ubahnverleih / WoBike

Documentation of Bike Sharing APIs 🚴🛴🛵
952 stars 131 forks source link

Accessing Lime API by email #253

Closed sean-mooney-jcbs closed 1 year ago

sean-mooney-jcbs commented 1 year ago

I'm able to access the lime API with a phone number but not with an email address. The command I'm using is: requests.post("https://web-production.lime.bike/api/rider/v2/onboarding/magic-link/email=email_address@gmail.com&user_agreement_country_code=US&user_agreement_version=4") This is giving a 404 error (same with replacing '@' with '%40') My understanding from the documentation is that I should be emailed a token which can then be used to access the API

Can someone explain what I'm doing wrong please?

Edit: I've also tried using the pycurl package but again, I'm not sure why it isn't working (403 error):

c = pycurl.Curl()
c.setopt(c.URL, 'https://web-production.lime.bike/api/rider/v2/onboarding/magic-link/')
data = {'Content-Type': 'application/x-www-form-urlencoded; charset=utf-8',
        'email': 'email_address%40gmail.com&user_agreement_country_code=US&user_agreement_version=4'}
pf = urlencode(data)

c.setopt(c.POSTFIELDS, pf)
c.perform()
print(f'Response Code: {c.getinfo(c.RESPONSE_CODE)}')
c.close()
BastelPichi commented 1 year ago

Hello, You did some things wrong. You added the data as params, like in a GET request, but you also forget an ?. However you're supposed to submit the data as body. Also you didn't set the correct headers, but this doesn't matter here. Heres some functioning code:

import requests

data = {
  "email": "nospam@pichisdns.com",
  "user_agreement_country_code": "US",
  "user_agreement_version": 4
}

r = requests.post("https://web-production.lime.bike/api/rider/v2/onboarding/magic-link", data=data)

print(r.json())

# returns {'errors': [{'status': 'send_magic_link_user_unavailable', 'title': 'Hmm...', 'detail': "We couldn't find an existing Lime account with this email. Please check the email entered or sign up for a new account using another method.", 'data': {}}]}
# If successful, returns {}
sean-mooney-jcbs commented 1 year ago

Thanks for your help. While I think I see where I went wrong I must still be missing something because I can't use the code I'm emailed to access the API.

From what I can see login should work the same way but the data dictionary should only contain the code I'm emailed. So when I get the email at the bottom I should do the following?

data = {"magic_link_token": "tak51fgzbbmc2gqkjKASZXnL"}
r = requests.post("https://web-production.lime.bike/api/rider/v2/onboarding/login", data=data)

However, this is giving a 422 error code with the following output: '{"errors":[{"status":"bad_argument_error","title":"Hmm...","detail":"Sorry, we can\'t process the request at the moment. Please try again later.","data":{}}]}'

Specifically its giving a bad_arguement_error but there's only one arguement which looks to me like the code I'm using is wrong but I'm copying and pasting it directly from the email so I'm not sure what's wrong here.

image

BastelPichi commented 1 year ago

Thanks for your help. While I think I see where I went wrong I must still be missing something because I can't use the code I'm emailed to access the API.

From what I can see login should work the same way but the data dictionary should only contain the code I'm emailed. So when I get the email at the bottom I should do the following?

data = {"magic_link_token": "tak51fgzbbmc2gqkjKASZXnL"}
r = requests.post("https://web-production.lime.bike/api/rider/v2/onboarding/login", data=data)

However, this is giving a 422 error code with the following output: '{"errors":[{"status":"bad_argument_error","title":"Hmm...","detail":"Sorry, we can\'t process the request at the moment. Please try again later.","data":{}}]}'

Specifically its giving a bad_arguement_error but there's only one arguement which looks to me like the code I'm using is wrong but I'm copying and pasting it directly from the email so I'm not sure what's wrong here.

image

Can help you tommorrow.

sean-mooney-jcbs commented 1 year ago

If you have the time, I'd really appreciate it (if not though, no worries). I've been looking at a few bike share schemes recently and this is the only one I've been having issues with.

BastelPichi commented 1 year ago

Sorry, completely forgot it. Will fix this later (pinky promise)

sean-mooney-jcbs commented 1 year ago

That's no problem. As you can probably tell I'm not too time pressured.

BastelPichi commented 1 year ago

Yeah... Now I forgot it, butt i will hopefully get this tmorrow lol

BastelPichi commented 1 year ago

Now I'm here. Gimme 20min...

BastelPichi commented 1 year ago

So, heres what I found out:

sean-mooney-jcbs commented 1 year ago

I'm finding that not receiving anything makes debugging this very difficult as I don't know where to start and I find that that effectively makes it a black box (to me at least).

BastelPichi commented 1 year ago

I'm finding that not receiving anything makes debugging this very difficult as I don't know where to start and I find that that effectively makes it a black box (to me at least).

Yep. Normally, if theres explicit SSL pinning, it doesnt work at all, and you can see an Connection aborted error in your sniffing software. Ill try with some older version. However, from previous posts it apperas that at least beack then sniffing on IOS is way easier. I do not have any IOS phone, or even PC, but if you do, would be cool if you could try out.

BastelPichi commented 1 year ago

Got it. The trick is to setup your proxy after the initial launch of the app.

Here's a working script. You might need the x-device-token for later requests.

import uuid
import requests

device_token = str(uuid.uuid4())
print("UUID:", device_token, "\n")

headers = {
    "x-device-token": device_token
}

data = {"magic_link_token": "ASJybRiOdiuZKYNhthSQyMKH"}
r = requests.post("https://web-production.lime.bike/api/rider/v2/onboarding/login", data=data, headers=headers)

print(r.json())
sean-mooney-jcbs commented 1 year ago

Works perfect. Thanks a million. I really appreciate your help. I'll keep this in mind for future APIs I'll be using.