git-hash-object

git hash-object [options] [file...]

git hash-object computes the object ID (SHA-1, or SHA-256 in repositories using that format) for a file and optionally stores it in the Git object database. This low-level plumbing command exposes Git's internal object storage mechanism.The command calculates the hash by formatting the file content as a Git object (with type and size header), then hashing it. With the -w option, it also writes the object to .git/objects/, making it part of the repository even if not yet referenced by any commit. The --stdin option enables hashing content from pipes or scripts; --stdin-paths reads a list of file paths instead.

-w

Write object to database.

Read from stdin.

Object type (blob, commit, tree, tag).

Hash as if at path.

Don't apply filters.

Read file paths from stdin, hashing each in turn.

Allow hashing content that wouldn't form a valid object (debugging).

Without -w, the object is only hashed, not stored. Filters (like `clean` in `.gitattributes`) are applied by default unless --no-filters is given, which can change the resulting hash.

git hash-object is a core plumbing command present since Git's earliest releases, exposing the object hashing used internally for blobs, trees, and commits.

git-cat-file(1), git-write-tree(1)

Source code · Homepage · Documentation