feat(main): finish data struct process

open_cortex/runtime/client.py

@@ -17,7 +17,7 @@ def stream_chat_events(message: list[ChatMessage]) -> Iterator[RuntimeEvent]:
     request_body = {
         "messages": to_llama_messages(message),
         "temperature": 0.2,
-        "max_tokens": 100,
+        "max_tokens": 1024,
         "stream": True,
         "stream_options": {"include_usage": True},
         "timings_per_token": True,

open_cortex/ui/gradio_history.py

@@ -0,0 +1,46 @@
+from open_cortex.runtime.messages import ChatMessage
+
+
+def normalize_gradio_content(content) -> str | None:
+    if isinstance(content, str):
+        return content
+
+    if isinstance(content, list):
+        parts = []
+
+        for item in content:
+            if isinstance(item, dict) and isinstance(item.get("text"), str):
+                parts.append(item["text"])
+            elif isinstance(item, str):
+                parts.append(item)
+
+        return "\n".join(parts)
+
+    if isinstance(content, dict) and isinstance(content.get("text"), str):
+        return content["text"]
+
+    return None
+
+
+def history_to_chat_messages(history: list) -> list[ChatMessage]:
+    messages = []
+
+    for item in history:
+        if isinstance(item, dict):
+            role = item.get("role")
+            content = normalize_gradio_content(item.get("content"))
+
+            if role in {"system", "user", "assistant"} and content and content.strip():
+                messages.append(ChatMessage(role=role, content=content.strip()))
+
+        elif isinstance(item, (list, tuple)) and len(item) == 2:
+            user_content = normalize_gradio_content(item[0])
+            assistant_content = normalize_gradio_content(item[1])
+
+            if user_content and user_content.strip():
+                messages.append(ChatMessage(role="user", content=user_content.strip()))
+
+            if assistant_content and assistant_content.strip():
+                messages.append(ChatMessage(role="assistant", content=assistant_content.strip()))
+
+    return messages

pyproject.toml

@@ -9,10 +9,15 @@ dependencies = [
     "httpx>=0.28.1",
 ]
 
-[[tool.uv.index]]
-url = "https://pypi.tuna.tsinghua.edu.cn/simple"
-default = true
-
 [build-system]
 requires = ["hatchling"]
 build-backend = "hatchling.build"
+
+[dependency-groups]
+dev = [
+    "pytest>=9.1.0",
+]
+
+[[tool.uv.index]]
+url = "https://pypi.tuna.tsinghua.edu.cn/simple"
+default = true

scripts/gradio_chat.py

@@ -0,0 +1,127 @@
+import gradio as gr
+from open_cortex.ui.gradio_history import history_to_chat_messages
+from open_cortex.runtime.client import stream_chat_events
+
+
+def format_runtime(event) -> str:
+    if event.kind == "request_started":
+        return "Phase: request started\nWaiting for first token..."
+
+    if event.kind == "first_token" and event.snapshot is not None:
+        snapshot = event.snapshot
+        context_tokens = (
+            snapshot.slot_context_tokens[0]
+            if snapshot.slot_context_tokens
+            else None
+        )
+
+        return "\n".join(
+            [
+                "Phase: first token",
+                f"TTFT: {event.ttft_ms:.1f} ms",
+                f"Context: {context_tokens} / {snapshot.slot_context_size}",
+                f"Token Stream: {snapshot.decode_tps} tok/s",
+                (
+                    "Engine: "
+                    f"processing={snapshot.requests_processing} "
+                    f"deferred={snapshot.requests_deferred}"
+                ),
+            ]
+        )
+
+    if event.kind == "request_completed":
+        return "\n".join(
+            [
+                "Phase: completed",
+                f"Prompt tokens: {event.prompt_tokens}",
+                f"Output tokens: {event.completion_tokens}",
+                f"Prefill: {event.prompt_tps:.1f} tok/s",
+                f"Decode: {event.decode_tps:.1f} tok/s",
+            ]
+        )
+
+    return "Phase: decoding"
+
+def user(user_message: str, history: list[dict]) -> tuple[str, list[dict]]:
+    if not user_message.strip():
+        return "", history
+
+    return "", history + [
+        {
+            "role": "user",
+            "content": user_message,
+        }
+    ]
+
+def bot(history: list):
+    messages = history_to_chat_messages(history)
+
+    history.append(
+        {
+            "role": "assistant",
+            "content": "",
+        }
+    )
+
+    history.append(
+        {
+            "role": "assistant",
+            "content": "",
+        }
+    )
+
+    runtime_text = "Phase: idle"
+
+    for event in stream_chat_events(messages):
+        runtime_text = format_runtime(event)
+
+        if event.text_delta:
+            history[-1]["content"] += event.text_delta
+
+        yield history, runtime_text
+
+
+with gr.Blocks() as demo:
+    gr.Markdown("# OpenCortex Minimal Chat")
+
+    with gr.Row():
+        with gr.Column(scale=2):
+            chatbot = gr.Chatbot(
+                height=480,
+            )
+            msg = gr.Textbox(
+                placeholder="Ask the local model...",
+                show_label=False,
+            )
+            clear = gr.Button("Clear")
+
+        with gr.Column(scale=1):
+            runtime = gr.Textbox(
+                label="Runtime",
+                value="Phase: idle",
+                lines=10,
+                interactive=False,
+            )
+
+
+    msg.submit(
+        user,
+        [msg, chatbot],
+        [msg, chatbot],
+        queue=False,
+    ).then(
+        bot,
+        chatbot,
+        [chatbot,runtime],
+    )
+
+    clear.click(
+        lambda: ([],"Phase: idle"),
+        None,
+        [chatbot, runtime],
+        queue=False,
+    )
+
+if __name__ == "__main__":
+    demo.queue()
+    demo.launch()

tests/runtime/test_metrics.py

@@ -0,0 +1,21 @@
+from open_cortex.runtime.metrics import parse_prometheus_value
+
+
+def test_parse_prometheus_value_reads_single_metric():
+    text = """
+# HELP llamacpp:requests_processing Number of requests processing.
+# TYPE llamacpp:requests_processing gauge
+llamacpp:requests_processing 1
+"""
+
+    value = parse_prometheus_value(text, "llamacpp:requests_processing")
+
+    assert value == 1.0
+
+
+def test_parse_prometheus_value_returns_none_when_missing():
+    text = "llamacpp:requests_processing 1\n"
+
+    value = parse_prometheus_value(text, "llamacpp:requests_deferred")
+
+    assert value is None

tests/ui/test_gradio_history.py

@@ -0,0 +1,55 @@
+from open_cortex.ui.gradio_history import history_to_chat_messages
+
+
+def test_history_to_chat_messages_extracts_text_from_gradio_list_content():
+    history = [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "text": "请解释 KV Cache",
+                    "type": "text",
+                }
+            ],
+        }
+    ]
+
+    messages = history_to_chat_messages(history)
+
+    assert len(messages) == 1
+    assert messages[0].role == "user"
+    assert messages[0].content == "请解释 KV Cache"
+
+
+def test_history_to_chat_messages_ignores_non_text_content():
+    history = [
+        {
+            "role": "user",
+            "content": [
+                {
+                    "path": "/tmp/image.png",
+                    "type": "image",
+                }
+            ],
+        }
+    ]
+
+    messages = history_to_chat_messages(history)
+
+    assert messages == []
+
+
+def test_history_to_chat_messages_supports_legacy_pair_format():
+    history = [
+        [
+            "hello",
+            "hi",
+        ]
+    ]
+
+    messages = history_to_chat_messages(history)
+
+    assert [(m.role, m.content) for m in messages] == [
+        ("user", "hello"),
+        ("assistant", "hi"),
+    ]

uv.lock

@@ -378,6 +378,15 @@ wheels = [
     { url = "https://pypi.tuna.tsinghua.edu.cn/packages/1e/5e/d4e9f1a599fb8e573b7b87160658329fbf28d19eac2718f51fc3def3aa5a/idna-3.18-py3-none-any.whl", hash = "sha256:7f952cbe720b688055e3f87de14f5c3e5fdaa8bc3928985c4077ca689de849a2", size = 65455, upload-time = "2026-06-02T14:34:06.319Z" },
 ]
 
+[[package]]
+name = "iniconfig"
+version = "2.3.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/72/34/14ca021ce8e5dfedc35312d08ba8bf51fdd999c576889fc2c24cb97f4f10/iniconfig-2.3.0.tar.gz", hash = "sha256:c76315c77db068650d49c5b56314774a7804df16fee4402c1f19d6d15d8c4730", size = 20503, upload-time = "2025-10-18T21:55:43.219Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/cb/b1/3846dd7f199d53cb17f49cba7e651e9ce294d8497c8c150530ed11865bb8/iniconfig-2.3.0-py3-none-any.whl", hash = "sha256:f631c04d2c48c52b84d0d0549c99ff3859c98df65b3101406327ecc7d53fbf12", size = 7484, upload-time = "2025-10-18T21:55:41.639Z" },
+]
+
 [[package]]
 name = "jinja2"
 version = "3.1.6"
@@ -544,12 +553,20 @@ dependencies = [
     { name = "httpx" },
 ]
 
+[package.dev-dependencies]
+dev = [
+    { name = "pytest" },
+]
+
 [package.metadata]
 requires-dist = [
     { name = "gradio", specifier = ">=6.18.0" },
     { name = "httpx", specifier = ">=0.28.1" },
 ]
 
+[package.metadata.requires-dev]
+dev = [{ name = "pytest", specifier = ">=9.1.0" }]
+
 [[package]]
 name = "orjson"
 version = "3.11.9"
@@ -733,6 +750,15 @@ wheels = [
     { url = "https://pypi.tuna.tsinghua.edu.cn/packages/ff/6e/cf826fae916b8658848d7b9f38d88da6396895c676e8086fc0988073aaf8/pillow-12.2.0-cp314-cp314t-win_arm64.whl", hash = "sha256:aa88ccfe4e32d362816319ed727a004423aab09c5cea43c01a4b435643fa34eb", size = 2556579, upload-time = "2026-04-01T14:45:52.529Z" },
 ]
 
+[[package]]
+name = "pluggy"
+version = "1.6.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f9/e2/3e91f31a7d2b083fe6ef3fa267035b518369d9511ffab804f839851d2779/pluggy-1.6.0.tar.gz", hash = "sha256:7dcc130b76258d33b90f61b658791dede3486c3e6bfb003ee5c9bfb396dd22f3", size = 69412, upload-time = "2025-05-15T12:30:07.975Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/54/20/4d324d65cc6d9205fabedc306948156824eb9f0ee1633355a8f7ec5c66bf/pluggy-1.6.0-py3-none-any.whl", hash = "sha256:e920276dd6813095e9377c0bc5566d94c932c33b27a3e3945d8389c374dd4746", size = 20538, upload-time = "2025-05-15T12:30:06.134Z" },
+]
+
 [[package]]
 name = "pydantic"
 version = "2.13.4"
@@ -841,6 +867,22 @@ wheels = [
     { url = "https://pypi.tuna.tsinghua.edu.cn/packages/f4/7e/a72dd26f3b0f4f2bf1dd8923c85f7ceb43172af56d63c7383eb62b332364/pygments-2.20.0-py3-none-any.whl", hash = "sha256:81a9e26dd42fd28a23a2d169d86d7ac03b46e2f8b59ed4698fb4785f946d0176", size = 1231151, upload-time = "2026-03-29T13:29:30.038Z" },
 ]
 
+[[package]]
+name = "pytest"
+version = "9.1.0"
+source = { registry = "https://pypi.tuna.tsinghua.edu.cn/simple" }
+dependencies = [
+    { name = "colorama", marker = "sys_platform == 'win32'" },
+    { name = "iniconfig" },
+    { name = "packaging" },
+    { name = "pluggy" },
+    { name = "pygments" },
+]
+sdist = { url = "https://pypi.tuna.tsinghua.edu.cn/packages/84/0e/b5858858d74958632c49b72cb25a3976ff9f632397626715be71c89d3971/pytest-9.1.0.tar.gz", hash = "sha256:41dd9148c08072446394cefd3d79701701335a9f4cae69ba92e39f6c7f5c061c", size = 1634181, upload-time = "2026-06-13T18:52:45.983Z" }
+wheels = [
+    { url = "https://pypi.tuna.tsinghua.edu.cn/packages/8b/5a/ba30a81239b909821b3153e303e7def45178bf353da4f72380e6c5e8793b/pytest-9.1.0-py3-none-any.whl", hash = "sha256:8ebb0e7888bdf2bdfc602ec51f8f62d50200af37356c74e503c79a94f5c81f32", size = 386453, upload-time = "2026-06-13T18:52:44.045Z" },
+]
+
 [[package]]
 name = "python-dateutil"
 version = "2.9.0.post0"