Fix lint errors

Signed-off-by: Ira <[email protected]>

Author: irar2

pkg/llm-d-inference-sim/metrics_test.go

@@ -322,11 +322,10 @@ var _ = Describe("Simulator metrics", Ordered, func() {
 			Expect(err).NotTo(HaveOccurred())
 		})
 		It("Should send correct kv cache usage metrics", func() {
-			modelName := "Qwen/Qwen2-0.5B"
 			// Three requests, there are should be two blocks in the kv cache, because
 			// the first and the second prompt share a block.
 			ctx := context.TODO()
-			args := []string{"cmd", "--model", modelName, "--mode", common.ModeRandom,
+			args := []string{"cmd", "--model", qwenModelName, "--mode", common.ModeRandom,
 				"--enable-kvcache", "true", "--kv-cache-size", "16", "--block-size", "8",
 				"--time-to-first-token", "5000", "--tokenizers-cache-dir", tmpDir}
 
@@ -342,19 +341,19 @@ var _ = Describe("Simulator metrics", Ordered, func() {
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in Haifa today? Is it cold?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 				{
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in Haifa today?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 				{
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in New York today?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 			}
 
@@ -402,9 +401,8 @@ var _ = Describe("Simulator metrics", Ordered, func() {
 		})
 
 		It("Should send correct kv cache usage metrics for sequentual requests", func() {
-			modelName := "Qwen/Qwen2-0.5B"
 			ctx := context.TODO()
-			args := []string{"cmd", "--model", modelName, "--mode", common.ModeRandom,
+			args := []string{"cmd", "--model", qwenModelName, "--mode", common.ModeRandom,
 				"--enable-kvcache", "true", "--kv-cache-size", "16", "--block-size", "8",
 				"--time-to-first-token", "5000", "--tokenizers-cache-dir", tmpDir, "--max-num-seqs", "2"}
 
@@ -420,19 +418,19 @@ var _ = Describe("Simulator metrics", Ordered, func() {
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in Haifa today? Is it cold?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 				{
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in Haifa today?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 				{
 					Prompt: openai.CompletionNewParamsPromptUnion{
 						OfString: openai.String("What is the weather like in New York today?"),
 					},
-					Model: openai.CompletionNewParamsModel(modelName),
+					Model: openai.CompletionNewParamsModel(qwenModelName),
 				},
 			}
 

pkg/llm-d-inference-sim/simulator_test.go

@@ -42,6 +42,7 @@ import (
 )
 
 const model = "my_model"
+const qwenModelName = "Qwen/Qwen2-0.5B"
 const baseURL = "http://localhost/v1"
 const userMessage = "This is a test."
 const invalidMaxTokensErrMsg = "Max completion tokens and max tokens should be positive"
@@ -1081,10 +1082,10 @@ var _ = Describe("Simulator", func() {
 
 	It("Should return correct response to /tokenize chat", func() {
 		ctx := context.TODO()
-		modelName := "Qwen/Qwen2-0.5B"
-		tmpDir := "./tests-tmp/"
+		tmpDir := "./tmp/"
+		//nolint
 		defer os.RemoveAll(tmpDir)
-		args := []string{"cmd", "--model", modelName, "--mode", common.ModeRandom,
+		args := []string{"cmd", "--model", qwenModelName, "--mode", common.ModeRandom,
 			"--tokenizers-cache-dir", tmpDir}
 		client, err := startServerWithArgs(ctx, common.ModeRandom, args, nil)
 		Expect(err).NotTo(HaveOccurred())
@@ -1113,10 +1114,10 @@ var _ = Describe("Simulator", func() {
 
 	It("Should return correct response to /tokenize text", func() {
 		ctx := context.TODO()
-		modelName := "Qwen/Qwen2-0.5B"
-		tmpDir := "./tests-tmp/"
+		tmpDir := "./tmp/"
+		//nolint
 		defer os.RemoveAll(tmpDir)
-		args := []string{"cmd", "--model", modelName, "--mode", common.ModeRandom,
+		args := []string{"cmd", "--model", qwenModelName, "--mode", common.ModeRandom,
 			"--tokenizers-cache-dir", tmpDir}
 		client, err := startServerWithArgs(ctx, common.ModeRandom, args, nil)
 		Expect(err).NotTo(HaveOccurred())

pkg/vllm-api/tokenize.go

@@ -41,7 +41,7 @@ func (t *TokenizeRequest) GetPrompt() string {
 		return t.Prompt
 	}
 
-	var messages []string
+	messages := make([]string, 0)
 	for _, message := range t.Messages {
 		messages = append(messages, message.Content.PlainText())
 	}