[Bug]: Newer export function vastly reduced messages output

[bahehs]

Version

2.3.4

Describe the bug

I want to report some different behavior compared to what I experienced previously. Maybe bug isn't the right issue to place it under but don't see anything else applicable.

When using the export function right now it calls the following which results in a small file (600kb) with a maximum of 99 responses

https://www.perplexity.ai/rest/thread/[threadUID]?version=2.18&source=default

Steps to reproduce

1. Go perplexity and login
2. open a thread with more than 100 messages or even 500 messages (this is a rare scenario but I can test it since I have some threads that ran longer)
3. observe it only exports the last 99 messages

Bug confirmation

• I am not able to reproduce the issue with the extension disabled
• I have re-installed the extension and the issue persists

Expected behavior

I think the call used to be something like this which was way better as it grabbed around 450 responses and came about to (1.8mb) .

https://www.perplexity.ai/p/api/v1/thread/[threadUID]?with_parent_info=true&source=web&limit=9999

The reason I ask is because I have some chats that I didn't bother to create a new thread and they ballooned to around 600 messages and I'm unable to export them at their full length, i think the maximum I was able to export previously was around 467 messages and right now only able to get 99.

Screenshots

No response

Browser

Firefox 141.0.3

Additional context

No response

[pnd280]

does https://www.perplexity.ai/rest/thread/[threadSlug]?version=2.18&source=default&limit=9999 work for you?

[bahehs]

Hi,
yes that works, I am using the console and looking at the exported data in the json using

const apiUrl = 'https://www.perplexity.ai/rest/thread/[threadSlug]?version=2.18&source=default&limit=9999';
const filename = 'perplexity_thread_data.json'; // Choose your desired filename

fetch(apiUrl)
  .then(response => {
    if (!response.ok) {
      throw new Error(`HTTP error! status: ${response.status}`);
    }
    // Get the response as a Blob (binary large object)
    // which is suitable for file download
    return response.blob();
  })
  .then(blob => {
    // Create a temporary URL for the Blob
    const url = window.URL.createObjectURL(blob);

    // Create a temporary anchor element (<a>)
    const a = document.createElement('a');
    a.style.display = 'none'; // Hide the element
    a.href = url;
    a.download = filename; // Set the filename for download

    // Append the anchor to the body, click it, and remove it
    document.body.appendChild(a);
    a.click();
    document.body.removeChild(a); // Clean up the anchor element

    // Revoke the temporary URL to free up memory
    window.URL.revokeObjectURL(url);

    console.log(`Download prompted for: ${filename}`);
  })
  .catch(error => {
    console.error('Error fetching or processing API for download:', error);
    // Attempt to log text response in case of non-blob error
    fetch(apiUrl).then(r => r.text()).then(t => console.log('Raw response text on error:', t));
  });

I am able to get a much more expansive export of messages.
However, I do prefer your previous method of using the API call as there is no rate limiting.
https://www.perplexity.ai/p/api/v1/thread/[threadSlug]?with_parent_info=true&source=web&limit=9999
using the rest call, it throttles fairly quickly (maybe 10? requests and I got limited but with the API call I don't observe that behavior.