MetaGPT icon indicating copy to clipboard operation
MetaGPT copied to clipboard
verified publisher icon geekan

JSONDecodeError

Open jtara1 opened this issue 1 year ago • 6 comments

Bug description

Project Manager transforms Architect's WriteDesign, but not as JSON then it tries to parse it as JSON

run command

metagpt.exe 'create flappy bird as a web app using only vanilla js, html, and css'

Environment information

  • LLM type and model name: ollama llama3
  • System version: Windows 10
  • Python version: 3.10.6
  • MetaGPT version or branch: main branch probably, likely commit 38cea1daf2b87ebc31a56c995d7857d07289fa70
  • packages version: pip 22.3.1
  • installation method: pip install --upgrade git+https://github.com/geekan/MetaGPT.git

Screenshots or logs

WriteDesign from Architect

{
    "Implementation approach": "We will implement Flappy Bird as a web app using vanilla JavaScript, HTML, and CSS. We will analyze the difficult points of the requirements, select the appropriate open-source framework for responsive design, and use keyboard or touch even
ts to control the bird's movement.",
    "File list": [
        "index.html",
        "style.css",
        "script.js"
    ],
    "Data structures and interfaces": "
classDiagram
    class FlappyBirdGame {
        -birdMovementAlgorithm
        +startGame()
        +updateGame()
        +renderGame()
    }
    class BirdMovementAlgorithm {
        -pipeMovementAlgorithm
        +calculateBirdPosition()
        +checkCollision()
    }
    class PipeMovementAlgorithm {
        +generatePipes()
        +movePipes()
    }
    FlappyBirdGame --> BirdMovementAlgorithm
    BirdMovementAlgorithm --> PipeMovementAlgorithm
",
    "Program call flow": "
sequenceDiagram
    participant FBG as FlappyBirdGame
    participant BMA as BirdMovementAlgorithm
    participant PMA as PipeMovementAlgorithm
    FBG->>BMA: startGame()
    BMA->>PMA: generatePipes()
    PMA-->>BMA: return pipes
    BMA->>FBG: updateGame()
    FBG->>BMA: renderGame()
",
    "Anything UNCLEAR": "Clarification needed on scoring system and game over logic."
}

WriteTasks from Project Manager

{
    "Implementation approach": "We will implement Flappy Bird as a web app using vanilla JavaScript, HTML, and CSS. We will analyze the difficult points of the requirements, select the appropriate open-ource framework for responsive design, and use keyboard or touch event
s to control the bird's movement.",
    "File list": ["index.html", "style.css", "script.js"],
    "Data structures and interfaces":
        classDiagram
        class FlappyBirdGame {
            -birdMovementAlgorithm
            +startGame()
            +updateGame()
            +renderGame()
        }
        class BirdMovementAlgorithm {
            -pipeMovementAlgorithm
            +calculateBirdPosition()
            +checkCollision()
        }
        class PipeMovementAlgorithm {
            +generatePipes()
            +movePipes()
        }
        FlappyBirdGame --> BirdMovementAlgorithm
        BirdMovementAlgorithm --> PipeMovementAlgorithm,
    "Program call flow":
        sequenceDiagram
        participant FBG as FlappyBirdGame
        participant BMA as BirdMovementAlgorithm
        participant PMA as PipeMovementAlgorithm
        FBG->>BMA: startGame()
        BMA->>PMA: generatePipes()
        PMA-->>BMA: return pipes
        BMA->>FBG: updateGame()
        FBG->>BMA: renderGame(),
    "Anything UNCLEAR": "Clarification needed on scoring system and game over logic."
}

actual error from stderr

json.decoder.JSONDecodeError: Expecting value: line 5 column 9 (char 415)

jtara1 avatar Jun 13 '24 00:06 jtara1

Ollama is not as effective as OpenAI. Try to set repair_llm_output: true in config2.yaml to get a better output.

voidking avatar Jun 17 '24 13:06 voidking

Still getting the same error with that config option. ~/.metagpt/config2.yaml

llm:
  api_type: 'ollama'
  base_url: 'http://127.0.0.1:11434/api'
  model: 'llama3'
  repair_llm_output: true

jtara1 avatar Jun 18 '24 02:06 jtara1

I'm getting a similar error with Ollama: json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0) running metagpt from the commandline. I have an issue: https://github.com/geekan/MetaGPT/issues/1383. I added repair_llm_output: true to ~/.metagpt/config2.yaml, but I still get the same error.

$ metagpt "Create a 2048 game"
2024-07-07 14:37:19.995 | INFO     | metagpt.const:get_metagpt_package_root:29 - Package root set to /Users/ken/devel/ai/agents/metagpt_novel
2024-07-07 14:37:27.413 | INFO     | metagpt.team:invest:90 - Investment: $3.0.
2024-07-07 14:37:27.414 | INFO     | metagpt.roles.role:_act:391 - Alice(Product Manager): to do PrepareDocuments(PrepareDocuments)
2024-07-07 14:37:27.535 | INFO     | metagpt.utils.file_repository:save:57 - save to: /Users/ken/devel/ai/agents/metagpt_novel/workspace/20240707143727/docs/requirement.txt
2024-07-07 14:37:27.538 | INFO     | metagpt.roles.role:_act:391 - Alice(Product Manager): to do WritePRD(WritePRD)
2024-07-07 14:37:27.540 | INFO     | metagpt.actions.write_prd:run:86 - New requirement detected: Create a 2048 game
Sure! Here's an example of how you could fill in the nodes for a 2048 game project:

### Language
Python

### Original Requirements
Create a 2048 game where players can move tiles around to combine them and reach the goal of having a 2048 tile.

### Project Name
Game_2048

### Product Goals

1. Create an engaging user experience
2. Improve accessibility, be responsive
3. More beautiful UI

### User Stories

1. As a player, I want to be able to choose difficulty levels so that the game is challenging but not too hard or too easy.
2. As a player, I want to see my score after each game so that I can track my progress and compete with others.
3. As a player, I want to get a restart button when I lose so that I can try again without feeling frustrated.
4. As a player, I want to see beautiful UI that makes me feel good and motivates me to play more.
5. As a player, I want to play the game via mobile phone so that I can play it anywhere, anytime.

### Competitive Analysis

1. 2048 Game A: Simple interface, lacks responsive features
2. Play2048.co: Beautiful and responsive UI with my best score shown
3. 2048game.com: Responsive UI with my best score shown, but many ads

### Competitive Quadrant Chart

    Title: Reach and engagement of campaigns
    X-axis: Low Reach --> High Reach
    Y-axis: Low Engagement --> High Engagement
    Quadrant 1: We should expand
    Quadrant 2: Need to promote
    Quadrant 3: Re-evaluate
    Quadrant 4: May be improved
    Campaign A: [0.3, 0.6]
    Campaign B: [0.45, 0.23]
    Campaign C: [0.57, 0.69]
    Campaign D: [0.78, 0.34]
    Campaign E: [0.40, 0.34]
    Campaign F: [0.35, 0.78]
    Our Target Product: [0.5, 0.6]

### Requirement Analysis

Not applicable for this project.

### Requirement Pool

1. P0: The main code of the game should be easy to understand and modify.
2. P0: The game algorithm should be efficient and fast.
3. P1: The UI should be visually appealing and easy to use.
4. P2: The game should be responsive and work well on different devices.

### UI Design draft
Basic function description with a simple style and layout.

### Anything UNCLEAR
Nothing is unclear for this project.
2024-07-07 14:40:39.306 | WARNING  | metagpt.utils.cost_manager:update_cost:49 - Model llama2:latest not found in TOKEN_COSTS.
2024-07-07 14:40:39.317 | WARNING  | metagpt.utils.repair_llm_raw_output:extract_content_from_output:320 - extract_content try another pattern: \[CONTENT\]([\s\S]*)\[/CONTENT\]
2024-07-07 14:40:39.318 | WARNING  | metagpt.utils.repair_llm_raw_output:run_and_passon:268 - parse json from content inside [CONTENT][/CONTENT] failed at retry 1, exp: Expecting value: line 1 column 1 (char 0)
2024-07-07 14:40:39.318 | INFO     | metagpt.utils.repair_llm_raw_output:repair_invalid_json:237 - repair_invalid_json, raw error: Expecting value: line 1 column 1 (char 0)
2024-07-07 14:40:39.319 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 191.776(s), this was the 1st time calling it. exp: RetryError[<Future at 0x130ca46d0 state=finished raised JSONDecodeError>]

kjenney avatar Jul 07 '24 18:07 kjenney

Ollama is definitely being used. The CPU usage is at 700% and my laptop is humming.

kjenney avatar Jul 07 '24 18:07 kjenney

The problem is at

    def _decode_and_load(self, chunk: bytes, encoding: str = "utf-8") -> dict:
        chunk = chunk.decode(encoding)
        return json.loads(chunk)

The error info is

chunk
'data: {"id":"chatcmpl-471","object":"chat.completion.chunk","created":1720838591,"model":"llama3","system_fingerprint":"fp_ollama","choices":[{"index":0,"delta":{"role":"assistant","content":"I"},"finish_reason":null}]}\n'
json.loads(chunk)
Traceback (most recent call last):
  File "<string>", line 1, in <module>
  File "/opt/homebrew/Cellar/[email protected]/3.9.19/Frameworks/Python.framework/Versions/3.9/lib/python3.9/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/opt/homebrew/Cellar/[email protected]/3.9.19/Frameworks/Python.framework/Versions/3.9/lib/python3.9/json/decoder.py", line 337, in decode
    obj, end = self.raw_decode(s, idx=_w(s, 0).end())
  File "/opt/homebrew/Cellar/[email protected]/3.9.19/Frameworks/Python.framework/Versions/3.9/lib/python3.9/json/decoder.py", line 355, in raw_decode
    raise JSONDecodeError("Expecting value", s, err.value) from None
json.decoder.JSONDecodeError: Expecting value: line 1 column 1 (char 0)

ZhengQiushi avatar Jul 13 '24 02:07 ZhengQiushi

maybe you can modify as follows:

    def _decode_and_load(self, chunk: bytes, encoding: str = "utf-8") -> dict:
        chunk = chunk.decode(encoding)
        json_data = chunk.removeprefix('data: ').strip()
        # print("json_data: ", json_data, "/n")
        if len(json_data) == 0: 
            return {}
        elif json_data.lower().find("done") != -1:
            return {"done": True}
        else:
            ret = json.loads(json_data)
            delta = ret.get('choices', [{}])[0].get('delta', {})
            ret["message"] = delta
        return ret

        async for raw_chunk in stream_resp:
            chunk = self._decode_and_load(raw_chunk)
            if chunk == {}:
                continue
            if not chunk.get("done", False):
                ...

ZhengQiushi avatar Jul 13 '24 06:07 ZhengQiushi

Due to the lack of updates or replies by the user for a long time, we will close it. Please reopen it if necessary.

better629 avatar Oct 09 '24 17:10 better629

This is still an issue. How many people use ollama or a local LLM?

jtara1 avatar Oct 10 '24 02:10 jtara1

This is still an issue. How many people use ollama or a local LLM?

This is still an issue in 2025/02/24? @better629 @jtara1 @kjenney @voidking @ZhengQiushi

  1. config config2.yaml 
llm:
  api_type: "ollama"
  model: "deepseek-r1:14b"  # or gpt-3.5-turbo-1106 / gpt-4-1106-previewgit
  base_url: "http://192.168.93.186:11434"  # or forward url / other llm url
repair_llm_output: true

mermaid:
  engine: "nodejs"
  path: "mmdc"
  puppeteer_config: "/app/metagpt/config/puppeteer-config.json"
  pyppeteer_path: "/usr/bin/chromium"
  1. launch docker metagpt
$ docker run --network host --privileged -v /opt/metagpt/config/config2.yaml:/app/metagpt/config/config2.yaml     -v /opt/metagpt/workspace:/app/metagpt/workspace     metagpt/metagpt:latest     metagpt "Write a cli snake game"
2025-02-24 11:01:12.126 | INFO     | metagpt.const:get_metagpt_package_root:21 - Package root set to /app/metagpt
2025-02-24 11:01:16.099 | INFO     | metagpt.team:invest:93 - Investment: $3.0.
2025-02-24 11:01:16.101 | INFO     | metagpt.roles.role:_act:403 - Alice(Product Manager): to do PrepareDocuments(PrepareDocuments)
2025-02-24 11:01:16.175 | INFO     | metagpt.utils.file_repository:save:57 - save to: /app/metagpt/workspace/20250224110116/docs/requirement.txt
2025-02-24 11:01:16.176 | INFO     | metagpt.roles.role:_act:403 - Alice(Product Manager): to do WritePRD(WritePRD)
2025-02-24 11:01:16.177 | INFO     | metagpt.actions.write_prd:run:86 - New requirement detected: Write a cli snake game
2025-02-24 11:01:16.196 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 0.018(s), this was the 1st time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:16.569 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 0.391(s), this was the 2nd time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:18.088 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 1.910(s), this was the 3rd time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:19.692 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 3.514(s), this was the 4th time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:20.731 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 4.553(s), this was the 5th time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:26.350 | ERROR    | metagpt.utils.common:log_it:554 - Finished call to 'metagpt.actions.action_node.ActionNode._aask_v1' after 10.172(s), this was the 6th time calling it. exp: Extra data: line 1 column 5 (char 4)
2025-02-24 11:01:26.351 | WARNING  | metagpt.utils.common:wrapper:673 - There is a exception in role's execution, in order to resume, we delete the newest role communication message in the role's memory.
2025-02-24 11:01:26.359 | ERROR    | metagpt.utils.common:wrapper:655 - Exception occurs, start to serialize the project, exp:
Traceback (most recent call last):
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 50, in __call__
    result = await fn(*args, **kwargs)
  File "/app/metagpt/metagpt/actions/action_node.py", line 437, in _aask_v1
    content = await self.llm.aask(prompt, system_msgs, images=images, timeout=timeout)
json.decoder.JSONDecodeError: Extra data: line 1 column 5 (char 4)

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/app/metagpt/metagpt/utils/common.py", line 664, in wrapper
    return await func(self, *args, **kwargs)
  File "/app/metagpt/metagpt/roles/role.py", line 551, in run
    rsp = await self.react()
tenacity.RetryError: RetryError[<Future at 0x7f3a2aa98dc0 state=finished raised JSONDecodeError>]

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File "/app/metagpt/metagpt/utils/common.py", line 650, in wrapper
    result = await func(self, *args, **kwargs)
  File "/app/metagpt/metagpt/team.py", line 134, in run
    await self.env.run()
Exception: Traceback (most recent call last):
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 50, in __call__
    result = await fn(*args, **kwargs)
  File "/app/metagpt/metagpt/actions/action_node.py", line 437, in _aask_v1
    content = await self.llm.aask(prompt, system_msgs, images=images, timeout=timeout)
  File "/app/metagpt/metagpt/provider/base_llm.py", line 152, in aask
    rsp = await self.acompletion_text(message, stream=stream, timeout=self.get_timeout(timeout))
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 88, in async_wrapped
    return await fn(*args, **kwargs)
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 47, in __call__
    do = self.iter(retry_state=retry_state)
  File "/usr/local/lib/python3.9/site-packages/tenacity/__init__.py", line 314, in iter
    return fut.result()
  File "/usr/local/lib/python3.9/concurrent/futures/_base.py", line 439, in result
    return self.__get_result()
  File "/usr/local/lib/python3.9/concurrent/futures/_base.py", line 391, in __get_result
    raise self._exception
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 50, in __call__
    result = await fn(*args, **kwargs)
  File "/app/metagpt/metagpt/provider/base_llm.py", line 202, in acompletion_text
    return await self._achat_completion_stream(messages, timeout=self.get_timeout(timeout))
  File "/app/metagpt/metagpt/provider/ollama_api.py", line 252, in _achat_completion_stream
    return self._processing_openai_response(resp)
  File "/app/metagpt/metagpt/provider/ollama_api.py", line 257, in _processing_openai_response
    resp = self.ollama_message.decode(openai_resp)
  File "/app/metagpt/metagpt/provider/ollama_api.py", line 41, in decode
    return json.loads(response.data.decode("utf-8"))
  File "/usr/local/lib/python3.9/json/__init__.py", line 346, in loads
    return _default_decoder.decode(s)
  File "/usr/local/lib/python3.9/json/decoder.py", line 340, in decode
    raise JSONDecodeError("Extra data", s, end)
json.decoder.JSONDecodeError: Extra data: line 1 column 5 (char 4)

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
  File "/app/metagpt/metagpt/utils/common.py", line 664, in wrapper
    return await func(self, *args, **kwargs)
  File "/app/metagpt/metagpt/roles/role.py", line 551, in run
    rsp = await self.react()
  File "/app/metagpt/metagpt/roles/role.py", line 520, in react
    rsp = await self._react()
  File "/app/metagpt/metagpt/roles/role.py", line 475, in _react
    rsp = await self._act()
  File "/app/metagpt/metagpt/roles/role.py", line 404, in _act
    response = await self.rc.todo.run(self.rc.history)
  File "/app/metagpt/metagpt/actions/write_prd.py", line 87, in run
    return await self._handle_new_requirement(req)
  File "/app/metagpt/metagpt/actions/write_prd.py", line 108, in _handle_new_requirement
    node = await WRITE_PRD_NODE.fill(context=context, llm=self.llm, exclude=exclude)  # schema=schema
  File "/app/metagpt/metagpt/actions/action_node.py", line 648, in fill
    return await self.simple_fill(schema=schema, mode=mode, images=images, timeout=timeout, exclude=exclude)
  File "/app/metagpt/metagpt/actions/action_node.py", line 473, in simple_fill
    content, scontent = await self._aask_v1(
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 88, in async_wrapped
    return await fn(*args, **kwargs)
  File "/usr/local/lib/python3.9/site-packages/tenacity/_asyncio.py", line 47, in __call__
    do = self.iter(retry_state=retry_state)
  File "/usr/local/lib/python3.9/site-packages/tenacity/__init__.py", line 326, in iter
    raise retry_exc from fut.exception()
tenacity.RetryError: RetryError[<Future at 0x7f3a2aa98dc0 state=finished raised JSONDecodeError>]

zjykzj avatar Feb 24 '25 11:02 zjykzj

This is no longer an issue for me using the latest version of Ollama and the latest version of MetaGPT.

kjenney avatar Feb 24 '25 14:02 kjenney

This is no longer an issue for me using the latest version of Ollama and the latest version of MetaGPT.

Yes! I am using the latest versions of Olama and Metagpt.

$ ollama --version
ollama version is 0.5.12
$ ollama run deepseek-r1:14b
>>> 1+1=
<think>

</think>

The sum of 1 and 1 is:

\[
1 + 1 = \boxed{2}
\]

>>> Send a message (/? for help)

$ proxychains docker pull metagpt/metagpt:latest
ProxyChains-3.1 (http://proxychains.sf.net)
latest: Pulling from metagpt/metagpt
Digest: sha256:65e94d386a8aba2a7d9cb99e171d0897f9fccc2f3b1e4c61307fc1a2807b098b
Status: Image is up to date for metagpt/metagpt:latest
docker.io/metagpt/metagpt:latest
$ cat config2.yaml
llm:
  api_type: "ollama"
  model: "deepseek-r1:14b"  # or gpt-3.5-turbo-1106 / gpt-4-1106-previewgit
  base_url: "http://0.0.0.0:11434"  # or forward url / other llm url
repair_llm_output: true

mermaid:
  engine: "nodejs"
  path: "mmdc"
  puppeteer_config: "/app/metagpt/config/puppeteer-config.json"
  pyppeteer_path: "/usr/bin/chromium"
$ docker run --network host --privileged -v /opt/metagpt/config/config2.yaml:/app/metagpt/config/config2.yaml     -v /opt/metagpt/workspace:/app/metagpt/workspace     metagpt/metagpt:latest     metagpt "Write a cli snake game"

zjykzj avatar Feb 25 '25 02:02 zjykzj

The ollama endpoint is http://127.0.0.1:11434/api

kjenney avatar Feb 25 '25 02:02 kjenney

The ollama endpoint is http://127.0.0.1:11434/api

@kjenney Thank you very much! It works

$ cat config2.yaml
llm:
  api_type: "ollama"
  model: "deepseek-r1:14b"  # or gpt-3.5-turbo-1106 / gpt-4-1106-previewgit
  base_url: "http://127.0.0.1:11434/api"  # or forward url / other llm url
repair_llm_output: true

mermaid:
  engine: "nodejs"
  path: "mmdc"
  puppeteer_config: "/app/metagpt/config/puppeteer-config.json"
  pyppeteer_path: "/usr/bin/chromium"

I am using deepseek-r1:14b for inference

$ docker run --network host --privileged -v /opt/metagpt/config/config2.yaml:/app/metagpt/config/config2.yaml     -v /opt/metagpt/workspace:/app/metagpt/workspace     metagpt/metagpt:latest     metagpt "Write a command-line program to input any number and print how many small stars"
...
...
...
$ python3 main.py
Enter a number between 1 and 100: 10
**********
$ python3 main.py
Enter a number between 1 and 100: 33
*********************************

zjykzj avatar Feb 25 '25 03:02 zjykzj