Struct Pipeline 

pub struct Pipeline<'a> { /* private fields */ }

Builder-style write queue. Returned by Store::pipeline; call fluent methods to enqueue + commit() to apply with batched AOF fsync.

Implementations

impl<'a> Pipeline<'a>

pub fn len(&self) -> usize

Number of ops queued so far.

pub fn is_empty(&self) -> bool

true when no ops are queued.

pub fn set(self, key: &[u8], value: &[u8]) -> Self

Queue SET key value. Nothing is applied (or AOF-logged) until commit — true of every enqueue method.

pub fn del(self, keys: &[&[u8]]) -> Self

Queue DEL key [key ...].

pub fn incr(self, key: &[u8]) -> Self

Queue INCR key (commit errors if the value is not an integer).

pub fn incr_by(self, key: &[u8], delta: i64) -> Self

Queue INCRBY key delta; negative delta does DECR-style work.

pub fn hset(self, key: &[u8], pairs: &[(&[u8], &[u8])]) -> Self

Queue HSET key field value [field value ...].

pub fn hdel(self, key: &[u8], fields: &[&[u8]]) -> Self

Queue HDEL key field [field ...].

pub fn hincrby(self, key: &[u8], field: &[u8], delta: i64) -> Self

Queue HINCRBY key field delta.

pub fn zadd(self, key: &[u8], pairs: &[(f64, &[u8])]) -> Self

Queue ZADD key score member [score member ...].

pub fn zadd_flags( self, key: &[u8], pairs: &[(f64, &[u8])], flags: ZaddFlags, ) -> Self

Flags-aware ZADD (v2.1) — e.g. the GT monotonic-heal form.

pub fn zrem(self, key: &[u8], members: &[&[u8]]) -> Self

Queue ZREM key member [member ...].

pub fn zincrby(self, key: &[u8], delta: f64, member: &[u8]) -> Self

Queue ZINCRBY key delta member.

pub fn sadd(self, key: &[u8], members: &[&[u8]]) -> Self

Queue SADD key member [member ...].

pub fn srem(self, key: &[u8], members: &[&[u8]]) -> Self

Queue SREM key member [member ...].

pub fn lpush(self, key: &[u8], values: &[&[u8]]) -> Self

Queue LPUSH key value [value ...].

pub fn rpush(self, key: &[u8], values: &[&[u8]]) -> Self

Queue RPUSH key value [value ...].

pub fn commit(self) -> Result<()>

Apply every queued op in order. Each op acquires its own per-shard write lock — other writers see intermediate states between ops; for transactional semantics use Store::atomic instead. AOF appends batch into one fsync per shard.