Erich Blume
1315b9ce18
A surface-owned, auto-spawned daemon can't be shared once the CLI is also a first-class client — so drop auto-spawn and manage the daemon as an OS service. - design §4: daemon lifecycle = explicit OS service; surfaces connect-only - heph-nvim.md: rewrite the daemon-lifecycle section (connect-only) + history - new how-to/run-the-daemon.md (heph daemon start/stop/restart/status); indexed - install-heph.md: post-install is `heph daemon start`; plugin no longer spawns - tech-spec §14: mark the managed-daemon entry superseded - changelog fragment Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
2.7 KiB
| title | modified | tags | |
|---|---|---|---|
| Install heph (and isolate dev) | 2026-06-02 |
|
Install heph (and isolate dev)
How to install heph/hephd from the forge and run the heph.nvim plugin,
isolated from in-repo development so the two never share data. No prebuilt
binaries yet — everything builds from source (works on macOS/arm64,
linux/arm64, linux/amd64).
1. Install the binaries from the forge
Build and install heph + hephd to ~/.cargo/bin (on PATH) from a forge
ref. Until v1 is tagged, install from the branch:
cargo install --locked \
--git ssh://forgejo@forge.ops.eblu.me:2222/eblume/hephaestus.git \
--branch feature/v1-prototype \
heph hephd
Re-run with --tag vX.Y.Z once a release is cut. This needs forge SSH access
(an unlocked 1Password / ssh-agent key).
The installed daemon owns the default paths — socket
$XDG_RUNTIME_DIR/heph/hephd.sock, DB $XDG_DATA_HOME/heph/heph.db — i.e. your
real data.
2. The Neovim plugin
heph.nvim lives in a subdirectory of the monorepo, and lazy.nvim can't load a
subdir plugin from a bare git URL (it puts the clone root on runtimepath).
For now, point lazy at a dedicated checkout via dir:
git clone --branch feature/v1-prototype \
ssh://forgejo@forge.ops.eblu.me:2222/eblume/hephaestus.git \
~/.local/share/heph/checkout
-- lazy.nvim spec
{
dir = vim.fn.expand("~/.local/share/heph/checkout/heph.nvim"),
config = function()
require("heph").setup({}) -- connect-only: talks to the daemon you started
end,
}
The plugin is connect-only — it talks to a hephd you run as a service, it
does not start one itself. Start the daemon once:
heph daemon start # launchd agent (macOS) / systemd user service (Linux)
See run-the-daemon for start/stop/restart/status. Update the plugin
by git pulling the checkout (and after a cargo install upgrade, heph daemon restart to pick up the new hephd). (A future split of heph.nvim into its own
forge repo will make this a normal { "eblume/heph.nvim" } spec.)
3. Isolate development
In-repo development must not touch the installed store. Run the dev daemon on separate paths and point a dev Neovim at it:
mise run dev # runs the working-tree hephd on .dev/hephd.sock + .dev/heph.db
# dev Neovim — the plugin reads these envs and targets the dev daemon
HEPH_SOCKET="$PWD/.dev/hephd.sock" HEPH_DB="$PWD/.dev/heph.db" nvim
The installed plugin (no envs) talks to the installed daemon; the dev Neovim
talks to the dev daemon. They never share a socket or DB. .dev/ is gitignored.