LensError: Could not parse response: TypeError: Cannot read properties of null (reading '0')

[TheRealStingo]

Sometimes when you i do requests . it doesnt retun the JSON object but instead it returns and error and some html body ,

it works like 3-4 times and then the issue appears , its like a russian roullete , sometimes i get it 3 times in a row and sometimes i dont get it in like 200 tries ( using the same picture over and over again )

[dimdenGD]

In order for me to debug this, you need to provide full body text

[TheRealStingo]

In order for me to debug this, you need to provide full body text

error_log.txt
here

[dimdenGD]

I couldn't reproduce your error with provided body. Make sure you're using latest version of the library

[TheRealStingo]

Make sure you're using latest version of the library

i already did , also a small question . what about rate limit . is there a way to add proxy support ?

[dimdenGD]

Proxies are already supported, check readme

[TheRealStingo]

im still getting the same error just by running the example in the readme . could you please watch this video to understand more
https://www.veed.io/view/ad9c1b4a-2c73-4373-9fdf-47b6647cb9bb?panel=share

[Arikips]

Same for me. I think it's because of CAPTCHa or VPN

[andreamagazzini]

Same for me. I think it's because of CAPTCHa or VPN

Did you find any solution?

[osmarks]

This happens inconsistently to me. Here's the body: https://osmarks.net/stuff/ocr_goog.txt

[Arikips]

Same for me. I think it's because of CAPTCHa or VPN

Did you find any solution?

This happens inconsistently to me. Here's the body: https://osmarks.net/stuff/ocr_goog.txt

Yes I think I did.
It seems like the problem is somewhere in the lib itself. I tried owocr(https://github.com/AuroraWright/owocr) and it works without any troubles. Also I have written a script of my own based off it where I added extra image preprocessing to speed up the scanning, so extracting big text from my 4k monitor takes about 1.2 seconds. I think I will upload it to my github and send you guys the link.

[dimdenGD]

This happens inconsistently to me. Here's the body: https://osmarks.net/stuff/ocr_goog.txt

I tried running the script on your body and couldn't reproduce the issue.

[Arikips]

Same for me. I think it's because of CAPTCHa or VPN

Did you find any solution?

This happens inconsistently to me. Here's the body: https://osmarks.net/stuff/ocr_goog.txt

I uploaded my script, you can give it a try if you want

[andreamagazzini]

I uploaded my script, you can give it a try if you want

@Arikips I only see the solution in python, any idea about how to fix it in javascript?

[dimdenGD]

You need to provide body of errored request for any fixes to be done. I checked their code and it does exact same thing as this library except it should be even more error prone because it doesn't account for second data structure

[andreamagazzini]

You need to provide body of errored request for any fixes to be done. I checked their code and it does exact same thing as this library except it should be even more error prone because it doesn't account for second data structure

I was also checking and I noticed it does the same.
I tried also to call the post from scratch but probably I miss something as I always get as response a page where to download google lens but no AF_initCallback data, which are the parameters that are needed in order for the post to be successful?

[dimdenGD]

I dont know specific parameters, try using same ones as used in code