maximum recursion depth exceeded in comparison

[Major-wagh]

Receiving the following error when :
2024-07-09 12:48:43.996 | SUCCESS | repo_agent.log:set_logger_level_from_config:74 - Log level set to INFO!
parsing parent relationship: 0%| parsing parent relationship: 100%|█████████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 8184.01it/s]
Loading MetaInfo: /home/jovyan/work/Documentation/Arya/wiki/hierarchy_files
MetaInfo is Refreshed and Saved
2024-07-09 12:48:44.015 | INFO | repo_agent.runner:first_generate:104 - Starting to generate documentation
parsing bidirectional reference: 0%| | 0/2 [00:00<?, ?it/s]2024-07-09 12:48:44.198 | INFO | repo_agent.doc_meta_info:find_all_referencer:293 - Error occurred: column parameter (6) is not in a valid range (0-0) for line 205 ('\n').
2024-07-09 12:48:44.198 | INFO | repo_agent.doc_meta_info:find_all_referencer:294 - Parameters: variable_name=Embed, file_path=wikiv1.py, line_number=205, column_number=6
2024-07-09 12:48:44.208 | INFO | repo_agent.doc_meta_info:find_all_referencer:293 - Error occurred: line parameter is not in a valid range.
2024-07-09 12:48:44.208 | INFO | repo_agent.doc_meta_info:find_all_referencer:294 - Parameters: variable_name=create_collection, file_path=wikiv1.py, line_number=284, column_number=4
parsing bidirectional reference: 50%|█████████████████████████████████████▌ | 1/2 [00:00<00:00, 5.17it/s]2024-07-09 12:48:44.212 | INFO | repo_agent.doc_meta_info:find_all_referencer:293 - Error occurred: column parameter (6) is not in a valid range (0-0) for line 205 ('\n').
2024-07-09 12:48:44.213 | INFO | repo_agent.doc_meta_info:find_all_referencer:294 - Parameters: variable_name=Embed, file_path=.ipynb_checkpoints/wikiv1-checkpoint.py, line_number=205, column_number=6
2024-07-09 12:48:44.220 | INFO | repo_agent.doc_meta_info:find_all_referencer:293 - Error occurred: line parameter is not in a valid range.
2024-07-09 12:48:44.220 | INFO | repo_agent.doc_meta_info:find_all_referencer:294 - Parameters: variable_name=create_collection, file_path=.ipynb_checkpoints/wikiv1-checkpoint.py, line_number=284, column_number=4
parsing bidirectional reference: 100%|███████████████████████████████████████████████████████████████████████████| 2/2 [00:00<00:00, 9.74it/s]
parsing topology task-list: 0%| | 0/10 [00:00<?, ?it/s]Traceback (most recent call last):
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/main.py", line 312, in
cli()
File "/opt/conda/envs/repoenv/lib/python3.11/site-packages/click/core.py", line 1157, in call
return self.main(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/repoenv/lib/python3.11/site-packages/click/core.py", line 1078, in main
rv = self.invoke(ctx)
^^^^^^^^^^^^^^^^
File "/opt/conda/envs/repoenv/lib/python3.11/site-packages/click/core.py", line 1688, in invoke
return _process_result(sub_ctx.command.invoke(sub_ctx))
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/repoenv/lib/python3.11/site-packages/click/core.py", line 1434, in invoke
return ctx.invoke(self.callback, **ctx.params)
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/opt/conda/envs/repoenv/lib/python3.11/site-packages/click/core.py", line 783, in invoke
return __callback(*args, **kwargs)
^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/main.py", line 260, in run
runner.run()
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/runner.py", line 240, in run
self.first_generate() # 如果是第一次做文档生成任务，就通过first_generate生成所有文档
^^^^^^^^^^^^^^^^^^^^^
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/runner.py", line 106, in first_generate
task_manager = self.meta_info.get_topology(
^^^^^^^^^^^^^^^^^^^^^^^^^^^^
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/doc_meta_info.py", line 619, in get_topology
task_manager = self.get_task_manager(
^^^^^^^^^^^^^^^^^^^^^^
File "/home/jovyan/work/repoagent/RepoAgent/repo_agent/doc_meta_info.py", line 577, in get_task_manager
if task_available_func(child) and (child not in deal_items):
^^^^^^^^^^^^^^^^^^^^^^^
File "", line 4, in eq
File "", line 4, in eq
File "", line 4, in eq
[Previous line repeated 851 more times]
RecursionError: maximum recursion depth exceeded in comparison
parsing topology task-list: 90%|███████████████████████████████████████████████████████████████████████ | 9/10 [00:00<00:00, 34.28it/s]

[DanielNSaga]

I hit the same crash in repo_agent/doc_meta_info.py at “parsing topology task-list”. I fixed it by switching the membership checks in get_task_manager to identity so DocItem.__eq__ never runs on cycles. I think this works because what we actually want is “did I already process this exact object instance?”, and id(...) answers that without recursing.
Try switching the get_task_manager method with this:

def get_task_manager(self, now_node: DocItem, task_available_func) -> TaskManager:
    """先写一个退化的版本，只考虑拓扑引用关系"""
    doc_items = now_node.get_travel_list()
    if self.white_list != None:

        def in_white_list(item: DocItem):
            for cont in self.white_list:
                if (
                    item.get_file_name() == cont["file_path"]
                    and item.obj_name == cont["id_text"]
                ):
                    return True
            return False

        doc_items = list(filter(in_white_list, doc_items))
    doc_items = list(filter(task_available_func, doc_items))
    doc_items = sorted(doc_items, key=lambda x: x.depth)  # 叶子节点在前面
    deal_items = []
    task_manager = TaskManager()
    bar = tqdm(total=len(doc_items), desc="parsing topology task-list")
    while doc_items:
        min_break_level = 1e7
        target_item = None
        for item in doc_items:
            """一个任务依赖于所有引用者和他的子节点,我们不能保证引用不成环(也许有些仓库的废代码会出现成环)。
            这时就只能选择一个相对来说遵守程度最好的了
            有特殊情况func-def中的param def可能会出现循环引用
            另外循环引用真实存在，对于一些bind类的接口真的会发生，比如：
            ChatDev/WareHouse/Gomoku_HumanAgentInteraction_20230920135038/main.py里面的: on-click、show-winner、restart
            """
            best_break_level = 0
            second_best_break_level = 0

            for _, child in item.children.items():
                if task_available_func(child) and all(id(child) != id(d) for d in deal_items):
                    best_break_level += 1

            for referenced, special in zip(item.reference_who, item.special_reference_type):
                if task_available_func(referenced) and all(id(referenced) != id(d) for d in deal_items):
                    best_break_level += 1
                if (
                    task_available_func(referenced)
                    and not special
                    and all(id(referenced) != id(d) for d in deal_items)
                ):
                    second_best_break_level += 1

            if best_break_level == 0:
                min_break_level = -1
                target_item = item
                break
            if second_best_break_level < min_break_level:
                target_item = item
                min_break_level = second_best_break_level

        if min_break_level > 0:
            print(
                f"circle-reference(second-best still failed), level={min_break_level}: {target_item.get_full_name()}"
            )

        item_denp_task_ids = []
        for _, child in target_item.children.items():
            if child.multithread_task_id != -1:
                assert child.multithread_task_id in task_manager.task_dict.keys()
                item_denp_task_ids.append(child.multithread_task_id)
        for referenced_item in target_item.reference_who:
            if referenced_item.multithread_task_id in task_manager.task_dict.keys():
                item_denp_task_ids.append(referenced_item.multithread_task_id)
        item_denp_task_ids = list(set(item_denp_task_ids))  # 去重
        if task_available_func == None or task_available_func(target_item):
            task_id = task_manager.add_task(
                dependency_task_id=item_denp_task_ids, extra=target_item
            )
            target_item.multithread_task_id = task_id
        deal_items.append(target_item)
        doc_items.remove(target_item)
        bar.update(1)

    return task_manager