Implementing Google Scholar Support

[shamssab]

Checked

• I searched existing ideas and did not find a similar one
• I added a very descriptive title
• I've clearly described the feature request and motivation for it

Feature request

We want to integrate Google Scholar functionality into the LangChain JS version, similar to the existing implementation in Python. This integration will provide tools for querying Google Scholar and handling the results efficiently within the JavaScript ecosystem.

Motivation

The motivation for creating a Google Scholar tool in JavaScript stems from the growing need for compatibility with JavaScript-based applications and the desire to have feature parity with the LangChain Python library. The Python version includes robust tools for interacting with Google Scholar, but the JS version currently lacks this functionality, limiting developers’ ability to leverage academic resources in their applications.

Proposal (If applicable)

We are a group of 4 students from the University of Toronto and we are eager to contribute to Langchain!

Here is a basic pseudocode implementation of the Google Scholar query tool.

class GoogleScholarQuery {
  constructor(query) {
    this.query = query;
  }

  async run() {
    // Placeholder for API request to Google Scholar
    const response = await fetch(`https://scholar.google.com/scholar?q=${encodeURIComponent(this.query)}`);
    const results = await response.json();

    return results.map(result => ({
      title: result.title,
      link: result.link,
      snippet: result.snippet,
    }));
  }
}

This is a basic pseudocode implementation of the Google Scholar query tool. We would also need to implement the necessary functions to handle API interactions and data formatting.

Example Usage

try {
  const query = new GoogleScholarQuery("Your search term here");
  const results = await query.run();
  
  console.log("Search Results:", results);
} catch (error) {
  console.error("Error querying Google Scholar:", error);
}

Our Next Steps:

• Define the complete class structure, including error handling and API rate limiting.
• Implement the necessary API requests to retrieve data from Google Scholar.
• Create corresponding documentation and examples for usage within the LangChain JS ecosystem.

[shamssab]

Hi @jacoblee93, we were wondering if this is something we should create an issue for. We've already started working on the coding side of it!

[shamssab]

Hey @jacoblee93! Just wanted to check in whenever you have a moment—I'd love to hear your thoughts on this proposal. Thanks so much!

[jacoblee93]

Hey sounds good, feel free to make a PR!

[codepeach]

Any updates on this? I'd love an alternative to serpAPI for google scholar searches.

[shamssab]

Hi @codepeach! Yes, we're working on this as part of our school project, and we’re aiming to have a pull request ready in about two weeks. If you have any APIs in mind that might be a good alternative to SerpAPI for Google Scholar searches, we'd love to hear your suggestions! Thanks for your interest! 😊

[praj18]

Hey, I have an alternative that I'm currently using Tavily, they're much more sparing in terms of API calls compared to SerpAPI. All you have to do is add google scholar's website to the list of domains for the 'includeDomains' field when calling the api.