git-ledger

Git-native record storage. Each record lives under its own ref with typed fields stored as blobs in a tree.

Usage

Create a record

git ledger create refs/issues --set title="Bug report" --set status=open

With an explicit ID:

git ledger create refs/issues my-id --set title="Named record"

With a content-addressed ID (hashes stdin):

git ledger create refs/issues --content-hash --set title="Deduplicated record"

With a custom commit message:

git ledger create refs/issues --set title="Bug" --set status=open -m "open issue: bug"

Read a record

git ledger read refs/issues/1

Update a record

Set a field:

git ledger update refs/issues/1 --set status=closed

Set multiple fields at once:

git ledger update refs/issues/1 --set status=closed --set resolution=fixed

Delete a field:

git ledger update refs/issues/1 --delete resolution

Mix sets and deletes:

git ledger update refs/issues/1 --set status=wontfix --delete assignee

List records

git ledger list refs/issues

Show history for a record

git ledger log refs/issues/1

Use a different repository

git ledger -C /path/to/repo list refs/issues

Design

A ledger record is a Git ref (e.g. refs/issues/1). Each field is a blob stored in the commit's tree under the field name. create and update both advance the ref with a new commit, so the full field history is preserved in the commit log.

IDs are assigned sequentially by default (1, 2, 3, …). Pass an explicit ID or --content-hash to override.