Merge pull request #2028 from janhq/s/chore/sync-dev-to-main

chore: sync dev to main

Author: vansangpfiev

.github/workflows/cortex-cpp-quality-gate.yml

@@ -168,7 +168,7 @@ jobs:
           cp build/cortex build/cortex-beta
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/main.py
+          python e2e-test/runner/main.py
           rm build/cortex-nightly
           rm build/cortex-beta
         env:
@@ -182,7 +182,7 @@ jobs:
           cp build/cortex.exe build/cortex-beta.exe
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/main.py
+          python e2e-test/runner/main.py
           rm build/cortex-nightly.exe
           rm build/cortex-beta.exe
         env:
@@ -196,7 +196,7 @@ jobs:
           cp build/cortex build/cortex-beta
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/cortex-llamacpp-e2e-nightly.py
+          python e2e-test/runner/cortex-llamacpp-e2e-nightly.py
           rm build/cortex-nightly
           rm build/cortex-beta
         env:
@@ -210,7 +210,7 @@ jobs:
           cp build/cortex.exe build/cortex-beta.exe
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/cortex-llamacpp-e2e-nightly.py
+          python e2e-test/runner/cortex-llamacpp-e2e-nightly.py
           rm build/cortex-nightly.exe
           rm build/cortex-beta.exe
         env:
@@ -443,7 +443,7 @@ jobs:
           cp build/cortex build/cortex-beta
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/main.py
+          python e2e-test/runner/main.py
           rm build/cortex-nightly
           rm build/cortex-beta
         env:
@@ -457,7 +457,7 @@ jobs:
           cp build/cortex.exe build/cortex-beta.exe
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/main.py
+          python e2e-test/runner/main.py
           rm build/cortex-nightly.exe
           rm build/cortex-beta.exe
         env:
@@ -471,7 +471,7 @@ jobs:
           cp build/cortex build/cortex-beta
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/cortex-llamacpp-e2e-nightly.py
+          python e2e-test/runner/cortex-llamacpp-e2e-nightly.py
           rm build/cortex-nightly
           rm build/cortex-beta
         env:
@@ -485,7 +485,7 @@ jobs:
           cp build/cortex.exe build/cortex-beta.exe
           python -m pip install --upgrade pip
           python -m pip install -r e2e-test/requirements.txt
-          python e2e-test/cortex-llamacpp-e2e-nightly.py
+          python e2e-test/runner/cortex-llamacpp-e2e-nightly.py
           rm build/cortex-nightly.exe
           rm build/cortex-beta.exe
         env:

docs/static/openapi/cortex.json

@@ -5356,7 +5356,7 @@
             "type": "string",
             "description": "The identifier or URL of the model to use. It can be a model ID on Cortexso (https://huggingface.co/cortexso) or a HuggingFace URL pointing to the model file. For example: 'gpt2' or 'https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/blob/main/mistral-7b-instruct-v0.1.Q2_K.gguf'",
             "examples": [
-              "tinyllama:gguf",
+              "tinyllama:1b",
               "https://huggingface.co/TheBloke/Mistral-7B-Instruct-v0.1-GGUF/blob/main/mistral-7b-instruct-v0.1.Q2_K.gguf"
             ]
           },

engine/common/download_task_queue.h

@@ -1,3 +1,4 @@
+#include <algorithm>
 #include <condition_variable>
 #include <deque>
 #include <mutex>

engine/controllers/models.cc

@@ -385,6 +385,10 @@ void Models::UpdateModel(const HttpRequestPtr& req,
       message = "Successfully update model ID '" + model_id +
                 "': " + json_body.toStyledString();
     } else if (model_config.engine == kPythonEngine) {
+      // Block changes to `command`
+      if (json_body.isMember("command")) {
+        json_body.removeMember("command");
+      }
       config::PythonModelConfig python_model_config;
       python_model_config.ReadFromYaml(yaml_fp.string());
       python_model_config.FromJson(json_body);
@@ -859,10 +863,11 @@ void Models::GetModelSource(
 void Models::GetRepositoryList(
     const HttpRequestPtr& req,
     std::function<void(const HttpResponsePtr&)>&& callback,
-    std::optional<std::string> author) {
+    std::optional<std::string> author, std::optional<std::string> tag) {
   if (!author.has_value())
     author = "cortexso";
-  auto res = model_src_svc_->GetRepositoryList(author.value());
+  auto res =
+      model_src_svc_->GetRepositoryList(author.value(), tag.value_or(""));
   if (res.has_error()) {
     Json::Value ret;
     ret["message"] = res.error();

engine/controllers/models.h

@@ -44,8 +44,8 @@ class Models : public drogon::HttpController<Models, false> {
   ADD_METHOD_TO(Models::DeleteModelSource, "/v1/models/sources", Delete);
   ADD_METHOD_TO(Models::GetModelSources, "/v1/models/sources", Get);
   ADD_METHOD_TO(Models::GetModelSource, "/v1/models/sources/{src}", Get);
-  ADD_METHOD_TO(Models::GetRepositoryList, "/v1/models/hub?author={author}",
-                Get);
+  ADD_METHOD_TO(Models::GetRepositoryList,
+                "/v1/models/hub?author={author}&tag={tag}", Get);
   METHOD_LIST_END
 
   explicit Models(std::shared_ptr<DatabaseService> db_service,
@@ -115,7 +115,8 @@ class Models : public drogon::HttpController<Models, false> {
 
   void GetRepositoryList(const HttpRequestPtr& req,
                          std::function<void(const HttpResponsePtr&)>&& callback,
-                         std::optional<std::string> author);
+                         std::optional<std::string> author,
+                         std::optional<std::string> tag);
 
  private:
   std::shared_ptr<DatabaseService> db_service_;

engine/e2e-test/test_api_engine.py engine/e2e-test/api/engines/test_api_engine.py

@@ -1,7 +1,7 @@
 import pytest
 import requests
 import time
-from test_runner import (
+from utils.test_runner import (
     start_server,
     stop_server,
     wait_for_websocket_download_success_event,

engine/e2e-test/test_api_engine_install_nightly.py engine/e2e-test/api/engines/test_api_engine_install_nightly.py

@@ -1,6 +1,6 @@
 import pytest
 import requests
-from test_runner import start_server, stop_server, get_latest_pre_release_tag
+from utils.test_runner import start_server, stop_server, get_latest_pre_release_tag
 
 latest_pre_release_tag = get_latest_pre_release_tag("janhq", "cortex.llamacpp")
 

engine/e2e-test/test_api_engine_update.py engine/e2e-test/api/engines/test_api_engine_update.py

@@ -1,6 +1,6 @@
 import pytest
 import requests
-from test_runner import (
+from utils.test_runner import (
     start_server,
     stop_server,
     wait_for_websocket_download_success_event,

engine/e2e-test/api/engines/test_api_get_default_engine.py

@@ -0,0 +1,82 @@
+import pytest
+import requests
+from utils.test_runner import start_server, stop_server
+import jsonschema
+from tenacity import retry, wait_exponential, stop_after_attempt
+from utils.logger import log_response
+from utils.assertion import assert_equal
+
+
+class TestApiDefaultEngine:
+
+    @pytest.fixture(autouse=True)
+    def setup_and_teardown(self):
+        # Setup
+        success = start_server()
+        if not success:
+            raise Exception("Failed to start server")
+
+        yield
+
+        # Teardown
+        stop_server()
+
+    def test_api_get_default_engine_successfully(self):
+        # Data test
+        engine= "llama-cpp"
+        name= "linux-amd64-avx-cuda-11-7"
+        version= "v0.1.35-27.10.24"
+    
+        data = {"version": version, "variant": name}
+        post_install_url = f"http://localhost:3928/v1/engines/{engine}/install"
+        response = requests.post(
+            post_install_url, json=data
+        )
+        assert_equal(response.status_code,200)
+        log_response(response.json(), "test_api_get_default_engine_successfully")
+        
+        get_list_url = f"http://localhost:3928/v1/engines/{engine}"
+        get_default_url = f"http://localhost:3928/v1/engines/{engine}/default"
+
+        @retry(
+            wait=wait_exponential(multiplier=2, min=2, max=30), 
+            stop=stop_after_attempt(5) 
+        )
+        def get_request(url):
+            response = requests.get(url)
+            assert len(response.json()) > 0
+
+        get_request(get_list_url)
+        
+        response_default_engine = requests.get(get_default_url)
+        json_data = response_default_engine.json()
+
+        log_response(json_data, "test_api_get_default_engine_successfully")
+        assert_equal(response_default_engine.status_code, 200)
+
+        schema = {
+            "type": "object",
+            "properties": {
+                "engine": {"type": "string"},
+                "variant": {"type": "string"},
+                "version": {"type": "string"}
+            },
+            "required": ["engine", "variant", "version"]
+        }
+
+        # Validate response schema
+        jsonschema.validate(instance=json_data, schema=schema)
+        
+    def test_api_get_default_engine_failed_invalid_engine(self):
+        # Data test
+        engine= "invalid"
+    
+        get_default_url = f"http://localhost:3928/v1/engines/{engine}/default"
+
+        response_default_engine = requests.get(get_default_url)
+        json_data_get_default = response_default_engine.json()
+
+        log_response(json_data_get_default, "test_api_get_default_engine_failed_invalid_engine")
+        assert_equal(response_default_engine.status_code, 400)
+
+        assert_equal(json_data_get_default["message"], f"Engine {engine} is not supported yet!")

engine/e2e-test/api/engines/test_api_get_engine_release.py

@@ -0,0 +1,76 @@
+import pytest
+import requests
+from utils.test_runner import start_server, stop_server
+import jsonschema
+from tenacity import retry, wait_exponential, stop_after_attempt
+from utils.logger import log_response
+from utils.assertion import assert_equal, assert_contains
+
+
+class TestApiEngineRelease:
+
+    @pytest.fixture(autouse=True)
+    def setup_and_teardown(self):
+        # Setup
+        success = start_server()
+        if not success:
+            raise Exception("Failed to start server")
+
+        yield
+
+        # Teardown
+        stop_server()
+
+    def test_api_get_engine_release_successfully(self):
+        # Data test
+        engine= "llama-cpp"
+        get_release_url = f"http://localhost:3928/v1/engines/{engine}/releases"
+
+        @retry(
+            wait=wait_exponential(multiplier=2, min=2, max=30), 
+            stop=stop_after_attempt(5) 
+        )
+        def get_request(url):
+            response = requests.get(url)
+            assert len(response.json()) > 0
+
+        get_request(get_release_url)
+        
+        response_engine_release = requests.get(get_release_url)
+        json_data = response_engine_release.json()
+
+        log_response(json_data, "test_api_get_engine_release_successfully")
+        assert_equal(response_engine_release.status_code, 200)
+
+        schema = {
+            "$schema": "http://json-schema.org/draft-07/schema#",
+            "type": "array",
+            "items": {
+                "type": "object",
+                "properties": {
+                "draft": { "type": "boolean" },
+                "name": { "type": "string" },
+                "prerelease": { "type": "boolean" },
+                "published_at": { "type": "string", "format": "date-time" },
+                "url": { "type": "string", "format": "uri" }
+                },
+                "required": ["draft", "name", "prerelease", "published_at", "url"]
+            }
+        }
+
+        # Validate response schema
+        jsonschema.validate(instance=json_data, schema=schema)
+        
+    def test_api_ge_engine_release_failed_invalid_engine(self):
+        # Data test
+        engine= "invalid"
+    
+        get_default_url = f"http://localhost:3928/v1/engines/{engine}/releases"
+
+        response_default_engine = requests.get(get_default_url)
+        json_data_get_default = response_default_engine.json()
+
+        log_response(json_data_get_default, "test_api_ge_engine_release_failed_invalid_engine")
+        assert_equal(response_default_engine.status_code, 400)
+
+        assert_contains(json_data_get_default["message"], "Not Found")

engine/e2e-test/api/engines/test_api_get_engine_release_latest.py

@@ -0,0 +1,73 @@
+import pytest
+import requests
+from utils.test_runner import start_server, stop_server
+import jsonschema
+from tenacity import retry, wait_exponential, stop_after_attempt
+from utils.logger import log_response
+from utils.assertion import assert_equal, assert_contains
+
+
+class TestApiEngineReleaseLatest:
+
+    @pytest.fixture(autouse=True)
+    def setup_and_teardown(self):
+        # Setup
+        success = start_server()
+        if not success:
+            raise Exception("Failed to start server")
+
+        yield
+
+        # Teardown
+        stop_server()
+
+    def test_api_get_engine_release_latest_successfully(self):
+        # Data test
+        engine= "llama-cpp"
+        get_release_url = f"http://localhost:3928/v1/engines/{engine}/releases/latest"
+
+        @retry(
+            wait=wait_exponential(multiplier=2, min=2, max=30), 
+            stop=stop_after_attempt(5) 
+        )
+        def get_request(url):
+            response = requests.get(url)
+            assert len(response.json()) > 0
+
+        get_request(get_release_url)
+        
+        response_engine_release = requests.get(get_release_url)
+        json_data = response_engine_release.json()
+
+        log_response(json_data, "test_api_get_engine_release_latest_successfully")
+        assert_equal(response_engine_release.status_code, 200)
+
+        schema = {
+            "$schema": "https://json-schema.org/draft/2020-12/schema",
+            "type": "array",
+            "items": {
+                "type": "object",
+                "properties": {
+                "created_at": {
+                    "type": "string",
+                    "format": "date-time"
+                },
+                "download_count": {
+                    "type": "integer",
+                    "minimum": 0
+                },
+                "name": {
+                    "type": "string"
+                },
+                "size": {
+                    "type": "integer",
+                    "minimum": 0
+                }
+                },
+                "required": ["created_at", "download_count", "name", "size"]
+            }
+        }
+
+
+        # Validate response schema
+        jsonschema.validate(instance=json_data, schema=schema)