unit-mesh / auto-dev-vscode

AutoDev - 🧙‍the AI-powered coding wizard . Put the most loved AutoDev AI assistant into your VSCode, and have things done quickly
http://vscode.unitmesh.cc/
MIT License
234 stars 38 forks source link

indexing codebase报错 #43

Closed anrgct closed 3 months ago

anrgct commented 3 months ago

Before submitting your bug report

Relevant environment info

- OS: macos 12.7.5
- AutoDev VSCode: v0.3.1
- IDE: vscode 1.89.1 (Universal)

Description

点击indexing后等待处理,之后报错,完整最后log信息:

To reproduce

No response

Log output

2024-05-24 21:08:58.217 [info] indexing progress: 0.7551490514905149 - Indexing LICENSE
2024-05-24 21:08:58.219 [info] indexing progress: 0.7552168021680217 - Indexing LICENSE
2024-05-24 21:08:58.359 [info] Failed to embed chunkbloop stands on the shoulders of the Rust ecosystem. Our search indexes are powered by [Tantivy](https://github.com/quickwit-oss/tantivy) and [Qdrant](https://github.com/qdrant/qdrant), and our multi-platform app is built with [Tauri](https://github.com/tauri-apps/tauri).

https://github.com/BloopAI/bloop/assets/7957964/93715188-d8d5-477b-8cd1-95d9cbd368cb

## Get Started

The simplest way to get started with bloop is to [download the app](https://github.com/BloopAI/bloop/releases) and follow the onboarding steps. Checkout our [getting started guide](https://bloop.ai/understand/docs/getting-started) and our references for [conversational](https://bloop.ai/understand/docs/natural-language-queries) and [regex](https://bloop.ai/understand/docs/regex-queries) search and [Code Studio](https://bloop.ai/understand/docs/code-studio).

For instructions on how to build from source or run bloop from the command line, check out these pages:

- [Build bloop app from source](./apps/desktop/README.md)
- [Run bloop from the command line](./server/README.md)

If you encounter any index issues you can wipe the bloop cache and reindex. Instructions on how to do this on different platforms [are here](./apps/desktop/README.md).

## Building From Source

You can build bloop from source and run it with your own OpenAI API key. Clone the repo, make sure the `oss` branch is checked out, and create a file called `local_config.json` at the top-level of the repo. `local_config.json` should contain the following fields:

Error: Non-zero status code returned while running Add node. Name:'/embeddings/Add_1' Status Message: /Users/runner/work/1/s/onnxruntime/core/providers/cpu/math/element_wise_ops.h:560 void onnxruntime::BroadcastIterator::Append(ptrdiff_t, ptrdiff_t) axis == 1 || axis == largest was false. Attempting to broadcast an axis by a dimension other than 1. 512 by 521

2024-05-24 21:08:58.519 [info] indexing progress: 0.7552845528455284 - Indexing README.md
2024-05-24 21:08:58.521 [info] indexing progress: 0.7553861788617886 - Indexing README.md
2024-05-24 21:08:58.522 [info] indexing progress: 0.7554878048780488 - Indexing README.md
2024-05-24 21:08:58.525 [info] indexing progress: 1 - Indexing Complete
phodal commented 3 months ago

看上去没啥问题,偶尔会有一些 chunk 超了,暂时没有更好的办法。

不过,现在不支持 Rust 。

anrgct commented 3 months ago

确实可以,但是试了几个问题都没有搜索到相关,可能不太会用。ps. 想要默认中文的输出的提示词,看英文费劲。

phodal commented 3 months ago

@anrgct 现在的设置里是如果 VSCode 用中文,就会用中文提示词。如果不行的话,可能是有 bug,可以来新 issue 。