mirror of
https://github.com/lordmathis/llamactl.git
synced 2025-11-05 16:44:22 +00:00
Improve comments
This commit is contained in:
@@ -12,11 +12,13 @@ import (
|
||||
"time"
|
||||
)
|
||||
|
||||
// errorResponse represents an error response returned by the API
|
||||
type errorResponse struct {
|
||||
Error string `json:"error"`
|
||||
Details string `json:"details,omitempty"`
|
||||
}
|
||||
|
||||
// writeError writes a JSON error response with the specified HTTP status code
|
||||
func writeError(w http.ResponseWriter, status int, code, details string) {
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
w.WriteHeader(status)
|
||||
@@ -25,6 +27,7 @@ func writeError(w http.ResponseWriter, status int, code, details string) {
|
||||
}
|
||||
}
|
||||
|
||||
// writeJSON writes a JSON response with the specified HTTP status code
|
||||
func writeJSON(w http.ResponseWriter, status int, data any) {
|
||||
w.Header().Set("Content-Type", "application/json")
|
||||
w.WriteHeader(status)
|
||||
@@ -33,6 +36,7 @@ func writeJSON(w http.ResponseWriter, status int, data any) {
|
||||
}
|
||||
}
|
||||
|
||||
// writeText writes a plain text response with the specified HTTP status code
|
||||
func writeText(w http.ResponseWriter, status int, data string) {
|
||||
w.Header().Set("Content-Type", "text/plain")
|
||||
w.WriteHeader(status)
|
||||
@@ -41,12 +45,14 @@ func writeText(w http.ResponseWriter, status int, data string) {
|
||||
}
|
||||
}
|
||||
|
||||
// Handler provides HTTP handlers for the llamactl server API
|
||||
type Handler struct {
|
||||
InstanceManager manager.InstanceManager
|
||||
cfg config.AppConfig
|
||||
httpClient *http.Client
|
||||
}
|
||||
|
||||
// NewHandler creates a new Handler instance with the provided instance manager and configuration
|
||||
func NewHandler(im manager.InstanceManager, cfg config.AppConfig) *Handler {
|
||||
return &Handler{
|
||||
InstanceManager: im,
|
||||
@@ -57,6 +63,7 @@ func NewHandler(im manager.InstanceManager, cfg config.AppConfig) *Handler {
|
||||
}
|
||||
}
|
||||
|
||||
// getInstance retrieves an instance by name from the request query parameters
|
||||
func (h *Handler) getInstance(r *http.Request) (*instance.Instance, error) {
|
||||
name := r.URL.Query().Get("name")
|
||||
validatedName, err := validation.ValidateInstanceName(name)
|
||||
@@ -72,6 +79,8 @@ func (h *Handler) getInstance(r *http.Request) (*instance.Instance, error) {
|
||||
return inst, nil
|
||||
}
|
||||
|
||||
// ensureInstanceRunning ensures the instance is running by starting it if on-demand start is enabled
|
||||
// It handles LRU eviction when the maximum number of running instances is reached
|
||||
func (h *Handler) ensureInstanceRunning(inst *instance.Instance) error {
|
||||
options := inst.GetOptions()
|
||||
allowOnDemand := options != nil && options.OnDemandStart != nil && *options.OnDemandStart
|
||||
|
||||
@@ -10,11 +10,12 @@ import (
|
||||
"strings"
|
||||
)
|
||||
|
||||
// ParseCommandRequest represents the request body for command parsing
|
||||
// ParseCommandRequest represents the request body for backend command parsing
|
||||
type ParseCommandRequest struct {
|
||||
Command string `json:"command"`
|
||||
}
|
||||
|
||||
// validateLlamaCppInstance validates that the instance specified in the request is a llama.cpp instance
|
||||
func (h *Handler) validateLlamaCppInstance(r *http.Request) (*instance.Instance, error) {
|
||||
inst, err := h.getInstance(r)
|
||||
if err != nil {
|
||||
@@ -33,6 +34,7 @@ func (h *Handler) validateLlamaCppInstance(r *http.Request) (*instance.Instance,
|
||||
return inst, nil
|
||||
}
|
||||
|
||||
// stripLlamaCppPrefix removes the llama.cpp proxy prefix from the request URL path
|
||||
func (h *Handler) stripLlamaCppPrefix(r *http.Request, instName string) {
|
||||
// Strip the "/llama-cpp/<name>" prefix from the request URL
|
||||
prefix := fmt.Sprintf("/llama-cpp/%s", instName)
|
||||
|
||||
@@ -6,7 +6,7 @@ import (
|
||||
"github.com/go-chi/chi/v5"
|
||||
)
|
||||
|
||||
// NodeResponse represents a sanitized node configuration for API responses
|
||||
// NodeResponse represents a node configuration in API responses
|
||||
type NodeResponse struct {
|
||||
Address string `json:"address"`
|
||||
}
|
||||
|
||||
@@ -8,13 +8,13 @@ import (
|
||||
"net/http"
|
||||
)
|
||||
|
||||
// OpenAIListInstancesResponse represents the response structure for listing instances (models) in OpenAI format
|
||||
// OpenAIListInstancesResponse represents the response structure for listing instances (models) in OpenAI-compatible format
|
||||
type OpenAIListInstancesResponse struct {
|
||||
Object string `json:"object"`
|
||||
Data []OpenAIInstance `json:"data"`
|
||||
}
|
||||
|
||||
// OpenAIInstance represents a single instance (model) in OpenAI format
|
||||
// OpenAIInstance represents a single instance (model) in OpenAI-compatible format
|
||||
type OpenAIInstance struct {
|
||||
ID string `json:"id"`
|
||||
Object string `json:"object"`
|
||||
|
||||
Reference in New Issue
Block a user