fix: e2e tests

sangjanai · sangjanai · commit e3dfea3202a6 · 2025-03-20T06:02:28.000+07:00
diff --git a/docs/static/openapi/cortex.json b/docs/static/openapi/cortex.json
@@ -2754,7 +2754,7 @@
                       },
                       "version": {
                         "type": "string",
-                        "example": "0.1.35-28.10.24"
+                        "example": "b4920"
                       }
                     }
                   }
@@ -2763,11 +2763,11 @@
                   {
                     "engine": "llama-cpp",
                     "name": "mac-arm64",
-                    "version": "0.1.35-28.10.24"
+                    "version": "b4920"
                   },
                   {
                     "engine": "llama-cpp",
-                    "name": "linux-amd64-avx",
+                    "name": "linux-avx-x64",
                     "version": "0.1.35-27.10.24"
                   }
                 ]
@@ -2901,7 +2901,7 @@
                       "name": {
                         "type": "string",
                         "description": "The name of the variant, including OS, architecture, and capabilities",
-                        "example": "linux-amd64-avx-cuda-11-7"
+                        "example": "linux-avx-x64-cuda-11-7"
                       },
                       "created_at": {
                         "type": "string",
@@ -2973,7 +2973,7 @@
                       },
                       "name": {
                         "type": "string",
-                        "example": "0.1.39-linux-amd64-avx-cuda-11-7"
+                        "example": "llama-b4920-bin-linux-avx-cuda-cu11.7"
                       },
                       "size": {
                         "type": "integer",
@@ -3250,7 +3250,7 @@
                     },
                     "version": {
                       "type": "string",
-                      "example": "0.1.35-28.10.24"
+                      "example": "b4920"
                     }
                   }
                 }
diff --git a/engine/e2e-test/api/engines/test_api_engine.py b/engine/e2e-test/api/engines/test_api_engine.py
@@ -28,14 +28,14 @@ def test_engines_get_llamacpp_should_be_successful(self):
         
     # engines install
     def test_engines_install_llamacpp_specific_version_and_variant(self):
-        data = {"version": "v0.1.40-b4354", "variant": "linux-amd64-avx"}
+        data = {"version": "b4920", "variant": "linux-avx-x64"}
         response = requests.post(
             "http://localhost:3928/v1/engines/llama-cpp/install", json=data
         )
         assert response.status_code == 200
 
     def test_engines_install_llamacpp_specific_version_and_null_variant(self):
-        data = {"version": "v0.1.40-b4354"}
+        data = {"version": "b4920"}
         response = requests.post(
             "http://localhost:3928/v1/engines/llama-cpp/install", json=data
         )
@@ -55,14 +55,14 @@ async def test_engines_install_uninstall_llamacpp_should_be_successful(self):
     @pytest.mark.asyncio
     async def test_engines_install_uninstall_llamacpp_with_only_version_should_be_failed(self):
         # install first
-        data = {"variant": "mac-arm64"}
+        data = {"variant": "linux-avx-x64"}
         install_response = requests.post(
             "http://127.0.0.1:3928/v1/engines/llama-cpp/install", json=data
         )
         await wait_for_websocket_download_success_event(timeout=120)
         assert install_response.status_code == 200
 
-        data = {"version": "v0.1.35"}
+        data = {"version": "b4920"}
         response = requests.delete(
             "http://localhost:3928/v1/engines/llama-cpp/install", json=data
         )
@@ -72,7 +72,7 @@ async def test_engines_install_uninstall_llamacpp_with_only_version_should_be_fa
     @pytest.mark.asyncio
     async def test_engines_install_uninstall_llamacpp_with_variant_should_be_successful(self):
         # install first
-        data = {"variant": "mac-arm64"}
+        data = {"variant": "linux-avx-x64"}
         install_response = requests.post(
             "http://127.0.0.1:3928/v1/engines/llama-cpp/install", json=data
         )
@@ -85,7 +85,7 @@ async def test_engines_install_uninstall_llamacpp_with_variant_should_be_success
     def test_engines_install_uninstall_llamacpp_with_specific_variant_and_version_should_be_successful(
         self,
     ):
-        data = {"variant": "mac-arm64", "version": "v0.1.35"}
+        data = {"variant": "linux-avx-x64", "version": "b4920"}
         # install first
         install_response = requests.post(
             "http://localhost:3928/v1/engines/llama-cpp/install", json=data
diff --git a/engine/e2e-test/api/engines/test_api_engine_install_nightly.py b/engine/e2e-test/api/engines/test_api_engine_install_nightly.py
@@ -2,7 +2,7 @@
 import requests
 from utils.test_runner import start_server, stop_server, get_latest_pre_release_tag
 
-latest_pre_release_tag = get_latest_pre_release_tag("menloresearch", "cortex.llamacpp")
+latest_pre_release_tag = get_latest_pre_release_tag("menloresearch", "llama.cpp")
 
 class TestApiEngineInstall:
 
@@ -23,7 +23,7 @@ def test_engines_install_llamacpp_should_be_successful(self):
         assert response.status_code == 200
 
     def test_engines_install_llamacpp_specific_version_and_variant(self):
-        data = {"version": latest_pre_release_tag, "variant": "linux-amd64-avx"}
+        data = {"version": latest_pre_release_tag, "variant": "linux-avx-x64"}
         response = requests.post(
             "http://localhost:3928/v1/engines/llama-cpp/install", json=data
         )
diff --git a/engine/e2e-test/api/engines/test_api_get_default_engine.py b/engine/e2e-test/api/engines/test_api_get_default_engine.py
@@ -24,8 +24,8 @@ def setup_and_teardown(self):
     def test_api_get_default_engine_successfully(self):
         # Data test
         engine= "llama-cpp"
-        name= "linux-amd64-avx"
-        version= "v0.1.35-27.10.24"
+        name= "linux-avx-x64"
+        version= "b4920"
     
         data = {"version": version, "variant": name}
         post_install_url = f"http://localhost:3928/v1/engines/{engine}/install"
diff --git a/engine/e2e-test/api/engines/test_api_get_list_engine.py b/engine/e2e-test/api/engines/test_api_get_list_engine.py
@@ -24,8 +24,8 @@ def setup_and_teardown(self):
     def test_api_get_list_engines_successfully(self):
         # Data test
         engine= "llama-cpp"
-        name= "linux-amd64-avx"
-        version= "v0.1.35-27.10.24"
+        name= "linux-avx-x64"
+        version= "b4920"
         
         post_install_url = f"http://localhost:3928/v1/engines/{engine}/install"
         response = requests.delete(
diff --git a/engine/e2e-test/api/engines/test_api_post_default_engine.py b/engine/e2e-test/api/engines/test_api_post_default_engine.py
@@ -23,8 +23,8 @@ def setup_and_teardown(self):
     def test_api_set_default_engine_successfully(self):
         # Data test
         engine= "llama-cpp"
-        name= "linux-amd64-avx"
-        version= "v0.1.35-27.10.24"
+        name= "linux-avx-x64"
+        version= "b4920"
     
         data = {"version": version, "variant": name}
         post_install_url = f"http://localhost:3928/v1/engines/{engine}/install"
diff --git a/engine/e2e-test/api/hardware/test_api_get_hardware.py b/engine/e2e-test/api/hardware/test_api_get_hardware.py
@@ -88,25 +88,6 @@ def test_api_get_hardware_successfully(self):
                     "example": True,
                     "description": "Indicates if the GPU is currently activated."
                     },
-                    "additional_information": {
-                    "type": "object",
-                    "properties": {
-                        "compute_cap": {
-                        "type": "string",
-                        "example": "8.6",
-                        "description": "The compute capability of the GPU."
-                        },
-                        "driver_version": {
-                        "type": "string",
-                        "example": "535.183",
-                        "description": "The version of the installed driver."
-                        }
-                    },
-                    "required": [
-                        "compute_cap",
-                        "driver_version"
-                    ]
-                    },
                     "free_vram": {
                     "type": "integer",
                     "example": 23983,
@@ -140,7 +121,6 @@ def test_api_get_hardware_successfully(self):
                 },
                 "required": [
                     "activated",
-                    "additional_information",
                     "free_vram",
                     "id",
                     "name",
diff --git a/engine/e2e-test/api/model/test_api_model.py b/engine/e2e-test/api/model/test_api_model.py
@@ -95,6 +95,7 @@ async def test_models_start_stop_should_be_successful(self):
         time.sleep(30)
 
         print("Pull model")
+        requests.delete("http://localhost:3928/v1/models/tinyllama:1b")
         json_body = {"model": "tinyllama:1b"}
         response = requests.post("http://localhost:3928/v1/models/pull", json=json_body)
         assert response.status_code == 200, f"Failed to pull model: tinyllama:1b"
diff --git a/engine/e2e-test/cli/engines/test_cli_engine_install.py b/engine/e2e-test/cli/engines/test_cli_engine_install.py
@@ -31,25 +31,9 @@ def test_engines_install_llamacpp_should_be_successfully(self):
         assert len(response.json()) > 0
         assert exit_code == 0, f"Install engine failed with error: {error}"
 
-    @pytest.mark.skipif(reason="Ignore onnx-runtime test")
-    def test_engines_install_onnx_on_macos_should_be_failed(self):
-        exit_code, output, error = run(
-            "Install Engine", ["engines", "install", "onnxruntime"]
-        )
-        assert "is not supported on" in output, "Should display error message"
-        assert exit_code == 0, f"Install engine failed with error: {error}"
-
-    @pytest.mark.skipif(reason="Ignore tensorrt-llm test")
-    def test_engines_install_onnx_on_tensorrt_should_be_failed(self):
-        exit_code, output, error = run(
-            "Install Engine", ["engines", "install", "tensorrt-llm"]
-        )
-        assert "is not supported on" in output, "Should display error message"
-        assert exit_code == 0, f"Install engine failed with error: {error}"
-
     @pytest.mark.skipif(platform.system() == "Windows", reason="Progress bar log issue on Windows")
     def test_engines_install_pre_release_llamacpp(self):
-        engine_version = "v0.1.43"
+        engine_version = "b4920"
         exit_code, output, error = run(
             "Install Engine",
             ["engines", "install", "llama-cpp", "-v", engine_version],
diff --git a/engine/e2e-test/runner/cortex-llamacpp-e2e-nightly.py b/engine/e2e-test/runner/cortex-llamacpp-e2e-nightly.py
@@ -21,7 +21,7 @@
 from api.engines.test_api_get_default_engine import TestApiDefaultEngine
 from api.engines.test_api_get_engine_release import TestApiEngineRelease
 from api.engines.test_api_get_engine_release_latest import TestApiEngineReleaseLatest
-from test_api_post_default_engine import TestApiSetDefaultEngine
+from api.engines.test_api_post_default_engine import TestApiSetDefaultEngine
 from api.model.test_api_model import TestApiModel
 from api.model.test_api_model_import import TestApiModelImport
 from api.files.test_api_create_file import TestApiCreateFile
diff --git a/engine/e2e-test/runner/main.py b/engine/e2e-test/runner/main.py
@@ -21,7 +21,7 @@
 from api.engines.test_api_get_default_engine import TestApiDefaultEngine
 from api.engines.test_api_get_engine_release import TestApiEngineRelease
 from api.engines.test_api_get_engine_release_latest import TestApiEngineReleaseLatest
-from test_api_post_default_engine import TestApiSetDefaultEngine
+from api.engines.test_api_post_default_engine import TestApiSetDefaultEngine
 from api.model.test_api_model import TestApiModel
 from api.model.test_api_model_import import TestApiModelImport
 from api.files.test_api_create_file import TestApiCreateFile
diff --git a/engine/e2e-test/test_api_cortexso_hub_llamacpp_engine.py b/engine/e2e-test/test_api_cortexso_hub_llamacpp_engine.py
@@ -125,7 +125,7 @@ async def test_models_on_cortexso_hub(self, model_url):
             "Install Engine", ["engines", "install", "llama-cpp"], timeout=None, capture = False
         )
         root = Path.home()
-        assert os.path.exists(root / "cortexcpp" / "engines" / "cortex.llamacpp" / "version.txt")
+        assert os.path.exists(root / "cortexcpp" / "engines" / "llama.cpp" / "version.txt")
         assert exit_code == 0, f"Install engine failed with error: {error}"
 
         # Start the model
diff --git a/engine/test/components/test_engine_matcher_utils.cc b/engine/test/components/test_engine_matcher_utils.cc
@@ -117,4 +117,11 @@ TEST_F(EngineMatcherUtilsTestSuite, TestGetVersionAndArch) {
     EXPECT_EQ(version, "b4920");
     EXPECT_EQ(arch, "macos-x64");
   }
+
+  {
+    std::string variant = "llama-b4920-bin-ubuntu-vulkan-x64.zip";
+    auto [version, arch] = engine_matcher_utils::GetVersionAndArch(variant);
+    EXPECT_EQ(version, "b4920");
+    EXPECT_EQ(arch, "ubuntu-vulkan-x64");
+  }
 }
diff --git a/engine/test/components/test_github_release_utils.cc b/engine/test/components/test_github_release_utils.cc
@@ -4,16 +4,16 @@
 class GitHubReleaseUtilsTest : public ::testing::Test {};
 
 TEST_F(GitHubReleaseUtilsTest, AbleToGetReleaseByVersion) {
-  auto version{"v0.1.36"};
+  auto version{"b4920"};
   auto result = github_release_utils::GetReleaseByVersion(
-      kMenloOrg, "cortex.llamacpp", version);
+      kMenloOrg, "llama.cpp", version);
 
   ASSERT_TRUE(result.has_value());
   ASSERT_EQ(result->tag_name, version);
 }
 
 TEST_F(GitHubReleaseUtilsTest, AbleToGetReleaseList) {
-  auto result = github_release_utils::GetReleases(kMenloOrg, "cortex.llamacpp");
+  auto result = github_release_utils::GetReleases(kMenloOrg, "llama.cpp");
 
   ASSERT_TRUE(result.has_value());
   ASSERT_TRUE(result->size() > 0);

Original file line number	Diff line number	Diff line change
`@@ -2754,7 +2754,7 @@`
`2754`	`2754`	`},`
`2755`	`2755`	`"version": {`
`2756`	`2756`	`"type": "string",`
`2757`		`- "example": "0.1.35-28.10.24"`
	`2757`	`+ "example": "b4920"`
`2758`	`2758`	`}`
`2759`	`2759`	`}`
`2760`	`2760`	`}`
`@@ -2763,11 +2763,11 @@`
`2763`	`2763`	`{`
`2764`	`2764`	`"engine": "llama-cpp",`
`2765`	`2765`	`"name": "mac-arm64",`
`2766`		`- "version": "0.1.35-28.10.24"`
	`2766`	`+ "version": "b4920"`
`2767`	`2767`	`},`
`2768`	`2768`	`{`
`2769`	`2769`	`"engine": "llama-cpp",`
`2770`		`- "name": "linux-amd64-avx",`
	`2770`	`+ "name": "linux-avx-x64",`
`2771`	`2771`	`"version": "0.1.35-27.10.24"`
`2772`	`2772`	`}`
`2773`	`2773`	`]`
`@@ -2901,7 +2901,7 @@`
`2901`	`2901`	`"name": {`
`2902`	`2902`	`"type": "string",`
`2903`	`2903`	`"description": "The name of the variant, including OS, architecture, and capabilities",`
`2904`		`- "example": "linux-amd64-avx-cuda-11-7"`
	`2904`	`+ "example": "linux-avx-x64-cuda-11-7"`
`2905`	`2905`	`},`
`2906`	`2906`	`"created_at": {`
`2907`	`2907`	`"type": "string",`
`@@ -2973,7 +2973,7 @@`
`2973`	`2973`	`},`
`2974`	`2974`	`"name": {`
`2975`	`2975`	`"type": "string",`
`2976`		`- "example": "0.1.39-linux-amd64-avx-cuda-11-7"`
	`2976`	`+ "example": "llama-b4920-bin-linux-avx-cuda-cu11.7"`
`2977`	`2977`	`},`
`2978`	`2978`	`"size": {`
`2979`	`2979`	`"type": "integer",`
`@@ -3250,7 +3250,7 @@`
`3250`	`3250`	`},`
`3251`	`3251`	`"version": {`
`3252`	`3252`	`"type": "string",`
`3253`		`- "example": "0.1.35-28.10.24"`
	`3253`	`+ "example": "b4920"`
`3254`	`3254`	`}`
`3255`	`3255`	`}`
`3256`	`3256`	`}`