private static void ParseResults(AssociationRule association, string searchTerm, HtmlDocument doc)
{
var nodes = doc.DocumentNode.SelectNodes("//div[@class = 'SearchResults-documents']/div/div");
if (nodes != null)
{
Console.WriteLine($"{association.Rule} - {searchTerm}: {nodes.Count()} results");
foreach (var node in nodes)
{
var nodeTitle = node.SelectSingleNode("./h2/a");
var nodePublicacao = node.SelectSingleNode("./div[@class = 'BaseSnippetWrapper-publisher']/span[@class = 'BaseSnippetWrapper-highlight-date']");
var nodeBody = node.SelectSingleNode("./div[@class = 'BaseSnippetWrapper-body']");
var result = new SearchResult
{
Title = nodeTitle.InnerText,
Url = nodeTitle.GetAttributeValue("href", string.Empty),
DataPublicacao = nodePublicacao != null?nodePublicacao.InnerText.Split(':')[1].Trim() : string.Empty,
Abstract = nodeBody.InnerText,
UsedTerm = searchTerm
};
association.Results.Add(result);
}
}
//else if (doc.DocumentNode.SelectSingleNode("//span[contains(text(), 'não encontrou nenhum documento')]") == null)
//{
// throw new Exception("Erro inesperado, possível bloqueio de IP.");
//}
}
private static AssociationRule ParseAssociationRule(string[] fields, int timeStamp)
{
var ruleItens = fields[0].Split('>');
var rule = new AssociationRule
{
Principal = Regex.Replace(ruleItens[1], @"\D", string.Empty),
Members = ruleItens[0].Split(',').Select(x => Regex.Replace(x, @"\D", string.Empty)).ToArray(),
Rule = fields[0],
Support = double.Parse(fields[1], CultureInfo.InvariantCulture),
Confidence = double.Parse(fields[2], CultureInfo.InvariantCulture),
Lift = double.Parse(fields[3], CultureInfo.InvariantCulture),
Count = int.Parse(fields[4], CultureInfo.InvariantCulture),
TimeStamp = timeStamp
};
return(rule);
}
