Struct OdbcBufferSettings 

pub struct OdbcBufferSettings {
    pub batch_size: usize,
    pub max_column_size: Option<usize>,
}

Available on crate feature odbc only.

Configuration for ODBC buffer settings that control memory usage and performance characteristics.

These settings affect how SQLx fetches and processes data from ODBC data sources. Careful tuning of these parameters can significantly impact memory usage and query performance.

Buffered vs Unbuffered Mode

The buffer settings control two different data fetching strategies based on max_column_size:

Buffered Mode (when max_column_size is Some(value)):

• Fetches rows in batches using columnar buffers for better performance with large result sets
• Pre-allocates memory for all columns and rows in each batch using ColumnarAnyBuffer
• WARNING: Long textual and binary field data will be truncated to max_column_size
• Better for high-throughput scenarios where memory usage is acceptable
• Uses batch_size to control how many rows are fetched at once

Unbuffered Mode (when max_column_size is None):

• Fetches rows in batches using the next_row() method for memory efficiency
• Processes rows in batches of batch_size but without pre-allocating columnar buffers
• No data truncation - handles variable-sized data dynamically
• More memory efficient than buffered mode but potentially slower
• Better for scenarios where data sizes vary significantly or memory is constrained

Fields

batch_size: usize

The number of rows to fetch in each batch during bulk operations.

Performance Impact:

• Higher values reduce the number of round-trips to the database but increase memory usage
• Lower values reduce memory usage but may increase latency due to more frequent fetches
• Typical range: 32-512 rows
• Used in both buffered and unbuffered modes

Memory Impact:

• In buffered mode: Each batch allocates buffers for batch_size * number_of_columns cells
• In unbuffered mode: Each batch processes batch_size rows without pre-allocation
• For wide result sets, this can consume significant memory

Default: 128 rows

max_column_size: Option<usize>

The maximum size (in characters) for text and binary columns when the database doesn’t specify a length.

Performance Impact:

• When Some(value): Enables buffered mode with batch fetching and pre-allocated buffers
• When None: Enables unbuffered mode with batched row-by-row processing
• Higher values ensure large text fields are fully captured but increase memory allocation
• Lower values may truncate data but reduce memory pressure
• Affects VARCHAR, NVARCHAR, TEXT, and BLOB column types

Memory Impact:

• When Some(value): Directly controls buffer size for variable-length columns
• When None: Minimal memory allocation per row, no truncation
• Setting too high can waste memory; setting too low can cause data truncation
• Consider your data characteristics when tuning this value

Default: Some(4096)

Trait Implementations

impl Clone for OdbcBufferSettings

fn clone(&self) -> OdbcBufferSettings

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for OdbcBufferSettings

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

Formats the value using the given formatter.

impl Default for OdbcBufferSettings

fn default() -> Self

Returns the “default value” for a type.

impl PartialEq for OdbcBufferSettings

fn eq(&self, other: &OdbcBufferSettings) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl Copy for OdbcBufferSettings

impl Eq for OdbcBufferSettings

impl StructuralPartialEq for OdbcBufferSettings