// 1. Load each file with links in memory
// 2. Filter the links
// 3. Write the filtered links to the same file (overwrite)
// This methid works only for raw web links. If it is used twice in a row, then results will be empty.
public static void FilterLinks(string pathDataDir)
{
Console.WriteLine("Starting links filtering . . .");
IEnumerable <string> letterDirectories = Directory.EnumerateDirectories(pathDataDir, "?", SearchOption.TopDirectoryOnly);
foreach (string pathLetterDir in letterDirectories)
{
/* 1. Get ch_links_?.csv from the current directory
* 2. Load all information from there into a Dictionary<Guid, List<string>>
* 3. Filter all the links for each entry in the dictionary, delete entry if list is empty
*/
Console.WriteLine("Filtering links in " + pathLetterDir);
List <List <string> > characterLinks = new List <List <string> >();
string pathChLinks = FileManager.GetLinksFilePath(pathLetterDir);
using (StreamReader reader = new StreamReader(pathChLinks))
{
while (!reader.EndOfStream)
{
string input = reader.ReadLine();
if (!input.Equals(String.Empty))
{
string[] linksInfo = input.Split(',');
List <string> newLinksInfo = new List <string>();
// Add ID to the new list of links
newLinksInfo.Add(linksInfo[0]);
// Add all links
for (int i = 1; i < linksInfo.Length; ++i)
{
string linkToTest = linksInfo[i];
if (IsValidLink(linkToTest))
{
newLinksInfo.Add(IDHelper.GetIDFromLink(linkToTest, Path.GetFileName(pathLetterDir)));
}
}
characterLinks.Add(newLinksInfo);
}
}
}
// 4. Write all the links to ch_links_?.csv
using (StreamWriter writer = new StreamWriter(pathChLinks))
{
foreach (List <string> listLinks in characterLinks)
{
if (listLinks.Count > 1)
{
writer.Write(listLinks[0]);
for (int i = 1; i < listLinks.Count; ++i)
{
writer.Write(String.Format(",{0}", listLinks[i]));
}
writer.WriteLine();
}
}
}
}
Console.WriteLine("Finished links filtering.");
}
