improve service names (#54)

- embedding-server => model_server
- public-types => public_types
- chatbot-ui => chatbot_ui
- function-calling => function_calling

envoyfilter/src/consts.rs

@@ -7,3 +7,4 @@ pub const USER_ROLE: &str = "user";
 pub const GPT_35_TURBO: &str = "gpt-3.5-turbo";
 pub const BOLT_FC_CLUSTER: &str = "bolt_fc_1b";
 pub const BOLT_FC_REQUEST_TIMEOUT_MS: u64 = 120000; // 2 minutes
+pub const MODEL_SERVER_NAME: &str = "model_server";

envoyfilter/src/filter_context.rs

@@ -1,4 +1,4 @@
-use crate::consts::DEFAULT_EMBEDDING_MODEL;
+use crate::consts::{DEFAULT_EMBEDDING_MODEL, MODEL_SERVER_NAME};
 use crate::ratelimit;
 use crate::stats::{Counter, Gauge, RecordingMetric};
 use crate::stream_context::StreamContext;
@@ -123,11 +123,11 @@ impl FilterContext {
 
         let json_data = to_string(&embeddings_input).unwrap();
         let token_id = match self.dispatch_http_call(
-            "embeddingserver",
+            MODEL_SERVER_NAME,
             vec![
                 (":method", "POST"),
                 (":path", "/embeddings"),
-                (":authority", "embeddingserver"),
+                (":authority", MODEL_SERVER_NAME),
                 ("content-type", "application/json"),
                 ("x-envoy-upstream-rq-timeout-ms", "60000"),
             ],
@@ -219,7 +219,7 @@ impl RootContext for FilterContext {
                     .insert(pt.name.clone(), pt.clone());
             }
 
-            debug!("set configuration object: {:?}", self.config);
+            debug!("set configuration object");
 
             if let Some(ratelimits_config) = self
                 .config

envoyfilter/src/stream_context.rs

@@ -1,7 +1,7 @@
 use crate::consts::{
     BOLT_FC_CLUSTER, BOLT_FC_REQUEST_TIMEOUT_MS, DEFAULT_EMBEDDING_MODEL, DEFAULT_INTENT_MODEL,
-    DEFAULT_PROMPT_TARGET_THRESHOLD, GPT_35_TURBO, RATELIMIT_SELECTOR_HEADER_KEY, SYSTEM_ROLE,
-    USER_ROLE,
+    DEFAULT_PROMPT_TARGET_THRESHOLD, GPT_35_TURBO, MODEL_SERVER_NAME,
+    RATELIMIT_SELECTOR_HEADER_KEY, SYSTEM_ROLE, USER_ROLE,
 };
 use crate::filter_context::{embeddings_store, WasmMetrics};
 use crate::ratelimit;
@@ -175,11 +175,11 @@ impl StreamContext {
         };
 
         let token_id = match self.dispatch_http_call(
-            "embeddingserver",
+            MODEL_SERVER_NAME,
             vec![
                 (":method", "POST"),
                 (":path", "/zeroshot"),
-                (":authority", "embeddingserver"),
+                (":authority", MODEL_SERVER_NAME),
                 ("content-type", "application/json"),
                 ("x-envoy-max-retries", "3"),
                 ("x-envoy-upstream-rq-timeout-ms", "60000"),
@@ -637,11 +637,11 @@ impl HttpContext for StreamContext {
         };
 
         let token_id = match self.dispatch_http_call(
-            "embeddingserver",
+            MODEL_SERVER_NAME,
             vec![
                 (":method", "POST"),
                 (":path", "/embeddings"),
-                (":authority", "embeddingserver"),
+                (":authority", MODEL_SERVER_NAME),
                 ("content-type", "application/json"),
                 ("x-envoy-max-retries", "3"),
                 ("x-envoy-upstream-rq-timeout-ms", "60000"),