aws / aws/bedrock-agentcore-sdk-python

Pagination discrepancy for MemorySessionManager ---> search_long_term_memories and list_long_term_memory_records

未关闭
#327 1 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看
主要语言
Python
星标
761
派生
147
平均合并
1 天 23 小时
30 天内合并 PR
7

描述

**Describe the bug**
MemorySessionManager has 2 memory listing methods named search_long_term_memories and list_long_term_memory_records

as per boto3 docs both the methods should support pagination with nextToken

i can see list_long_term_memory_records has some pagination inside it and fetches all the records in loop where as search_long_term_memories doesnt have one

both methods pagination logic should be similar as per boto3 actual methods? if yes, search_long_term_memories also needs same looping logic and fetching all records

Also as list_long_term_memory_records brings up all the records in a loop and then do the slicing as per max_results,

- what if the records are huge in number?
- Is sequential pagination will be available in future? as even if i increase max_results value to a bigger number, i will get the values from the beginning till max_result always

**Screenshots**

Image

Image

boto3 links:
https://docs.aws.amazon.com/boto3/latest/reference/services/bedrock-agentcore/client/retrieve_memory_records.html
https://docs.aws.amazon.com/boto3/latest/reference/services/bedrock-agentcore/client/list_memory_records.html

贡献指南

打开贡献指南

调研方向

从 MemorySessionManager.search_long_term_memories 和 list_long_term_memory_records 开始,然后将它们对 nextToken 的处理与链接的 boto3 retrieve_memory_records 和 list_memory_records 文档进行比较。确认预期的分页和 max_results 行为;完成的标准是两个方法都具有一致且有文档说明的结果,不存在含糊的检索语义。

由索引模型根据 Issue 内容生成。

评估

技术栈
aws, python
领域
api, backend
Issue 类型
缺陷
难度
4/5
预计耗时
3-5 天
活跃度
停滞
描述清晰度
基本清楚
新手友好度
42/100

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。