Parsing tables, cells with Html agility in C#

Question

I need to parse Html code. More specifically, parse each cell of every rows in all tables. Each row represent a single object and each cell represent different properties.

Answer 1

What I had meant in my comment was that you're doing in code (the nested loops) what having the right XPath can do for you. Using LINQ-to-XML can make this even more simpler to write. But now that we see how you want your XML file formatted, we can offer our own answers. I'd write the ParseHtml() method like so:

public void ParseHtml()
{
    var htmlDoc = new HtmlDocument();
    htmlDoc.LoadHtml(htmlCode);
    var cells = htmlDoc.DocumentNode
                                    // use the right XPath rather than looping manually
                       .SelectNodes(@"//tr/tr/td[@class='statBox']")
                       .Select(node => node.InnerText.Trim())
                       .ToList();
    var elementNames = new[] { "Name", "Team", "Pos", "GP", "G", "A", "PlusMinus", "PIM", "PP", "SH", "GW", "OT", "Shots", "ShotPctg", "TOIPerGame", "ShiftsPerGame", "FOWinPctg", "UnknownField" };
    var xmlDoc =
        new XElement("Stats", new XAttribute("Date", DateTime.Now.ToShortDateString()),
            new XElement("Player", new XAttribute("Rank", cells.First()),
                // generate the elements based on the parsed cells
                cells.Skip(1)
                     .Zip(elementNames, (Value, Name) => new XElement(Name, Value))
                     .Where(element => !String.IsNullOrEmpty(element.Value))
            )
        );

    // save to your file
    xmlDoc.Save(filepath);
}

Produces the output:

  
    Sidney Crosby
    PIT
    C
    39
    32
    33
    20
    29
    10
    1
    3
    0
    154
    20.8
    21:54
    22.6
    55.7