The library: triples, content addressing, mounts and the command line.
The fastest way to see tie is the local test environment in the source repo. It builds the binaries, starts a triplestore and a …
tie has a small vocabulary. Once these six ideas click, the CLI and the apps stop feeling like a database with a filesystem bolted …
The file A .tie file is a flat log of fixed 50-byte little-endian records: Type Size Payload TYPE_ENTRY (11) 50 a 24-byte trie …
tie can mount two different things, both over FUSE: tie mount <dir-hash> ~/snapshot # one immutable tiedir, read-only tie …
tie is a good personal media library because it does not need a media server. The files stay where they are (or in the blob …
tie has free-text search over named relations. There is no sidecar index and no external service: the corpus is built in memory …
tie -f <file.tie> runs the whole client against a local .tie file, with no triplestore daemon. The filehost is still used …
tie-smb serves the live tree over SMB2/3 to Windows Explorer, macOS Finder and Linux CIFS clients. It is pure Go: no Samba, no …
tie verify is fsck for a collection. It cross-checks the forward and reverse indexes, then scans the virtual tree for problems: …
From source git clone https://github.com/uidbz/tie cd tie make build # binaries in dist/ make install # client only, rootless, to …
Both HTTP services use Basic auth with three roles: none < read < write Users live in the service config: [[Users]] Username …
A tie deployment has two data sets: the triplestore — one .tie file per collection (plus _prev history); the filehost — the …
tie-runner is optional. It turns content-addressed trees into a compute substrate: a job takes an input hash, runs in a container, …
Generated from tie --help at v0.9.0. The binary is the source of truth; this page is a convenience. tie [global options] [command …
tie is pre-1.0. The full, canonical changelog is CHANGELOG.md in the source repo. The highlights below are the ones that changed …