run "yarn run ingest" Japanese punctuation marks were converted to Korean about gpt4-pdf-chatbot-langchain

[dosubot]

🤖

Thank you for providing the detailed error log. It seems like the issue is related to the text processing part of the ingestion process. The yarn run ingest command uses a script that includes a text splitting step, which might be where the punctuation conversion is happening.

The script uses RecursiveCharacterTextSplitter to split the text into chunks. If this splitter is not handling Japanese punctuation correctly, it could be converting it to Korean punctuation during the splitting process.

However, without access to the specific implementation of RecursiveCharacterTextSplitter and the rest of the codebase, it's hard to pinpoint the exact cause of the issue. It would be helpful if you could provide more information about the RecursiveCharacterTextSplitter and how it's implemented, or any other part of the code that might be handling text processing.

In the meantime, you might want to check the configuration of the text splitter and see if there's a setting that's causing this issue. If the splitter is using a library or service for text processing, you might also want to check the documentation for that to see if there's any known issues or settings that could cause this.

If you're unable to find a solution, it might be worth considering using a different method for text splitting that can handle Japanese punctuation correctly. There are many libraries and services available that can handle text processing for different languages, so you might be able to find one that works better for your needs.

I hope this helps! If you have any more questions or need further clarification, please don't hesitate to ask.

Source Code

Sources

• README.md
• scripts/ingest-data.ts

About Dosu

This response is meant to be useful and save you time. It is not meant to be a precise solution, but rather a starting point for your own research.

Help me be more useful! Please leave a 👍 if this is helpful and 👎 if it is irrelevant.

If you want to continue the conversation, start your reply with @dosu-bot.