snap: add server app

scripts/server.sh

@@ -0,0 +1,7 @@
+#!/usr/bin/env bash
+set -euo pipefail
+
+# Add additional command line arguments as configured
+SERVER_OPTS=()
+read -r -a SERVER_OPTS <<< "$(snapctl get server-opts)";
+exec "${SNAP}"/bin/cli-wrapper server "${SERVER_OPTS[@]}"

snap/hooks/configure

@@ -19,3 +19,8 @@ esac
 if [ -z "$backend" ]; then
   snapctl set backend=auto
 fi
+
+# Set the default value for server-opts
+if [ -z "$(snapctl get server-opts)" ]; then
+  snapctl set server-opts=""
+fi

snap/snapcraft.yaml

@@ -21,6 +21,21 @@ description: |
   or override it per invocation with the env var:
 
     LLAMA_CPP_BACKEND=cpu llama-cpp cli ...
+
+  The llama.cpp web server can also be run using the llama-cpp.server app. This
+  is disabled by default but can be enabled by running:
+
+    snap start --enable llama-cpp.server
+
+  By default, no additional arguments are passed to the server, meaning it will
+  listen on the default port and address. Arguments can be specified by setting
+  the server-opts configuration value, e.g.:
+
+    snap set llama-cpp server-opts="--host 0.0.0.0"
+
+  Afterwards, the server must be restarted for the new arguments to take effect:
+
+    snap restart llama-cpp.server
 license: MIT
 source-code:
   - https://github.com/talhaHavadar/snap-llama.cpp
@@ -88,10 +103,10 @@ components:
 apps:
   llama-cpp:
     command: bin/cli-wrapper
-    extensions:
+    extensions: &extensions
       # Automatically binds host Vulkan/OpenGL maps
       - gpu
-    plugs:
+    plugs: &plugs
       - home
       - network
       - network-bind
@@ -106,15 +121,23 @@ apps:
       # llama.cpp pins worker threads with sched_setaffinity. Without this
       # application fails to init.
       - process-control
+  server:
+    command: bin/server-wrapper
+    daemon: simple
+    install-mode: disable
+    extensions: *extensions
+    plugs: *plugs
 parts:
   # --- Wrapper script ------------------------------------------------------
   cli:
     plugin: dump
-    source: .
+    source: scripts
     stage:
       - bin/cli-wrapper
+      - bin/server-wrapper
     organize:
       cli.sh: bin/cli-wrapper
+      server.sh: bin/server-wrapper
   # --- ggml: CPU + loader libs, lives in the main snap --------------------
   # Builds llama.cpp's bundled ggml/ subdirectory standalone with the
   # backend-dl loader enabled and every CPU variant emitted as its own .so.