server: logs: allow to choose log format in json or plain text

examples/server/README.md

@@ -39,9 +39,11 @@ see https://github.com/ggerganov/llama.cpp/issues/1437
 - `--mmproj MMPROJ_FILE`: Path to a multimodal projector file for LLaVA.
 - `--grp-attn-n`: Set the group attention factor to extend context size through self-extend(default: 1=disabled), used together with group attention width `--grp-attn-w`
 - `--grp-attn-w`: Set the group attention width to extend context size through self-extend(default: 512), used together with group attention factor `--grp-attn-n`
-- `-n, --n-predict`: Set the maximum tokens to predict (default: -1)
+- `-n N, --n-predict N`: Set the maximum tokens to predict (default: -1)
 - `--slots-endpoint-disable`: To disable slots state monitoring endpoint. Slots state may contain user data, prompts included.
 - `--chat-template JINJA_TEMPLATE`: Set custom jinja chat template. This parameter accepts a string, not a file name (default: template taken from model's metadata). We only support [some pre-defined templates](https://github.com/ggerganov/llama.cpp/wiki/Templates-supported-by-llama_chat_apply_template)
+- `--log-disable`: Output logs to stdout only, default: enabled.
+- `--log-format FORMAT`: Define the log output to FORMAT: json or text (default: json)
 
 ## Build
 

examples/server/server.cpp

@@ -46,6 +46,7 @@ struct server_params
 };
 
 bool server_verbose = false;
+bool server_log_json = true;
 
 static size_t common_part(const std::vector<llama_token> &a, const std::vector<llama_token> &b)
 {
@@ -2037,6 +2038,7 @@ static void server_print_usage(const char *argv0, const gpt_params &params,
     printf("  -ctv TYPE, --cache-type-v TYPE\n");
     printf("                            KV cache data type for V (default: f16)\n");
     printf("  --mmproj MMPROJ_FILE      path to a multimodal projector file for LLaVA.\n");
+    printf("  --log-format              log output format: json or text (default: json)\n");
     printf("  --log-disable             disables logging to a file.\n");
     printf("  --slots-endpoint-disable  disables slots monitoring endpoint.\n");
     printf("\n");
@@ -2489,6 +2491,27 @@ static void server_params_parse(int argc, char **argv, server_params &sparams,
             }
             params.mmproj = argv[i];
         }
+        else if (arg == "--log-format")
+        {
+            if (++i >= argc)
+            {
+                invalid_param = true;
+                break;
+            }
+            if (std::strcmp(argv[i], "json") == 0)
+            {
+                server_log_json = true;
+            }
+            else if (std::strcmp(argv[i], "text") == 0)
+            {
+                server_log_json = false;
+            }
+            else
+            {
+                invalid_param = true;
+                break;
+            }
+        }
         else if (arg == "--log-disable")
         {
             log_set_target(stdout);

examples/server/utils.hpp

@@ -14,6 +14,7 @@
 using json = nlohmann::json;
 
 extern bool server_verbose;
+extern bool server_log_json;
 
 #ifndef SERVER_VERBOSE
 #define SERVER_VERBOSE 1
@@ -133,36 +134,49 @@ struct completion_token_output
     std::string text_to_send;
 };
 
-static inline void server_log(const char *level, const char *function, int /*line*/, const char *message, const nlohmann::ordered_json &extra)
+static void server_log(const char *level, const char *function, int line, const char *message, const nlohmann::ordered_json &extra)
 {
-    char buf[1024];
+    std::stringstream ss_tid;
-    snprintf(buf, 1024, "%24s %4s: %-80s", function, level, message);
+    ss_tid << std::this_thread::get_id();
-
+    json log = nlohmann::ordered_json{
-    nlohmann::ordered_json log;
+        {"tid", ss_tid.str()},
-
+        {"timestamp", time(nullptr)},
-    {
+    };
-        std::stringstream ss_thread_id;
-        ss_thread_id << std::this_thread::get_id();
-        log.push_back({"tid", ss_thread_id.str()});
-    }
 
+    if (server_log_json) {
         log.merge_patch(
                 {
-        //{"timestamp", time(nullptr)},
+                        {"level",     level},
-        //{"level",     level},
+                        {"function",  function},
-        //{"function",  function},
+                        {"line",      line},
-        //{"line",      line},
+                        {"msg",       message},
-        {"msg",   buf},
                 });
-
+        if (!extra.empty()) {
-    if (!extra.empty())
-    {
             log.merge_patch(extra);
         }
 
         const std::string str = log.dump(-1, ' ', false, json::error_handler_t::replace);
         printf("%.*s\n", (int)str.size(), str.data());
         fflush(stdout);
+    } else {
+        char buf[1024];
+        snprintf(buf, 1024, "%24s %4s: %-80s", function, level, message);
+
+        if (!extra.empty()) {
+            log.merge_patch(extra);
+        }
+        std::stringstream ss;
+        ss << buf;
+        for (const auto& el : log.items())
+        {
+            snprintf(buf, 1024, " %s=%s", el.key().c_str(), el.value().dump().c_str());
+            ss << buf;
+        }
+
+        const std::string str = ss.str();
+        printf("%.*s\n", (int)str.size(), str.data());
+        fflush(stdout);
+    }
 }
 
 //