Struct ConfigValueGroup 

pub struct ConfigValueGroup {

    pub min_spacing_between_global_dedup_queries: usize,
    pub local_cas_scheme: String,
    pub max_concurrent_file_ingestion: usize,
    pub max_concurrent_file_downloads: usize,
    pub ingestion_block_size: ByteSize,
    pub progress_update_interval: Duration,
    pub progress_update_speed_sampling_window: Duration,
    pub progress_update_speed_min_observations: u32,
    pub session_xorb_metadata_flush_interval: Duration,
    pub session_xorb_metadata_flush_max_count: usize,
    pub default_cas_endpoint: String,
    pub aggregate_progress: bool,
    pub default_prefix: String,
    pub staging_subdir: String,
}

ConfigValueGroup struct containing all configurable values

Fields

min_spacing_between_global_dedup_queries: usize

Gives the minimum spacing in number of chunks between global dedup queries sent to the server to limit the number of simultaneous queries.

The default value is 256, which means that the server will receive a query at most for every 256 chunks or 4MB of data.

Use the environment variable HF_XET_DATA_MIN_SPACING_BETWEEN_GLOBAL_DEDUP_QUERIES to set this value.

local_cas_scheme: String

scheme for a local filesystem based CAS server

The default value is “local://”.

Use the environment variable HF_XET_DATA_LOCAL_CAS_SCHEME to set this value.

max_concurrent_file_ingestion: usize

The maximum number of files to ingest at once on the upload path. High performance mode (enabled via HF_XET_HIGH_PERFORMANCE or HF_XET_HP) automatically sets this to 100 via XetConfig::with_high_performance().

The default value is 8.

Use the environment variable HF_XET_DATA_MAX_CONCURRENT_FILE_INGESTION to set this value.

max_concurrent_file_downloads: usize

The maximum number of files to ingest at once on the download path.

The default value is 8.

Use the environment variable HF_XET_DATA_MAX_CONCURRENT_FILE_DOWNLOADS to set this value.

ingestion_block_size: ByteSize

The maximum block size from a file to process at once.

The default value is 8mb.

Use the environment variable HF_XET_DATA_INGESTION_BLOCK_SIZE to set this value.

progress_update_interval: Duration

How often to send updates on file progress, in milliseconds. Disables batching if set to 0.

The default value is 200ms.

Use the environment variable HF_XET_DATA_PROGRESS_UPDATE_INTERVAL to set this value.

progress_update_speed_sampling_window: Duration

Half-life duration for the exponentially weighted moving average used to estimate progress completion speed. Older rate observations are exponentially decayed with this half-life.

The default value is 10sec.

Use the environment variable HF_XET_DATA_PROGRESS_UPDATE_SPEED_SAMPLING_WINDOW to set this value.

progress_update_speed_min_observations: u32

Minimum number of speed observations before reporting a rate. Until this many updates have been recorded, the completion rate is reported as unknown (None). This avoids displaying noisy initial estimates.

The default value is 4.

Use the environment variable HF_XET_DATA_PROGRESS_UPDATE_SPEED_MIN_OBSERVATIONS to set this value.

session_xorb_metadata_flush_interval: Duration

How often do we flush new xorb data to disk on a long running upload session?

The default value is 20sec.

Use the environment variable HF_XET_DATA_SESSION_XORB_METADATA_FLUSH_INTERVAL to set this value.

session_xorb_metadata_flush_max_count: usize

Force a flush of the xorb metadata every this many xorbs, if more are created in this time window.

The default value is 64.

Use the environment variable HF_XET_DATA_SESSION_XORB_METADATA_FLUSH_MAX_COUNT to set this value.

default_cas_endpoint: String

Default CAS endpoint

The default value is “http://localhost:8080”.

Use the environment variable HF_XET_DATA_DEFAULT_CAS_ENDPOINT to set this value.

aggregate_progress: bool

Whether to aggregate progress updates before sending them. When enabled, progress updates are batched and sent at regular intervals to reduce overhead.

The default value is true.

Use the environment variable HF_XET_DATA_AGGREGATE_PROGRESS to set this value.

default_prefix: String

Default prefix used for CAS and shard operations.

The default value is “default”.

Use the environment variable HF_XET_DATA_DEFAULT_PREFIX to set this value.

staging_subdir: String

Subdirectory name for staging data within the endpoint cache directory.

The default value is “staging”.

Use the environment variable HF_XET_DATA_STAGING_SUBDIR to set this value.

Implementations

impl ConfigValueGroup

pub fn new() -> Self

Create a new instance with default values only (no environment variable overrides).

pub fn apply_env_overrides(&mut self)

Apply environment variable overrides to this configuration group.

The group name is derived from the module path. For example, in module xet_config::groups::data, the env var for TEST_INT would be HF_XET_DATA_TEST_INT.

pub fn field_names() -> &'static [&'static str]

Returns the list of field names in this configuration group.

pub fn get(&self, name: &str) -> Result<String, ConfigError>

Get a configuration field’s string representation by name.

Trait Implementations

impl AsRef<ConfigValueGroup> for ConfigValueGroup

fn as_ref(&self) -> &ConfigValueGroup

Converts this type into a shared reference of the (usually inferred) input type.

impl Clone for ConfigValueGroup

fn clone(&self) -> ConfigValueGroup

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for ConfigValueGroup

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

Formats the value using the given formatter.

impl Default for ConfigValueGroup

fn default() -> Self

Returns the “default value” for a type.