Scraping einer Online-Film-Website
This article was translated from English: Does it need improvement?
TranslatedView the article in English
Beginnen wir mit einem weiteren Beispiel von einer echten Website. Wir werden eine Film-Website scrapen.
Fügen wir eine neue Klasse hinzu und nennen sie "MovieScraper":
Werfen wir nun einen Blick auf die Website, die wir scrapen werden:
Dies ist ein Teil der HTML-Homepage, die wir auf der Website sehen:
<div id="movie-featured" class="movies-list movies-list-full tab-pane in fade active">
<div data-movie-id="20746" class="ml-item">
<a href="https://website.com/film/king-arthur-legend-of-the-sword-20746/">
<span class="mli-quality">CAM</span>
<img data-original="https://img.gocdn.online/2017/05/16/poster/2116d6719c710eabe83b377463230fbe-king-arthur-legend-of-the-sword.jpg"
class="lazy thumb mli-thumb" alt="King Arthur: Legend of the Sword"
src="https://img.gocdn.online/2017/05/16/poster/2116d6719c710eabe83b377463230fbe-king-arthur-legend-of-the-sword.jpg"
style="display: inline-block;">
<span class="mli-info"><h2>King Arthur: Legend of the Sword</h2></span>
</a>
</div>
<div data-movie-id="20724" class="ml-item">
<a href="https://website.com/film/snatched-20724/" >
<span class="mli-quality">CAM</span>
<img data-original="https://img.gocdn.online/2017/05/16/poster/5ef66403dc331009bdb5aa37cfe819ba-snatched.jpg"
class="lazy thumb mli-thumb" alt="Snatched"
src="https://img.gocdn.online/2017/05/16/poster/5ef66403dc331009bdb5aa37cfe819ba-snatched.jpg"
style="display: inline-block;">
<span class="mli-info"><h2>Snatched</h2></span>
</a>
</div>
</div><div id="movie-featured" class="movies-list movies-list-full tab-pane in fade active">
<div data-movie-id="20746" class="ml-item">
<a href="https://website.com/film/king-arthur-legend-of-the-sword-20746/">
<span class="mli-quality">CAM</span>
<img data-original="https://img.gocdn.online/2017/05/16/poster/2116d6719c710eabe83b377463230fbe-king-arthur-legend-of-the-sword.jpg"
class="lazy thumb mli-thumb" alt="King Arthur: Legend of the Sword"
src="https://img.gocdn.online/2017/05/16/poster/2116d6719c710eabe83b377463230fbe-king-arthur-legend-of-the-sword.jpg"
style="display: inline-block;">
<span class="mli-info"><h2>King Arthur: Legend of the Sword</h2></span>
</a>
</div>
<div data-movie-id="20724" class="ml-item">
<a href="https://website.com/film/snatched-20724/" >
<span class="mli-quality">CAM</span>
<img data-original="https://img.gocdn.online/2017/05/16/poster/5ef66403dc331009bdb5aa37cfe819ba-snatched.jpg"
class="lazy thumb mli-thumb" alt="Snatched"
src="https://img.gocdn.online/2017/05/16/poster/5ef66403dc331009bdb5aa37cfe819ba-snatched.jpg"
style="display: inline-block;">
<span class="mli-info"><h2>Snatched</h2></span>
</a>
</div>
</div>HTML
Wie wir sehen können, haben wir eine Film-ID, einen Titel und einen Link zur Detailseite.
Beginnen wir mit dem Scrapen dieses Datensatzes:
public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("www.website.com", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var MovieId = Divs.GetAttribute("data-movie-id");
var link = Divs.Css("a")[0];
var MovieTitle = link.TextContentClean;
Scrape(new ScrapedData() { { "MovieId", MovieId }, { "MovieTitle", MovieTitle } }, "Movie.Jsonl");
}
}
}
}public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("www.website.com", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var MovieId = Divs.GetAttribute("data-movie-id");
var link = Divs.Css("a")[0];
var MovieTitle = link.TextContentClean;
Scrape(new ScrapedData() { { "MovieId", MovieId }, { "MovieTitle", MovieTitle } }, "Movie.Jsonl");
}
}
}
}Public Class MovieScraper
Inherits WebScraper
Public Overrides Sub Init()
License.LicenseKey = "LicenseKey"
Me.LoggingLevel = WebScraper.LogLevel.All
Me.WorkingDirectory = AppSetting.GetAppRoot() & "\MovieSample\Output\"
Me.Request("www.website.com", AddressOf Parse)
End Sub
Public Overrides Sub Parse(ByVal response As Response)
For Each Divs In response.Css("#movie-featured > div")
If Divs.Attributes ("class") <> "clearfix" Then
Dim MovieId = Divs.GetAttribute("data-movie-id")
Dim link = Divs.Css("a")(0)
Dim MovieTitle = link.TextContentClean
Scrape(New ScrapedData() From {
{ "MovieId", MovieId },
{ "MovieTitle", MovieTitle }
},
"Movie.Jsonl")
End If
Next Divs
End Sub
End ClassVB C#
Was ist neu an diesem Code?
Die Eigenschaft Arbeitsverzeichnis wird verwendet, um das Hauptarbeitsverzeichnis für alle gescrapten Daten und die zugehörigen Dateien festzulegen.
Lasst uns mehr tun.
Was, wenn wir typisierte Objekte erstellen müssen, die gescrapte Daten in formatierten Objekten enthalten?
Implementieren wir eine Filmklasse, die unsere formatierten Daten aufnehmen wird:
public class Movie
{
public int Id { get; set; }
public string Title { get; set; }
public string URL { get; set; }
}public class Movie
{
public int Id { get; set; }
public string Title { get; set; }
public string URL { get; set; }
}Public Class Movie
Public Property Id() As Integer
Public Property Title() As String
Public Property URL() As String
End ClassVB C#
Jetzt werden wir unseren Code aktualisieren:
public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("https://website.com/", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var movie = new Movie();
movie.Id = Convert.ToInt32( Divs.GetAttribute("data-movie-id"));
var link = Divs.Css("a")[0];
movie.Title = link.TextContentClean;
movie.URL = link.Attributes ["href"];
Scrape(movie, "Movie.Jsonl");
}
}
}
}public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("https://website.com/", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var movie = new Movie();
movie.Id = Convert.ToInt32( Divs.GetAttribute("data-movie-id"));
var link = Divs.Css("a")[0];
movie.Title = link.TextContentClean;
movie.URL = link.Attributes ["href"];
Scrape(movie, "Movie.Jsonl");
}
}
}
}Public Class MovieScraper
Inherits WebScraper
Public Overrides Sub Init()
License.LicenseKey = "LicenseKey"
Me.LoggingLevel = WebScraper.LogLevel.All
Me.WorkingDirectory = AppSetting.GetAppRoot() & "\MovieSample\Output\"
Me.Request("https://website.com/", AddressOf Parse)
End Sub
Public Overrides Sub Parse(ByVal response As Response)
For Each Divs In response.Css("#movie-featured > div")
If Divs.Attributes ("class") <> "clearfix" Then
Dim movie As New Movie()
movie.Id = Convert.ToInt32(Divs.GetAttribute("data-movie-id"))
Dim link = Divs.Css("a")(0)
movie.Title = link.TextContentClean
movie.URL = link.Attributes ("href")
Scrape(movie, "Movie.Jsonl")
End If
Next Divs
End Sub
End ClassVB C#
Was gibt es Neues?
Wir implementieren Movie Class, um unsere gescrapten Daten zu speichern
- Wir übergeben Filmobjekte an die Scrape-Methode und sie versteht unser Format und speichert in einem definierten Format, wie wir hier sehen können:
Beginnen wir mit dem Scannen einer detaillierteren Seite.
Die Filmseite sieht wie folgt aus:
<div class="mvi-content">
<div class="thumb mvic-thumb"
style="background-image: url(https://img.gocdn.online/2017/04/28/poster/5a08e94ba02118f22dc30f298c603210-guardians-of-the-galaxy-vol-2.jpg);"></div>
<div class="mvic-desc">
<h3>Guardians of the Galaxy Vol. 2</h3>
<div class="desc">
Set to the backdrop of Awesome Mixtape #2, Marvel's Guardians of the Galaxy Vol. 2 continues the team's adventures as they travel throughout the cosmos to help Peter Quill learn more about his true parentage.
</div>
<div class="mvic-info">
<div class="mvici-left">
<p>
<strong>Genre: </strong>
<a href="https://Domain/genre/action/" title="Action">Action</a>,
<a href="https://Domain/genre/adventure/" title="Adventure">Adventure</a>,
<a href="https://Domain/genre/sci-fi/" title="Sci-Fi">Sci-Fi</a>
</p>
<p>
<strong>Actor: </strong>
<a target="_blank" href="https://Domain/actor/chris-pratt" title="Chris Pratt">Chris Pratt</a>,
<a target="_blank" href="https://Domain/actor/-zoe-saldana" title="Zoe Saldana">Zoe Saldana</a>,
<a target="_blank" href="https://Domain/actor/-dave-bautista-" title="Dave Bautista">Dave Bautista</a>
</p>
<p>
<strong>Director: </strong>
<a href="#" title="James Gunn">James Gunn</a>
</p>
<p>
<strong>Country: </strong>
<a href="https://Domain/country/us" title="United States">United States</a>
</p>
</div>
<div class="mvici-right">
<p><strong>Duration:</strong> 136 min</p>
<p><strong>Quality:</strong> <span class="quality">CAM</span></p>
<p><strong>Release:</strong> 2017</p>
<p><strong>IMDb:</strong> 8.3</p>
</div>
<div class="clearfix"></div>
</div>
<div class="clearfix"></div>
</div>
<div class="clearfix"></div>
</div><div class="mvi-content">
<div class="thumb mvic-thumb"
style="background-image: url(https://img.gocdn.online/2017/04/28/poster/5a08e94ba02118f22dc30f298c603210-guardians-of-the-galaxy-vol-2.jpg);"></div>
<div class="mvic-desc">
<h3>Guardians of the Galaxy Vol. 2</h3>
<div class="desc">
Set to the backdrop of Awesome Mixtape #2, Marvel's Guardians of the Galaxy Vol. 2 continues the team's adventures as they travel throughout the cosmos to help Peter Quill learn more about his true parentage.
</div>
<div class="mvic-info">
<div class="mvici-left">
<p>
<strong>Genre: </strong>
<a href="https://Domain/genre/action/" title="Action">Action</a>,
<a href="https://Domain/genre/adventure/" title="Adventure">Adventure</a>,
<a href="https://Domain/genre/sci-fi/" title="Sci-Fi">Sci-Fi</a>
</p>
<p>
<strong>Actor: </strong>
<a target="_blank" href="https://Domain/actor/chris-pratt" title="Chris Pratt">Chris Pratt</a>,
<a target="_blank" href="https://Domain/actor/-zoe-saldana" title="Zoe Saldana">Zoe Saldana</a>,
<a target="_blank" href="https://Domain/actor/-dave-bautista-" title="Dave Bautista">Dave Bautista</a>
</p>
<p>
<strong>Director: </strong>
<a href="#" title="James Gunn">James Gunn</a>
</p>
<p>
<strong>Country: </strong>
<a href="https://Domain/country/us" title="United States">United States</a>
</p>
</div>
<div class="mvici-right">
<p><strong>Duration:</strong> 136 min</p>
<p><strong>Quality:</strong> <span class="quality">CAM</span></p>
<p><strong>Release:</strong> 2017</p>
<p><strong>IMDb:</strong> 8.3</p>
</div>
<div class="clearfix"></div>
</div>
<div class="clearfix"></div>
</div>
<div class="clearfix"></div>
</div>HTML
Wir können unsere Filmklasse um neue Eigenschaften erweitern(Beschreibung, Genre, Darsteller, Regisseur, Land, Dauer, IMDB Score) aber wir verwenden(Beschreibung, Genre, Akteur) nur für unsere Stichprobe.
public class Movie
{
public int Id { get; set; }
public string Title { get; set; }
public string URL { get; set; }
public string Description { get; set; }
public List<string> Genre { get; set; }
public List<string> Actor { get; set; }
}public class Movie
{
public int Id { get; set; }
public string Title { get; set; }
public string URL { get; set; }
public string Description { get; set; }
public List<string> Genre { get; set; }
public List<string> Actor { get; set; }
}Public Class Movie
Public Property Id() As Integer
Public Property Title() As String
Public Property URL() As String
Public Property Description() As String
Public Property Genre() As List(Of String)
Public Property Actor() As List(Of String)
End ClassVB C#
Jetzt navigieren wir zu der Seite Detailliert, um sie zu scrapen.
IronWebscraper ermöglicht es Ihnen, die Scrape-Funktion zu erweitern, um verschiedene Arten von Seitenformaten zu scrapen
Wie wir hier sehen können:
public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("https://domain/", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var movie = new Movie();
movie.Id = Convert.ToInt32( Divs.GetAttribute("data-movie-id"));
var link = Divs.Css("a")[0];
movie.Title = link.TextContentClean;
movie.URL = link.Attributes ["href"];
this.Request(movie.URL, ParseDetails, new MetaData() { { "movie", movie } });// to scrap Detailed Page
}
}
}
public void ParseDetails(Response response)
{
var movie = response.MetaData.Get<Movie>("movie");
var Div = response.Css("div.mvic-desc")[0];
movie.Description = Div.Css("div.desc")[0].TextContentClean;
foreach(var Genre in Div.Css("div > p > a"))
{
movie.Genre.Add(Genre.TextContentClean);
}
foreach (var Actor in Div.Css("div > p:nth-child(2) > a"))
{
movie.Actor.Add(Actor.TextContentClean);
}
Scrape(movie, "Movie.Jsonl");
}
}public class MovieScraper : WebScraper
{
public override void Init()
{
License.LicenseKey = "LicenseKey";
this.LoggingLevel = WebScraper.LogLevel.All;
this.WorkingDirectory = AppSetting.GetAppRoot() + @"\MovieSample\Output\";
this.Request("https://domain/", Parse);
}
public override void Parse(Response response)
{
foreach (var Divs in response.Css("#movie-featured > div"))
{
if (Divs.Attributes ["class"] != "clearfix")
{
var movie = new Movie();
movie.Id = Convert.ToInt32( Divs.GetAttribute("data-movie-id"));
var link = Divs.Css("a")[0];
movie.Title = link.TextContentClean;
movie.URL = link.Attributes ["href"];
this.Request(movie.URL, ParseDetails, new MetaData() { { "movie", movie } });// to scrap Detailed Page
}
}
}
public void ParseDetails(Response response)
{
var movie = response.MetaData.Get<Movie>("movie");
var Div = response.Css("div.mvic-desc")[0];
movie.Description = Div.Css("div.desc")[0].TextContentClean;
foreach(var Genre in Div.Css("div > p > a"))
{
movie.Genre.Add(Genre.TextContentClean);
}
foreach (var Actor in Div.Css("div > p:nth-child(2) > a"))
{
movie.Actor.Add(Actor.TextContentClean);
}
Scrape(movie, "Movie.Jsonl");
}
}Public Class MovieScraper
Inherits WebScraper
Public Overrides Sub Init()
License.LicenseKey = "LicenseKey"
Me.LoggingLevel = WebScraper.LogLevel.All
Me.WorkingDirectory = AppSetting.GetAppRoot() & "\MovieSample\Output\"
Me.Request("https://domain/", AddressOf Parse)
End Sub
Public Overrides Sub Parse(ByVal response As Response)
For Each Divs In response.Css("#movie-featured > div")
If Divs.Attributes ("class") <> "clearfix" Then
Dim movie As New Movie()
movie.Id = Convert.ToInt32(Divs.GetAttribute("data-movie-id"))
Dim link = Divs.Css("a")(0)
movie.Title = link.TextContentClean
movie.URL = link.Attributes ("href")
Me.Request(movie.URL, AddressOf ParseDetails, New MetaData() From {
{ "movie", movie }
}) ' to scrap Detailed Page
End If
Next Divs
End Sub
Public Sub ParseDetails(ByVal response As Response)
Dim movie = response.MetaData.Get(Of Movie)("movie")
Dim Div = response.Css("div.mvic-desc")(0)
movie.Description = Div.Css("div.desc")(0).TextContentClean
For Each Genre In Div.Css("div > p > a")
movie.Genre.Add(Genre.TextContentClean)
Next Genre
For Each Actor In Div.Css("div > p:nth-child(2) > a")
movie.Actor.Add(Actor.TextContentClean)
Next Actor
Scrape(movie, "Movie.Jsonl")
End Sub
End ClassVB C#
Was gibt es Neues?
Wir können Scrape-Funktionen hinzufügen(ParseDetails) zum Scrapen von Detailseiten
Wir haben die Scrape-Funktion, die unsere Datei erzeugt, in die neue Funktion
Wir haben die IronWebScraper-Funktion(MetaDaten) um unser Filmobjekt an die neue Scrape-Funktion zu übergeben
- Wir haben die Seite ausgewertet und unsere Filmobjektdaten in einer Datei gespeichert
