This repository was archived by the owner on Dec 28, 2025. It is now read-only.
Refactor: Refactor hugegraph-ai to using CGraph & port some usecase in web demo#49
Merged
weijinglin merged 9 commits intoagenticrag/devfrom Sep 25, 2025
Merged
Refactor: Refactor hugegraph-ai to using CGraph & port some usecase in web demo#49weijinglin merged 9 commits intoagenticrag/devfrom
weijinglin merged 9 commits intoagenticrag/devfrom
Conversation
|
Important Review skippedAuto reviews are disabled on base/target branches other than the default branch. Please check the settings in the CodeRabbit UI or the You can disable this status message by setting the Walkthrough此次变更引入固定流程框架文档与需求说明；新增并重构多条固定流程（构建向量索引、图抽取、导入数据、更新VID向量、获取索引信息、构建模式、提示词生成）；新增统一的Node层与若干节点类；调整Operators由Node解耦为纯操作；扩展Scheduler注册与调度；更新状态结构与若干工具/演示入口。 Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Client as 调用方
participant S as Scheduler
participant M as GPipelineManager
participant F as Flow
participant P as GPipeline
participant N as Node
participant O as Operator
Client->>S: schedule_flow(task, inputs)
S->>M: get_manager(task)
alt 首次/无可复用Pipeline
S->>F: build_flow(inputs)
F->>P: 创建并装配Pipeline
P-->>S: 返回Pipeline
else 复用Pipeline
S->>P: 准备输入并运行
end
S->>P: run()
loop 节点执行
P->>N: run()
N->>O: operator_schedule(state_json)
O-->>N: 处理结果
N-->>P: CStatus
end
P-->>S: wkflow_state
S-->>Client: 返回结果
sequenceDiagram
autonumber
actor UI as Web Demo
participant S as Scheduler
participant F as PromptGenerateFlow
participant P as GPipeline
participant N as PromptGenerateNode
participant O as PromptGenerate(Operator)
participant L as LLM
UI->>S: prompt_generate(source_text, scenario, example_name)
S->>F: build_flow(...)
F->>P: 注册 PromptGenerateNode
S->>P: run()
P->>N: run()
N->>O: operator_schedule(state_json)
O->>L: 调用LLM生成提示词
L-->>O: 文本
O-->>N: {"generated_extract_prompt": "..."}
N-->>P: CStatus
P-->>S: wkflow_state
S-->>UI: 返回prompt
Estimated code review effort🎯 4 (Complex) | ⏱️ ~75 minutes Poem
Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
|
@codecov-ai-reviewer review |
![github-actions[bot]](https://avatars.githubusercontent.com/in/15368?s=60&v=4){kind=small}
There was a problem hiding this comment.
license-eye has checked 353 files.
| Valid | Invalid | Ignored | Fixed |
|---|---|---|---|
| 285 | 1 | 67 | 0 |
Click to see the invalid file list
- hugegraph-llm/test_prompt_generate_workflow.py
Use this command to fix any missing license headers
```bash
docker run -it --rm -v $(pwd):/github/workspace apache/skywalking-eyes header fix
</details>
Collaborator
Author
|
@coderabbitai review |
✅ Actions performedReview triggered.
|
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 16
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (2)
hugegraph-llm/src/hugegraph_llm/state/ai_state.py (1)
103-110: 安全与稳健性：assign_from_json 允许任意键注入，可能遮蔽方法或破坏实例状态当前实现会把任意键设置为实例属性，可能覆盖如 to_json 等方法名，或注入未知字段，带来稳定性与安全隐患。建议仅赋值已存在且非可调用的属性。
应用如下补丁：
- # Implement a method that assigns keys from data_json as WkFlowState member variables - def assign_from_json(self, data_json: dict): - """ - Assigns each key in the input json object as a member variable of WkFlowState. - """ - for k, v in data_json.items(): - setattr(self, k, v) + # Implement a method that assigns keys from data_json as WkFlowState member variables + def assign_from_json(self, data_json: dict): + """ + Assign only to existing non-callable attributes; ignore unknown/private keys. + """ + for k, v in data_json.items(): + if k.startswith("_"): + continue + try: + attr = getattr(self, k) + except AttributeError: + continue + if callable(attr): + continue + setattr(self, k, v)hugegraph-llm/src/hugegraph_llm/nodes/llm_node/extract_info.py (1)
48-53: 调度阶段缺少异常兜底。
run()抛出的异常会穿透到调度器。请捕获并返回CStatus，保证流程稳定。def operator_schedule(self, data_json): if self.extract_type == "triples": - return self.info_extract.run(data_json) + try: + return self.info_extract.run(data_json) + except Exception as exc: + return CStatus(-1, f"InfoExtract failed: {exc}") elif self.extract_type == "property_graph": - return self.property_graph_extract.run(data_json) + try: + return self.property_graph_extract.run(data_json) + except Exception as exc: + return CStatus(-1, f"PropertyGraphExtract failed: {exc}")
🧹 Nitpick comments (20)
hugegraph-llm/src/hugegraph_llm/operators/hugegraph_op/commit_to_hugegraph.py (1)
43-47: 请移除直接的生产路径中使用
log。- print(f"get schema {schema}") + log.debug("Received schema: %s", schema)hugegraph-llm/src/hugegraph_llm/nodes/util.py (1)
19-27: 细化缺参错误信息，便于定位问题当前返回“Required workflow parameters not found”无法区分缺失项。建议分别检测并回传具体缺少的参数键，提升可观测性。
应用如下补丁：
def init_context(obj) -> CStatus: try: - obj.context = obj.getGParamWithNoEmpty("wkflow_state") - obj.wk_input = obj.getGParamWithNoEmpty("wkflow_input") - if obj.context is None or obj.wk_input is None: - return CStatus(-1, "Required workflow parameters not found") - return CStatus() + context = obj.getGParamWithNoEmpty("wkflow_state") + wk_input = obj.getGParamWithNoEmpty("wkflow_input") + if context is None: + return CStatus(-1, "Required parameter 'wkflow_state' not found") + if wk_input is None: + return CStatus(-1, "Required parameter 'wkflow_input' not found") + obj.context = context + obj.wk_input = wk_input + return CStatus() except Exception as e: return CStatus(-1, f"Failed to initialize context: {str(e)}")hugegraph-llm/src/hugegraph_llm/state/ai_state.py (1)
28-35: 为新增输入字段补充类型注解，提升可读性与静态检查效果这些字段目前缺少类型注解，建议补全为 Optional[...]。
应用如下补丁：
- data_json = None - extract_type = None - query_examples = None - few_shot_schema = None + data_json: Optional[dict] = None + extract_type: Optional[str] = None + query_examples: Optional[List[Any]] = None + few_shot_schema: Optional[Any] = None - # Fields related to PromptGenerate - source_text: str = None # Original text - scenario: str = None # Scenario description - example_name: str = None # Example name + # Fields related to PromptGenerate + source_text: Optional[str] = None # Original text + scenario: Optional[str] = None # Scenario description + example_name: Optional[str] = None # Example name.vibedev/spec/hugegraph-llm/fixed_flow/design.md (1)
515-524: 文档命名与实现不一致：SchemaManagerNode vs SchemaNode实现文件为 SchemaNode（nodes/hugegraph_node/schema.py），文档此处称 SchemaManagerNode/CheckSchemaNode。建议统一为 SchemaNode，并在描述中说明其内部路由到 SchemaManager 或 CheckSchema。
hugegraph-llm/src/hugegraph_llm/nodes/hugegraph_node/schema.py (1)
64-75: 避免使用 print；按初始化结果路由到对应算子根据节点初始化结果路由至 check_schema 或 schema_manager，且使用统一日志。
应用如下补丁：
def operator_schedule(self, data_json): - print(f"check data json {data_json}") - if self.schema.startswith("{"): - try: - return self.check_schema.run(data_json) - except json.JSONDecodeError as exc: - log.error("Invalid JSON format in schema. Please check it again.") - raise ValueError("Invalid JSON format in schema.") from exc - else: - log.info("Get schema '%s' from graphdb.", self.schema) - return self.schema_manager.run(data_json) + log.debug("SchemaNode.operator_schedule input: %s", data_json) + if getattr(self, "check_schema", None) is not None: + return self.check_schema.run(data_json) + if getattr(self, "schema_manager", None) is not None: + return self.schema_manager.run(data_json) + return CStatus(-1, "SchemaNode is not initialized")hugegraph-llm/src/hugegraph_llm/flows/import_graph_data.py (2)
54-57: 避免在核心 Flow 中引入 UI 副作用（gr.Info）Flow 层应保持纯业务编排，UI 通知应放在调用方（web/demo）以免引入运行时 UI 依赖和测试困难。
建议移除 gr.Info 调用：
def post_deal(self, pipeline=None): res = pipeline.getGParamWithNoEmpty("wkflow_state").to_json() - gr.Info("Import graph data successfully!") return json.dumps(res, ensure_ascii=False, indent=2)
18-19: 移除未使用的 UI 依赖若按上条建议移除 gr.Info，则可同步去掉 gradio 依赖，减小运行面。
-import gradio as gr from PyCGraph import GPipelinehugegraph-llm/src/hugegraph_llm/flows/scheduler.py (2)
75-100: 避免局部变量名遮蔽入参 flow，提升可读性本地变量 flow 覆盖了方法入参 flow（字符串），易混淆。建议更名。
- flow: BaseFlow = self.pipeline_pool[flow]["flow"] + flow_obj: BaseFlow = self.pipeline_pool[flow]["flow"] pipeline: GPipeline = manager.fetch() if pipeline is None: # call coresponding flow_func to create new workflow - pipeline = flow.build_flow(*args, **kwargs) + pipeline = flow_obj.build_flow(*args, **kwargs) status = pipeline.init() @@ - res = flow.post_deal(pipeline) + res = flow_obj.post_deal(pipeline) manager.add(pipeline) return res else: # fetch pipeline & prepare input for flow prepared_input = pipeline.getGParamWithNoEmpty("wkflow_input") - flow.prepare(prepared_input, *args, **kwargs) + flow_obj.prepare(prepared_input, *args, **kwargs) status = pipeline.run() if status.isErr(): raise RuntimeError(f"Error in flow execution {status.getInfo()}") - res = flow.post_deal(pipeline) + res = flow_obj.post_deal(pipeline) manager.release(pipeline) return res
31-66: 未使用的 max_pipeline 参数max_pipeline 目前仅保存未使用。建议后续用于限制 manager 内活跃/缓存的 pipeline 数量，或移除该参数。
hugegraph-llm/src/hugegraph_llm/nodes/document_node/chunk_split.py (1)
37-41: 为空文本输入添加快速失败当 texts 为空时，尽早返回错误可避免后续算子无效执行。
if isinstance(texts, str): texts = [texts] + if not texts: + return CStatus(-1, "No texts provided for chunk split") self.chunk_split_op = ChunkSplit(texts, split_type, language) return CStatus()hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.py (4)
47-50: 移除未使用的 split_text 辅助函数当前模块未引用，建议删除以减小干扰。
-def split_text(text: str) -> List[str]: - chunk_splitter = ChunkSplitter(split_type="paragraph", language="zh") - chunks = chunk_splitter.split(text) - return chunks
65-66: 降低冗长日志级别以免污染 INFOproperties_map 体量可能较大，建议改为 debug。
- log.info("properties_map: %s", properties_map) + log.debug("properties_map: %s", properties_map)
92-99: 为上下文必需键添加快速校验与简化初始化避免 KeyError，同时用 setdefault 精简初始化。
- def run(self, context: Dict[str, Any]) -> Dict[str, List[Any]]: - schema = context["schema"] - chunks = context["chunks"] - if "vertices" not in context: - context["vertices"] = [] - if "edges" not in context: - context["edges"] = [] + def run(self, context: Dict[str, Any]) -> Dict[str, List[Any]]: + if "schema" not in context or "chunks" not in context: + raise KeyError("context requires 'schema' and 'chunks'") + schema = context["schema"] + chunks = context["chunks"] + context.setdefault("vertices", []) + context.setdefault("edges", []) items = []
120-123: 避免局部变量名与模块 prompt 同名造成可读性混淆重命名为 prompt_text 更直观。
- prompt = generate_extract_property_graph_prompt(chunk, schema) - if self.example_prompt is not None: - prompt = self.example_prompt + prompt - return self.llm.generate(prompt=prompt) + prompt_text = generate_extract_property_graph_prompt(chunk, schema) + if self.example_prompt is not None: + prompt_text = self.example_prompt + prompt_text + return self.llm.generate(prompt=prompt_text)hugegraph-llm/src/hugegraph_llm/flows/update_vid_embeddings.py (1)
27-35: prepare() 的返回状态未处理，可能掩盖初始化失败build_flow 调用 prepare() 但未检查 CStatus，初始化失败仍继续注册节点，后续异常更隐蔽。
建议在此处检查 isErr() 并中止/抛错，或统一至 BaseFlow 的约定（如果有）。例如：
- self.prepare(prepared_input) + sts = self.prepare(prepared_input) + if hasattr(sts, "isErr") and sts.isErr(): + # 可根据你们的约定选择抛错或返回 None 交由上层处理 + raise RuntimeError("prepare() failed for UpdateVidEmbeddingsFlows")hugegraph-llm/src/hugegraph_llm/nodes/base_node.py (1)
52-57: 统一 operator_schedule 返回约定，避免歧义注释宣称返回 CStatus，但 run() 中当作数据字典使用。应明确：返回 dict 合并上下文，或返回 CStatus 表示失败。
- 节点调度operator的接口，子类可重写。 - 返回CStatus对象，表示调度是否成功。 + 节点调度 operator 的接口，子类可重写。 + 返回: + - dict: 将合并进 context + - CStatus: 表示错误/状态（非 OK 时中止）hugegraph-llm/src/hugegraph_llm/flows/prompt_generate.py (1)
25-28: 移除空的 init 以简化代码无状态且无逻辑的构造器可去除。
-class PromptGenerateFlow(BaseFlow): - def __init__(self): - pass +class PromptGenerateFlow(BaseFlow):hugegraph-llm/src/hugegraph_llm/nodes/llm_node/prompt_generate.py (1)
33-45: 补充健壮性：校验 wk_input/LLM 实例
- 若 init_context 未正确注入 wk_input，访问属性会抛 AttributeError。
- get_chat_llm 失败返回 None 时，后续 PromptGenerate 初始化将失败。
[scheggest_recommended_refactor]- llm = get_chat_llm(llm_settings) - if not all( + llm = get_chat_llm(llm_settings) + if llm is None: + return CStatus(-1, "LLM initialization failed") + if not self.wk_input or not all( [ self.wk_input.source_text, self.wk_input.scenario, self.wk_input.example_name, ] ): return CStatus( -1, "Missing required parameters: source_text, scenario, or example_name", )hugegraph-llm/src/hugegraph_llm/utils/graph_index_utils.py (1)
186-196: 异常处理不应打印堆栈到 stdout；保持统一的用户提示traceback.print_exc() 在 Web 场景下噪声较大；建议仅日志文件记录详细堆栈，UI 用 gr.Warning。
- except Exception as e: # pylint: disable=W0718 - log.error(e) - traceback.print_exc() + except Exception as e: # pylint: disable=W0718 + log.exception("import_graph_data failed") # Note: can't use gr.Error here - gr.Warning(str(e) + " Please check the graph data format/type carefully.") + gr.Warning(str(e) + " Please check the graph data format/type carefully.") return datahugegraph-llm/src/hugegraph_llm/nodes/llm_node/schema_build.py (1)
33-46: 建议补充对 wk_input/LLM 的健壮性校验为与 PromptGenerateNode 一致，建议在 node_init 初始阶段校验 llm 与 wk_input，并在缺失时返回 CStatus(-1, msg)。
📜 Review details
Configuration used: CodeRabbit UI
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (33)
.vibedev/spec/hugegraph-llm/fixed_flow/design.md(1 hunks).vibedev/spec/hugegraph-llm/fixed_flow/requirements.md(1 hunks).vibedev/spec/hugegraph-llm/fixed_flow/tasks.md(1 hunks)hugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py(2 hunks)hugegraph-llm/src/hugegraph_llm/flows/build_schema.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/build_vector_index.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/get_graph_index_info.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/graph_extract.py(3 hunks)hugegraph-llm/src/hugegraph_llm/flows/import_graph_data.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/prompt_generate.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/scheduler.py(3 hunks)hugegraph-llm/src/hugegraph_llm/flows/update_vid_embeddings.py(1 hunks)hugegraph-llm/src/hugegraph_llm/flows/utils.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/base_node.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/document_node/chunk_split.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/hugegraph_node/commit_to_hugegraph.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/hugegraph_node/fetch_graph_data.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/hugegraph_node/schema.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/index_node/build_semantic_index.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/index_node/build_vector_index.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/llm_node/extract_info.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/llm_node/prompt_generate.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/llm_node/schema_build.py(1 hunks)hugegraph-llm/src/hugegraph_llm/nodes/util.py(1 hunks)hugegraph-llm/src/hugegraph_llm/operators/common_op/check_schema.py(0 hunks)hugegraph-llm/src/hugegraph_llm/operators/document_op/chunk_split.py(0 hunks)hugegraph-llm/src/hugegraph_llm/operators/hugegraph_op/commit_to_hugegraph.py(10 hunks)hugegraph-llm/src/hugegraph_llm/operators/hugegraph_op/schema_manager.py(0 hunks)hugegraph-llm/src/hugegraph_llm/operators/index_op/build_vector_index.py(0 hunks)hugegraph-llm/src/hugegraph_llm/operators/llm_op/info_extract.py(0 hunks)hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.py(1 hunks)hugegraph-llm/src/hugegraph_llm/state/ai_state.py(5 hunks)hugegraph-llm/src/hugegraph_llm/utils/graph_index_utils.py(4 hunks)
💤 Files with no reviewable changes (5)
- hugegraph-llm/src/hugegraph_llm/operators/hugegraph_op/schema_manager.py
- hugegraph-llm/src/hugegraph_llm/operators/index_op/build_vector_index.py
- hugegraph-llm/src/hugegraph_llm/operators/common_op/check_schema.py
- hugegraph-llm/src/hugegraph_llm/operators/llm_op/info_extract.py
- hugegraph-llm/src/hugegraph_llm/operators/document_op/chunk_split.py
🧰 Additional context used
🧠 Learnings (8)
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/operators/**/*.py : Put core processing pipelines under src/hugegraph_llm/operators/
Applied to files:
hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.pyhugegraph-llm/src/hugegraph_llm/flows/build_vector_index.py
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/operators/graph_rag_task.py : Maintain the Graph RAG pipeline in src/hugegraph_llm/operators/graph_rag_task.py
Applied to files:
hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.pyhugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py
📚 Learning: 2025-06-25T09:50:06.213Z
Learnt from: day0n
PR: hugegraph/hugegraph-ai#16
File: hugegraph-llm/src/hugegraph_llm/config/models/base_prompt_config.py:124-137
Timestamp: 2025-06-25T09:50:06.213Z
Learning: Language-specific prompt attributes (answer_prompt_CN, answer_prompt_EN, extract_graph_prompt_CN, extract_graph_prompt_EN, gremlin_generate_prompt_CN, gremlin_generate_prompt_EN, keywords_extract_prompt_CN, keywords_extract_prompt_EN, doc_input_text_CN, doc_input_text_EN) are defined in the PromptConfig class in hugegraph-llm/src/hugegraph_llm/config/prompt_config.py, which inherits from BasePromptConfig, making these attributes accessible in the parent class methods.
Applied to files:
hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.pyhugegraph-llm/src/hugegraph_llm/flows/prompt_generate.pyhugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.pyhugegraph-llm/src/hugegraph_llm/nodes/llm_node/prompt_generate.py
📚 Learning: 2025-06-25T09:45:10.751Z
Learnt from: day0n
PR: hugegraph/hugegraph-ai#16
File: hugegraph-llm/src/hugegraph_llm/config/models/base_prompt_config.py:100-116
Timestamp: 2025-06-25T09:45:10.751Z
Learning: In hugegraph-llm BasePromptConfig class, llm_settings is a runtime property that is loaded from config through dependency injection during object initialization, not a static class attribute. Static analysis tools may flag this as missing but it's intentional design.
Applied to files:
hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.pyhugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/resources/demo/config_prompt.yaml : Keep prompt configuration in src/hugegraph_llm/resources/demo/config_prompt.yaml
Applied to files:
hugegraph-llm/src/hugegraph_llm/operators/llm_op/property_graph_extract.pyhugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/models/**/*.py : Implement LLM, embedding, and reranker models under src/hugegraph_llm/models/
Applied to files:
hugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/indices/**/*.py : Store vector and graph indexing code under src/hugegraph_llm/indices/
Applied to files:
hugegraph-llm/src/hugegraph_llm/flows/build_vector_index.pyhugegraph-llm/src/hugegraph_llm/utils/graph_index_utils.py
📚 Learning: 2025-09-16T06:40:44.968Z
Learnt from: CR
PR: hugegraph/hugegraph-ai#0
File: hugegraph-llm/AGENTS.md:0-0
Timestamp: 2025-09-16T06:40:44.968Z
Learning: Applies to hugegraph-llm/src/hugegraph_llm/operators/kg_construction_task.py : Maintain the KG Construction pipeline in src/hugegraph_llm/operators/kg_construction_task.py
Applied to files:
hugegraph-llm/src/hugegraph_llm/flows/build_vector_index.py
🪛 markdownlint-cli2 (0.18.1)
.vibedev/spec/hugegraph-llm/fixed_flow/tasks.md
21-21: Fenced code blocks should have a language specified
(MD040, fenced-code-language)
🔇 Additional comments (8)
.vibedev/spec/hugegraph-llm/fixed_flow/tasks.md (1)
1-36: 文档结构清晰、任务拆分合理任务列表对齐本次重构目标，勾选项直观可追踪，便于后续迭代。
.vibedev/spec/hugegraph-llm/fixed_flow/requirements.md (1)
1-25: 文档条理清晰需求与验收标准明确，已完成功能列表完整。
hugegraph-llm/src/hugegraph_llm/flows/build_vector_index.py (1)
17-18: 批准节点层迁移调整 引用已从 operators 切换到 nodes，符合 Node/Operator 解耦方向，依赖关系清晰。已确认 ChunkSplitNode 与 BuildVectorIndexNode 正确定义并可导入。hugegraph-llm/src/hugegraph_llm/nodes/hugegraph_node/commit_to_hugegraph.py (1)
27-33: 无需在 node_init 中判空 context
context 已由 BaseNode.init 调用 init_context 保证注入并非空，无需额外校验。Likely an incorrect or invalid review comment.
hugegraph-llm/src/hugegraph_llm/flows/graph_extract.py (1)
39-58: 确认 ExtractNode 已正确消费 extract_typeExtractNode 在 node_init 中通过
self.wk_input.extract_type读取并赋值给self.extract_type，operator_schedule已按该值分支执行，参数无遗漏。hugegraph-llm/src/hugegraph_llm/demo/rag_demo/vector_graph_block.py (1)
71-78: 确认调度结果类型与 UI 输出组件一致generate_prompt_for_ui 返回 scheduler.schedule_flow 的结果；该结果需为字符串以填充 gr.Code(language="markdown")。若返回 dict/对象，将显示异常。
请确认 PromptGenerateFlow.post_deal 返回为字符串（见 flows/prompt_generate.py 的实现）。如存在不一致，建议在此处统一转为字符串。
hugegraph-llm/src/hugegraph_llm/flows/prompt_generate.py (1)
60-63: 确认状态键名与算子输出一致post_deal 读取 generated_extract_prompt。请确认 PromptGenerate.run 的返回/上下文写入使用相同键名，否则将返回默认失败文案。
如需增强健壮性，可提供备用键或回退逻辑（如从 context 中不同字段容错读取）。
hugegraph-llm/src/hugegraph_llm/nodes/base_node.py (1)
16-16: 确认PyCGraph导入与依赖名一致
仓库中多处使用from PyCGraph import …，但pyproject.toml中声明依赖为pycgraph（小写），请确认安装包名与导入名称及大小写一致，避免 CI/部署 报错。
hugegraph-llm/src/hugegraph_llm/operators/hugegraph_op/commit_to_hugegraph.py
Show resolved
Hide resolved
This was referenced Sep 27, 2025
This was referenced Oct 9, 2025
Closed
weijinglin
added a commit
that referenced
this pull request
Oct 22, 2025
weijinglin
added a commit
that referenced
this pull request
Oct 23, 2025
…usecases in web demo (#49)
github-actions bot
pushed a commit
that referenced
this pull request
Oct 23, 2025
…usecases in web demo (#49)
Closed
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Usecase
Add necessary doc for ai development
Summary by CodeRabbit