It scrappes N review pages for an specific dealer and evaluates them
1. Select all rating values equal to 5
2. Sum all detailed ratings
3. Calculate mean of that sum
4. Filter candidates that would be a fake review using 5 as a threshold
docker compose up
It is possible to configure Scrapper using these envs:
| Name | Description | Default |
|---|---|---|
| DealerRater__PageCount | Total pages to fetch | 5 |
| DealerRater__BaseUrl | Base url | https://www.dealerrater.com/dealer/ |
| DealerRater__Dealer | Dealer url | McKaig-Chevrolet-Buick-A-Dealer-For-The-People-dealer-reviews-23685 |
| DealerRater__Rank | Rank Quantity | 3 |
| DealerRater__Threshold | Rank Threshold | 5 |
It is possible to run unit tests using Docker
Linux
docker run --rm -v $(pwd):/app -w /app/tests/Scrapper.Application.Tests mcr.microsoft.com/dotnet/sdk:6.0 dotnet test --logger:trx
Windows Powershell
docker run --rm -v ${pwd}:/app -w /app/tests/Scrapper.Application.Tests mcr.microsoft.com/dotnet/sdk:6.0 dotnet test --logger:trx
