, to save requests should fetch
these as they are, since what lies between them is not wanted. The
default is read_many; results are the same bytes
either way, only the requests differ.Source§fn concurrency(&self) -> usize
fn concurrency(&self) -> usize
How many ranges this reader can usefully have in flight at once — the
planner’s hint for probe-vs-scan and batch-size decisions (a phone’s
serial sync-XHR reader reports 1; the CLI’s threaded HTTP client and the
browser’s concurrent-fetch variants report their fan-out). Purely
advisory: correctness never depends on it. Defaults to 1 (sequential).
Source§fn resident_prefix(&self) -> u64
fn resident_prefix(&self) -> u64
How many leading bytes this reader already holds in memory: a read that
lies wholly inside
[0, resident_prefix()) costs no round trip. An HTTP
reader that learnt the length from a ranged GET of the first 128 KiB
reports those 128 KiB. A caching layer above it serves such a read
straight through instead of fetching a block-aligned span that would
re-read what is already here (the open’s first read used to cost a
second, larger request for block 0). Purely advisory; defaults to 0.Auto Trait Implementations§
impl<R> !Freeze for CountingReader<R>
impl<R> Forget for CountingReader<R>
impl<R> Move for CountingReader<R>
impl<R> RefUnwindSafe for CountingReader<R>where
R: RefUnwindSafe,
impl<R> Send for CountingReader<R>where
R: Send,
impl<R> Sync for CountingReader<R>where
R: Sync,
impl<R> Unpin for CountingReader<R>where
R: Unpin,
impl<R> UnsafeUnpin for CountingReader<R>where
R: UnsafeUnpin,
impl<R> UnwindSafe for CountingReader<R>where
R: UnwindSafe,
Blanket Implementations§
Source§impl<T> BorrowMut<T> for Twhere
T: ?Sized,
impl<T> BorrowMut<T> for Twhere
T: ?Sized,
Source§fn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more