Struct PartitionIndexEntry 

pub struct PartitionIndexEntry {
    pub key_digest: Arc<[u8]>,
    pub raw_key: Option<Arc<[u8]>>,
    pub data_offset: u64,
    pub data_size: u32,
    pub promoted_index: Option<PromotedIndexData>,
}

Partition index entry in Index.db

Fields

key_digest: Arc<[u8]>

Raw partition key bytes (length-prefixed in the on-disk BIG/NB Index.db format).

NOTE (Issue #552): Despite the historical field name key_digest, this holds the RAW partition key bytes, not an MD5 digest. The real Cassandra 5.0 NB Index.db entry format is [key_len: u16 BE][raw key bytes][data_offset: vint][promoted_len: vint]. There is no 0x0010 marker and no MD5 digest on disk. The field name is retained to avoid churn in the zero-copy lookup table and downstream callers; it is used directly as the partition key (e.g. for RowKey). The leading u16 is the key length (e.g. 0x0010 for a 16-byte UUID, 0x0026 for a 38-byte composite key).

raw_key: Option<Arc<[u8]>>

Raw partition key bytes (mirror of key_digest, kept for API compatibility). Always Some now that all entries carry their raw key.

data_offset: u64

Offset in Data.db file

data_size: u32

Size of partition data

promoted_index: Option<PromotedIndexData>

Promoted index entries for wide partitions (optional)

Trait Implementations

impl Clone for PartitionIndexEntry

fn clone(&self) -> PartitionIndexEntry

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for PartitionIndexEntry

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.