Reputation: 102
I'm trying to get data from this website
I want to get: Level, Vocation and Name from the table. They are located directly in tr class -> td. How can I get those informations out? This is how data looks like:
<table width="100%" class="tabi">
<tr>
<td colspan=7>
Characters
</td>
</tr>
<tr>
<td height='30' style='background-color:#9f8f6d;'>
<a href=?page=whoisonline&ord=name&sort=DESC&id=1>↑Name</a>
</td>
<td width='240' style='background-color:#9f8f6d;'>
<a href=?page=whoisonline&ord=voc&sort=DESC&id=1>Vocation</a>
</td>
<td width='120' style='background-color:#9f8f6d;'>
<a href=?page=whoisonline&ord=lvl&sort=DESC&id=1>Level</a>
</td>
</tr>
<tr class='hover'>
<td>
<a href='?page=character&name=Abe' class='menulink_hs'>Abe</a>
</td>
<td>
Elder Druid
</td>
<td>
19
</td>
</tr>
Right now I'm stuck on getting this data out of tds using Nodes, with bad results. My htmlNodes is either NULL or it gives more than one Node(that I cant actually get out of it for some reason). What might be good solution to this?
My code:
var html = @"https://tibiantis.online/?page=whoisonline";
HtmlWeb web = new HtmlWeb();
var htmlDoc = web.Load(html);
HtmlNode htmlNodes = htmlDoc.DocumentNode.SelectSingleNode("/html/body/div[2]/table/tbody/tr[1]/td[3]/div[2]/div[2]/table/tbody/tr[3]");
foreach (var node in htmlNodes)
{
foreach (var cell in htmlNodes.SelectNodes(".//td"))
{
listBox1.Items.Add(cell.InnerText);
}
}
**I'm stuck with this .SelectNodes thing which no metter what gives me either null or too many Nodes. I tried many combinations both with .SelectSingleNode and .SelectNode **
Second thing is that I've got no clue how to get number of items that I will receive.
I was looking for the anwser on stack and google with some results, but noone of them was close to my situation
Upvotes: 1
Views: 438
Reputation: 2371
Try with this:
public class Person
{
public string Name { get; set; }
public string Vocation { get; set; }
public int Level { get; set; }
public static List<Person> LoadPersons(HtmlAgilityPack.HtmlDocument doc)
{
var persons = new List<Person>();
var rowsNodes = doc.DocumentNode.SelectNodes("//table//tr[contains(@class, 'hover')]");
if (rowsNodes == null)
{
return persons;
}
foreach (var rowNode in rowsNodes)
{
var cells = rowNode.SelectNodes(".//td");
if (cells != null && cells.Count >= 3)
{
var name = cells[0].InnerText;
var vocation = cells[1].InnerText;
var levelText = cells[2].InnerText;
if (int.TryParse(levelText, out int level))
{
persons.Add(new Person
{
Name = name,
Vocation = vocation,
Level = level
});
}
}
}
return persons;
}
}
This class represent a person (a row in the table) and include a method to scrap the table. When you make scraping you must try to be a bit general because putting all tags in the query makes the query to fail with a bit HTML change.
I simply search in the document (//) a table and, inside a table (// because maybe some browsers add tbody or not automatically), select all rows (tr) with the "hover" class (your persons).
Iterate each row getting the 3 cells texts. The last one (the level), convert to integer. And then, create the person.
Now, you can create a class to define each item in your list. I almost always create a class to work with the class when I get an item from the ListBox (get selected item as PersonItem and do any work with it...):
public class PersonItem
{
public PersonItem(Person person)
{
this.Person = person;
}
public Person Person { get; }
public override string ToString()
{
return $"{this.Person.Name} ({this.Person.Level})";
}
}
It's simply a wrapper around Person. Override ToString with the text to show in the ListBox.
Test it:
var web = new HtmlWeb();
var doc = web.Load("https://tibiantis.online/?page=whoisonline");
var persons = Person.LoadPersons(doc);
foreach (var person in persons)
{
var item = new PersonItem(person);
listBox1.Items.Add(item);
}
// In any moment, you may do things like this:
var personItem = listBox1.SelectedItem as PersonItem;
if (personItem != null)
{
var person = personItem.Person;
// ...
}
Upvotes: 1