A simple service that allows a user to pass in a Lending Tree lender's url, and returns all the reviews for that lender. Built in python3 with Flask.

This service requires python3 to be installed on your machine.

1. Clone this repo and navigate to the directory
2. Create a virtual environment for packages:

   python3 -m venv venv

3. Activate your virtual environment

   . ./venv/bin/activate

4. Install the requirements

   pip install -r requirements.txt

Start the application with

python3 -m app

The server runs on localhost:5000.

Run tests with

python3 -m unittest discover ./test/ 

1. Find your desired lender's URL on Lending Tree's website. https://www.lendingtree.com/reviews/personal/ has the list of available lenders.
2. Make a GET request to localhost:5000/review/{lender_url}

Responses are JSON formatted. An example response is as follows:

[
  {
    "id": "5e416ae37f520f0001508adc",
    "productId": "42825",
    "productType": "lender",
    "title": "It was a really good experience. ",
    "text": "It was so easy to complete the application and the get the money I needed. Very helpful. Thanks so much.",
    "isRecommended": true,
    "authorId": "",
    "authorEmail": "",
    "anonymousUId": "",
    "authorName": "Vicky",
    "userLocation": "GROTTOES, VA",
    "brandId": "42825",
    "lenderId": 81638970,
    "isAuthenticated": false,
    "isVerifiedCustomer": false,
    "socialSurveyReviewId": null,
    "votesUp": 0,
    "votesDown": 0,
    "isFlagged": false,
    "comments": null,
    "primaryRating": {
      "value": 4,
      "name": "Overallrating"
    },
    "secondaryRatings": [
      {
        "name": "RateRating",
        "value": 3.0
      },
      {
        "name": "FeesAndCostsRating",
        "value": 3.0
      },
      {
        "name": "ResponsivenessRating",
        "value": 5.0
      },
      {
        "name": "CustomerServiceRating",
        "value": 5.0
      }
    ],
    "properties": [
      {
        "name": "RequestType",
        "value": "Personal Loan"
      },
      {
        "name": "IsLoanClosed",
        "value": "true"
      }
    ],
    "submissionDateTime": "2020-02-10T09:38:26.509+00:00"
  }
]

• Persistent storage would likely be the most impactful addition to this project. After a lender is looked up once, we could store the brand_id that is used by Lending Tree's API. Then, for future lookups, we would not need to scrape that value from the webpage.
• After storage is in place, a cron job could be run to keep the database of reviews and lenders up to date.
• Docker would be a welcome addition to bundle the service.
• Flask on it's own is not production-ready. Adding waitress would be a simple way to get the app ready for production.
• As the API grows, Sphinx can be added to generate documentation. As there's only one route currently, a GitHub readme will suffice.