initial commit

Signed-off-by: Alex Chi <iskyzh@gmail.com>
This commit is contained in:
Alex Chi
2022-12-23 14:29:32 -05:00
commit 68303e99b8
17 changed files with 849 additions and 0 deletions

View File

@@ -0,0 +1,57 @@
use bytes::BufMut;
use super::{Block, SIZEOF_U16};
/// Builds a block
pub struct BlockBuilder {
offsets: Vec<u16>,
data: Vec<u8>,
target_size: usize,
}
impl BlockBuilder {
/// Creates a new block builder
pub fn new(target_size: usize) -> Self {
Self {
offsets: Vec::new(),
data: Vec::new(),
target_size,
}
}
fn estimated_size(&self) -> usize {
self.offsets.len() * SIZEOF_U16 + self.data.len() + SIZEOF_U16
}
/// Adds a key-value pair to the block
#[must_use]
pub fn add(&mut self, key: &[u8], value: &[u8]) -> bool {
assert!(!key.is_empty(), "key must not be empty");
if self.estimated_size() + key.len() + value.len() + SIZEOF_U16 * 3 > self.target_size
&& !self.is_empty()
{
return false;
}
self.offsets.push(self.data.len() as u16);
self.data.put_u16(key.len() as u16);
self.data.put(key);
self.data.put_u16(value.len() as u16);
self.data.put(value);
true
}
pub fn is_empty(&self) -> bool {
self.offsets.is_empty()
}
/// Builds a block
pub fn build(self) -> Block {
if self.is_empty() {
panic!("block should not be empty");
}
Block {
data: self.data,
offsets: self.offsets,
}
}
}

View File

@@ -0,0 +1,110 @@
use std::sync::Arc;
use bytes::Buf;
use super::Block;
pub struct BlockIterator {
block: Arc<Block>,
key: Vec<u8>,
value: Vec<u8>,
idx: usize,
}
impl BlockIterator {
fn new(block: Arc<Block>) -> Self {
Self {
block,
key: Vec::new(),
value: Vec::new(),
idx: 0,
}
}
pub fn create_and_seek_to_first(block: Arc<Block>) -> Self {
let mut iter = Self::new(block);
iter.seek_to_first();
iter
}
pub fn create_and_seek_to_key(block: Arc<Block>, key: &[u8]) -> Self {
let mut iter = Self::new(block);
iter.seek_to_key(key);
iter
}
pub fn key(&self) -> &[u8] {
debug_assert!(!self.key.is_empty(), "invalid iterator");
&self.key
}
pub fn value(&self) -> &[u8] {
debug_assert!(!self.key.is_empty(), "invalid iterator");
&self.value
}
pub fn is_valid(&self) -> bool {
!self.key.is_empty()
}
pub fn seek_to_first(&mut self) {
self.seek_to(0);
}
pub fn seek_to_last(&mut self) {
self.seek_to(self.block.offsets.len() - 1);
}
pub fn len(&self) -> usize {
self.block.offsets.len()
}
pub fn is_empty(&self) -> bool {
self.block.offsets.is_empty()
}
pub fn seek_to(&mut self, idx: usize) {
if idx >= self.block.offsets.len() {
self.key.clear();
self.value.clear();
return;
}
let offset = self.block.offsets[idx] as usize;
self.seek_to_offset(offset);
self.idx = idx;
}
pub fn next(&mut self) {
self.idx += 1;
self.seek_to(self.idx);
}
fn seek_to_offset(&mut self, offset: usize) {
let mut entry = &self.block.data[offset..];
let key_len = entry.get_u16() as usize;
let key = entry[..key_len].to_vec();
entry.advance(key_len);
self.key.clear();
self.key.extend(key);
let value_len = entry.get_u16() as usize;
let value = entry[..value_len].to_vec();
entry.advance(value_len);
self.value.clear();
self.value.extend(value);
}
pub fn seek_to_key(&mut self, key: &[u8]) {
let mut low = 0;
let mut high = self.block.offsets.len();
while low < high {
let mid = low + (high - low) / 2;
self.seek_to(mid);
assert!(self.is_valid());
match self.key().cmp(key) {
std::cmp::Ordering::Less => low = mid + 1,
std::cmp::Ordering::Greater => high = mid,
std::cmp::Ordering::Equal => return,
}
}
}
}