Hashing

The content hashes an asset is identified by — its identity, computed once.

The hashing nodes compute content hashes of a media file. They are the foundation of asset identity in MetaLoom: the SHA-512 hash is the content key used to look an asset up, avoid recomputation, and match against blacklists. Four kinds live in the same module and can be toggled independently.

Kinds

md5, sha256, sha512, chunk-hash

Applies to

Any file

Input ports

media (typically the first processing step after the source)

Output ports

hash — one port per kind, typed hash/md5, hash/sha256, hash/sha512 or hash/chunk. The type is what downstream nodes match on, so a dedup node never has to name the node that produced it

Requirements

CPU only. Streams the file, so memory use is low regardless of file size. Reads the full file bytes from storage.

Persists to

asset row hash columns + asset_node_result ledger

chunk-hash computes a content-based chunked hash used for near-duplicate detection on partially modified files.

Configuration

Figure 1. The node’s settings in the pipeline editor

Per-node options enable each algorithm individually. Set them in the panel above, or in the node’s options block in a pipeline definition:

Option Default Meaning

Option	Default	Meaning
`md5`	off	Compute the MD5 hash
`sha256`	off	Compute the SHA-256 hash
`sha512`	on	Compute the SHA-512 hash (the system-wide content identity)
`chunkHash`	off	Compute the chunked content hash

md5

off

Compute the MD5 hash

sha256

off

Compute the SHA-256 hash

sha512

Compute the SHA-512 hash (the system-wide content identity)

chunkHash

off

Compute the chunked content hash

In online mode a hash node checks Loom first and short-circuits when the hash is already stored.

Seeing it run

Turn on Debug Mode and every node keeps what it produced, on the card itself. Below is a real run of this node over pexels-photo-2379005.jpeg.

The strip on the card lists what each output port carried — hash.

Use Cases

Asset identity — resolve an asset by its SHA-512 across the platform.
Exact deduplication — find byte-identical files (see Deduplication).
Near-duplicate detection — chunk-hash flags files that were partially modified.
Blacklisting — match incoming media against a hash blacklist.

Looking for something else?

All documentation Feature overview