Struct Dump 

pub struct Dump { /* private fields */ }

A PostgreSQL dump archive.

Implementations

impl Dump

pub fn load<P: AsRef<Path>>(path: P) -> Result<Self>

Load a dump from a file or directory.

Automatically detects the format:

• Directory → directory format (-Fd)
• File starting with PGDMP → custom format (-Fc)
• File with ustar header → tar format (-Ft)

pub fn new(dbname: &str, encoding: &str, appear_as: &str) -> Result<Self>

Create a new empty dump.

pub fn save<P: AsRef<Path>>(&self, path: P) -> Result<()>

Save the dump to a file or directory.

For custom format (-Fc), writes to a temporary file first, then atomically renames to avoid leaving a partial file on failure. For directory format (-Fd), writes directly to the directory.

pub fn set_format(&mut self, format: Format)

Set the output format for writing.

Note: Tar format does not support compression. If compression is set when switching to tar format, it will be reset to None.

pub fn version(&self) -> ArchiveVersion

The archive format version.

pub fn compression(&self) -> CompressionAlgorithm

The compression algorithm used.

pub fn dbname(&self) -> &str

The database name.

pub fn server_version(&self) -> &str

The PostgreSQL server version string.

pub fn dump_version(&self) -> &str

The pg_dump version string.

pub fn timestamp(&self) -> &Timestamp

The archive creation timestamp.

pub fn entries(&self) -> &[Entry]

All TOC entries.

pub fn lookup_entry( &self, desc: &ObjectType, namespace: &str, tag: &str, ) -> Option<&Entry>

Look up an entry by object type, namespace, and tag.

pub fn get_entry(&self, dump_id: i32) -> Option<&Entry>

Get an entry by dump_id.

pub fn get_entry_mut(&mut self, dump_id: i32) -> Option<&mut Entry>

Get a mutable reference to an entry by dump_id.

pub fn table_data( &self, namespace: &str, table: &str, ) -> Result<impl Iterator<Item = &str>>

Iterate over table data rows for the given namespace and table.

Each yielded item is a line from the COPY data (without the trailing newline).

pub fn blobs(&self) -> Vec<(i32, &[u8])>

Iterate over all large objects (blobs) across all BLOBS entries.

Yields (oid, data) pairs where oid is the large object OID and data is the decompressed blob content.

pub fn add_blob(&mut self, oid: i32, data: Vec<u8>) -> Result<i32>

Add a large object (blob) to a BLOBS entry.

If no BLOBS entry exists yet, one is created automatically. Returns the OID of the added blob.

pub fn entry_data(&self, dump_id: i32) -> Option<&[u8]>

Get the raw data bytes for an entry by dump_id.

pub fn add_entry( &mut self, desc: ObjectType, namespace: Option<&str>, tag: Option<&str>, owner: Option<&str>, defn: Option<&str>, drop_stmt: Option<&str>, copy_stmt: Option<&str>, dependencies: &[i32], ) -> Result<i32>

Add a new TOC entry.

pub fn set_entry_data(&mut self, dump_id: i32, data: Vec<u8>) -> Result<()>

Set the data for an entry (raw COPY format bytes).

pub fn set_compression(&mut self, alg: CompressionAlgorithm)

Set compression algorithm for writing.

pub fn sort_entries(&mut self)

Sort TOC entries using the same weighted topological sort as pg_dump.

Entries are first sorted by object-type priority (schema before table, table before index, etc.), then by namespace and name. A topological sort pass then reorders only as needed to satisfy the dependency graph, preserving the cosmetic ordering wherever possible.

This is called automatically by Dump::save, but can also be called manually if you need the sorted order before writing.

Trait Implementations

impl Debug for Dump

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

Formats the value using the given formatter.