community: Corrected aload func to be asynchronous from webBaseLoader

[yeounhak]

• Description: The aload function, contrary to its name, is not an asynchronous function, so it cannot work concurrently with other asynchronous functions.

• Issue: 🐛Bug: The aload function, contrary to its name, is not an asynchronous function, so it cannot work concurrently with other asynchronous functions. #28336

• **Test: **: Done

• **Docs: ** here

• **Lint: ** All checks passed

If no one reviews your PR within a few days, please @-mention one of baskaryan, efriis, eyurtsev, ccurme, vbarda, hwchase17.

[vercel]

The latest updates on your projects. Learn more about Vercel for Git ↗︎

Name	Status	Preview	Comments	Updated (UTC)
langchain	✅ Ready (Inspect)	Visit Preview	💬 Add feedback	Dec 20, 2024 7:32pm

[yeounhak]

I understand you might be busy, but I would greatly appreciate it if you could provide comments on this PR.

@baskaryan @efriis @eyurtsev @ccurme @vbarda @hwchase17

[efriis]

tapping in @eyurtsev on the WebBaseLoader type ignore in #19936

I think this one might be funky for backwards compatibility reasons. I'm open to a breaking bugfix here

[eyurtsev]

Thanks for spending the time on this! I suggest a way to partially handle the problem without making a breaking change to this code since it's used by a lot of people as is

[eyurtsev]

sorry forgot to actually submit the comments

[yeounhak]

I will make the revisions based on the comments. Thank you!

[yeounhak]

✅ [What I've done]

1. Implementing the load() function to execute concurrently in a new thread for Jupyter environment users.

2. Keep aload() as a synchronous version with a warning message stating the following:

   • Users should replace aload() with load().
   • Or they can replace it with alazy_load().

3. Add 'a' prefix for the scrape_all() function

⏳ [What remains to do]

• Include a warning message specifying in which version the aload() function will be updated to an asynchronous version.
• Deleting the aload() function definition from web_base.py when updating to a major version.

Could you please provide comments on the revised code?
Thank you in advance for your kind feedback! 🙏
@eyurtsev

[ccurme]

Thank you @yeounhak!

I've made some updates:

• Keep .scrape_all as this is in the public API (your updating the tests in other integrations previews what developers would have to do if we deleted it).
• Keep behavior of synchronous loading unchanged.
• Update docs.

So in effect we are

1. Implementing .ascrape_all (includes a slight refactor)
2. Implementing .alazy_load
3. Deprecating aload.