String slices must be integers.

[rV4nxZ]

Describe the bug try: images = [i["file_url"] for i in data]

except: images = [i["file"]["url"] for i in data]

Gives the error. This is yours interpreter.

To Reproduce Steps to reproduce the behavior:

1. Use parse_image(result) on: result = await r34.search_image("neko", limit=2) # test

Expected behavior Give me only the url.

Desktop (please complete the following information):

• OS: Arch Linux
• Python version: 3.10.10 and 3.11.3

Additional context No.

[da-vinci-bot[bot]]

Hey! @rV4nxZ You'll get a response soon! UwU

Click here to make your day UmU


[sinkaroid]

I tested on 3.10.11 everything works fine until some of my testing network marked as spam thus captcha were found on r34, but it should throwing some JSONDecodeError instead String slices must be integers

Could you tell some moreinfo pls

[rV4nxZ]

hi, i have good news, i just discover that this:

async def main():
    r34 = Rule34()
    img = await r34.search(query="sexy", limit=1)
    img = json.loads(img)
    urls = [item['file_url'] for item in img if item['file_url'].startswith('https://api-cdn.rule34.xxx/images/')]
    print(urls)

asyncio.run(main())

parses the images correctly :D

[rV4nxZ]

it only works without the "gacha" tag, works with the "limit" tag

[sinkaroid]

When using gacha method, all returns only 1 object/dict, so your loop statements is no use then

[rV4nxZ]

fixed it

    img = await r34.search(query="sexy", gacha=True)

    if isinstance(img, str):
        # if img is a string, parse it as JSON
        data = json.loads(img)
        urls = [data['file_url']] if data['file_url'].startswith('https://api-cdn.') else []
    else:
        # if img is a list of image objects, extract the file urls
        data = json.loads(img)
        urls = [item['file_url'] for item in data if item['file_url'].startswith('https://api-cdn.')]

    print(urls)

now i'll try to add it to the parse_image function and use the resolve instead of data = json.loads(img) to follow your code

[rV4nxZ]

updated function:

def parse_image_r34(raw_object: dict):
    if isinstance(raw_object, str):
        data = json.loads(raw_object)
        urls = [data['file_url']] if data['file_url'].startswith('https://api-cdn.') else []
    else:
        data = json.loads(raw_object)
        urls = [item['file_url'] for item in data if item['file_url'].startswith('https://api-cdn.')]

output: ['https://api-cdn.rule34.xxx/images/6747/48766c19aac641cae347e3a4180e817a.png']

[sinkaroid]

You can always submit PR if the current flow is a bit clunky on your side

[sinkaroid]

fixed it

    img = await r34.search(query="sexy", gacha=True)

    if isinstance(img, str):
        # if img is a string, parse it as JSON
        data = json.loads(img)
        urls = [data['file_url']] if data['file_url'].startswith('https://api-cdn.') else []
    else:
        # if img is a list of image objects, extract the file urls
        data = json.loads(img)
        urls = [item['file_url'] for item in data if item['file_url'].startswith('https://api-cdn.')]

    print(urls)

now i'll try to add it to the parse_image function and use the resolve instead of data = json.loads(img) to follow your code

What is the point reinventing urls with loop, while img is only single object, there is still make sense if you re-validating them in normal search not gacha

[rV4nxZ]

well for me it works for what i need