ddg search error when using fastapi's pip version

I've been trying to get images from duckduckgo using

from duckduckgo_search import DDGS
from fastcore.all import *

ddgs = DDGS()
def search_images(term, max_images=30):
    print(f"Searching for '{term}'")
    keywords = 'stop sign'
    return L(ddgs.images((keywords, term, max_results=max_images)).itemgot('image'))

but it gives an error:  Cell In[30], line 8
    return L(ddgs.images((keywords, term, max_results=max_images)).itemgot('image'))
                                          ^
SyntaxError: invalid syntax. Maybe you meant '==' or ':=' instead of '='?

did some looking up and seems there's some issue with the pip version of fastbook-0.0.16.tar.gz where the file: init.py overwrites and breaks the working version of search_images_ddg
Is there any way to fix this?
If not, is there another search engine API that's good for getting images?
I'm making a basic image recognition thing with fastapi
in jupyter notebook

The images() function of DDGS no longer supports the max_results argument, so we have to do it in a separate step after creating the generator with the images() function.
A generator has no method itemgot(), so we also need to get the urls in another step.

Here is a drop-in replacement for the function search_images based on the current generator implementation of duckduckgo_search.

from duckduckgo_search import DDGS
from fastcore.all import *

def search_images(term, max_images=30):
    print(f"Searching for '{term}'")
    with DDGS() as ddgs:
        # generator which yields dicts with:
        # {'title','image','thumbnail','url','height','width','source'}
        search_results = ddgs.images(keywords=term)       
        # grap number of max_images urls
        image_urls = [next(search_results).get("image") for _ in range(max_images)]
        # convert to L (functionally extended list class from fastai)
        return L(image_urls)

# example usage:
urls = search_images("dog images", max_images=10)
print(urls[0])

If you have problems with the pip version of fastbook, maybe just install the fastai and duckduckgo_search packages directly.

For installing from a Jupyter Notebook:

%pip install -U fastai duckduckgo_search

# if you use Anaconda locally it is better to use 
# conda or mamba (mambaforge) instead of pip:
!mamba install fastai 
%pip install -U duckduckgo_search 

# or
%conda install fastai
%pip install -U duckduckgo_search

I had the same issue. Turns out the function used in the tutorial is outdated, and the updated duckduckgo image() function returns a generator instead of a list, so just changing the function call doesn't solve the problem. What you can do instead is to work around the new function. Try the following:

from duckduckgo_search import DDGS
from fastcore.all import *

def search_images(term, max_images=30):
    print(f"Searching for '{term}'")
    with DDGS() as ddgs:
        ddgs_images_gen = ddgs.images(term)
        count = 0
        ddgs_images_list = []
        while count < max_images:
            image = next(ddgs_images_gen)
            ddgs_images_list.append(image.get('image'))
            count = count+1
        return ddgs_images_list

All the code does is add the desired number of images to a list, then return that list.

urls = search_images('stop sign', max_images=1)

urls[0]

This should now give the proper result. Hope this helps!