PickAll 1.3.0

PickAll

.NET agile and extensible web searching API. Built with AngleSharp.

Philosophy

PickAll is primarily designed to collect a limited amount of results (possibly the more relavant) from different sources and process these in a chain of steps. Results are essentially URLs and descriptions, but more data can be handled.

Documentation

Documentation is available in the project wiki.

Targets

• .NET Standard 2.0
• .NET Core 3.1
• .NET 5.0

Install via NuGet

$ dotnet add package PickAll --version 1.3.0
  Determining projects to restore...
  ...

Build and sample

# clone the repository
$ git clone https://github.com/gsscoder/pickall.git

# build the package
$ cd pickall/src/PickAll
$ dotnet build -c release

# execute sample
$ cd pickall/samples/PickAll.Sample
$ dotnet build -c release
$ cd ../../artifacts/PickAll.Sample/Release/netcoreapp3.0/PickAll.Sample
./PickAll.Sample "Steve Jobs" -e bing:duckduckgo
Searching 'Steve Jobs' ...
[0] Bing: "Steve Jobs - Wikipedia": "https://it.wikipedia.org/wiki/Steve_Jobs"
[0] DuckDuckGo: "Steve Jobs - Wikipedia": "https://en.wikipedia.org/wiki/Steve_Jobs"
[1] DuckDuckGo: "Steve Jobs - Apple, Family & Death - Biography": "https://www.biography.com/business-figure/steve-jobs"
[2] Bing: "CC-BY-SA licenza": "http://creativecommons.org/licenses/by-sa/3.0/"
[2] DuckDuckGo: "Steve Jobs - IMDb": "https://www.imdb.com/name/nm0423418/"
[3] Bing: "Biografia di Steve Jobs - Biografieonline": "https://biografieonline.it/biografia.htm?BioID=1560&biografia=Steve+Jobs"

Test

# change to tests directory
$ cd pickall/tests/PickAll.Specs

# build with debug configuration
$ dotnet build -c debug
...

# execute tests
$ dotnet test
...

At a glance

CSharp:

using PickAll;

var context = new SearchContext()
    .WithEvents()
    .With<Google>() // search on google.com
    .With<Yahoo>() // search on yahoo.com
    .With<Uniqueness>() // remove duplicates
    .With<Order>() // prioritize results
    // match Levenshtein distance with maximum of 15
    .With<FuzzyMatch>(new FuzzyMatchSettings { Text = "mechanics", MaximumDistance = 15 });
    // repeat a search using more frequent words of previous results
    .With<Improve>(new ImproveSettings { WordCount = 2, NoiseLength = 3 })
    // scrape result pages and extract all text
    .With<Textify>(new TextifySettings { IncludeTitle = true, NoiseLength = 3 });
// attach events
context.ResultCreated += (sender, e) => Console.WriteLine($"Result created from {e.Result.Originator}");
// execute services (order of addition)
var results = await context.SearchAsync("quantum physics");
// do anything you need with LINQ
var scientific = results.Where(result => result.Url.Contains("wikipedia"));
foreach (var result in scientific) {
    Console.WriteLine($"{result.Url} {result.Description}");
}

FSharp:

let context = new SearchContext(typeof<Google>,
                                typeof<DuckDuckGo>,
                                typeof<Yahoo>)
let results = context.SearchAsync("quantum physics")
              |> Async.AwaitTask
              |> Async.RunSynchronously

results |> Seq.iter (fun x -> printfn "%s %s" x.Url x.Description)

Libraries

• AngleSharp
• AngleSharp.Io
• SharpX
• CommandLineParser
• xUnit.net
• FluentAssertions
• WaffleGenerator
• Bogus

Tools

• Paket

Icon

• Search Engine icon designed by Vectors Market from The Noun Project.