static void TestBlum(string path)
{
var buf = File.ReadAllBytes(path);
var tests = LoadTestFiles(spath_tests_folder);
// load
int maxStringLen = BitConverter.ToInt32(DataPack.Unpack(buf, 0), 0);
//int count = BitConverter.ToInt32(DataPack.Unpack(buf, 0),sizeof(int));
int filterBufLen = BitConverter.ToInt32(DataPack.Unpack(buf, 0), sizeof(int));
byte[] arr = DataPack.Unpack(buf, 1);
var excs2 = SplitString(ASCIIEncoding.ASCII.GetString(DataPack.Unpack(buf, 2)));
var ends = SplitString(ASCIIEncoding.ASCII.GetString(DataPack.Unpack(buf, 3)));
var d2b = SplitString(ASCIIEncoding.ASCII.GetString(DataPack.Unpack(buf, 4)));
var bf = new BlumFilter(filterBufLen);
bf.SetArray(arr);
var flt = new PreFilter(excs2, ends, d2b, maxStringLen);
int errcount = 0;
foreach (var t in tests)
{
bool isWord = false;
var s = t.Key.ToLower().Trim();
string sb;
var sf = flt.WordStatus(s, out sb);
if (sf == 0)
{
isWord = false;
}
else if (sf == 1)
{
isWord = true;
}
else
{
isWord = bf.Contains(sb);
}
if (isWord && !t.Value)
{
errcount++;
}
if (!isWord && t.Value)
{
errcount++;
}
}
Console.WriteLine("Errors:{0} from {1}, {2}",
errcount, tests.Count, (float)errcount / (float)(tests.Count) * 100.0F);
}
static void Main(string[] args)
{
// loadTests(spath_tests_folder);
// CombineTests(spath_tests_folder, spath_tests_2);
int filterBufLen = 506464;
int maxStringLen = 12;
int endingsCount = 23;
double err = 0.04;
var sz = LoadAndOrderAndSave(spath0, spath2, spathe, maxStringLen);
SaveAvailable2Beg(spath2, spath2e, spath3, err);
TreeAnalyze(spath3, spath_ends, maxStringLen, filterBufLen, endingsCount);
var buf = File.ReadAllLines(spath3);
var excs2 = File.ReadAllLines(spathe).ToList();
var ends = File.ReadAllLines(spath_ends).ToList();
var d2b = File.ReadAllLines(spath2e).ToList();
Console.WriteLine("d2b:{0}", d2b.Count);
var flt = new PreFilter(excs2, ends, d2b, maxStringLen);
Dictionary <string, bool> dicf = new Dictionary <string, bool>();
List <string> buf2 = new List <string>();
Console.WriteLine("Words before filtering:{0}", buf.Length);
// filter buffer
foreach (var sb in buf)
{
string sbase;
if (flt.WordStatus(sb, out sbase) == 2)
{
dicf[sbase] = true;
}
}
buf2 = dicf.Keys.ToList();
buf2.Sort();
Console.WriteLine("Words after {0} level filtering:{1}", ends.Count, buf2.Count);
var bf = new BlumFilter(filterBufLen);
bf.AddRange(buf2);
if (File.Exists(spath_filter))
{
File.Delete(spath_filter);
}
File.WriteAllBytes(spath_filter, bf.GetArray());
// pack data
DataPack.Pack(maxStringLen, buf2.Count, filterBufLen, spath_data, spath_filter, spathe, spath_ends, spath2e);
TestBlum(spath_data);
var cs = CalcSources();
Console.WriteLine("Packet size:{0} Kb, overflow:{1} bytes", cs / 1024.0, cs - 64 * 1024);
Console.ReadLine();
return;
}
static void TreeAnalyze(string path, string path2, int maxStringLen, int arrLen, int endsLimit)
{
var dic = new Dictionary <string, Pair3>();
var dic2 = new Dictionary <string, bool>();
PreFilter flt = new PreFilter(new List <string>(), new List <string>(), new List <string>(), maxStringLen);
var fsl = new List <string>();
foreach (var ss in File.ReadAllLines(path))
{
string bs;
if (flt.WordStatus(ss, out bs) == 2)
{
fsl.Add(ss);
}
}
foreach (var s in fsl)
{
dic2[s] = true;
}
// *********************************************
int sizeLimit = arrLen;
int elementsCount = dic2.Count;
foreach (var s in fsl)
{
// get all endings
for (int i = 1; i < (s.Length - 1); i++)
{
var sb = s.Substring(i);
var sbase = s.Substring(0, i);
Pair3 v;
if (dic2.ContainsKey(sbase))
{
if (!dic.TryGetValue(sb, out v))
{
v = new Pair3(sb, 0, 0, sizeLimit, elementsCount);
dic[sb] = v;
}
v.Correlations++;
}
if (dic.TryGetValue(sb, out v))
{
v.Count++;
}
}
}
Debug.WriteLine("Endings:{0}", dic.Count);
var lst = dic.Values.ToList();
lst.Sort((x, y) =>
{
return(x.Error.CompareTo(y.Error));
});
var dic3 = new Dictionary <string, Pair3>();
lst.ForEach(l =>
{
bool isFound = false;
for (int i = 0; i < (l.Ends.Length - 1); i++)
{
var sv = l.Ends.Substring(i);
if (dic3.ContainsKey(sv))
{
isFound = true;
break;
}
}
if (!isFound)
{
dic3.Add(l.Ends, l);
}
});
lst = dic3.Values.ToList();
lst.Sort((x, y) =>
{
return(x.Error.CompareTo(y.Error));
});
Debug.WriteLine("Endings reduced:{0}", lst.Count);
dic3.Clear();
// go from the end
for (int j = lst.Count - 1; j >= 0; j--)
{
for (int i = 0; i < (lst[j].Ends.Length - 1); i++)
{
var sv = lst[j].Ends.Substring(i);
if (dic3.ContainsKey(sv))
{
dic3.Remove(sv);
}
}
dic3.Add(lst[j].Ends, lst[j]);
}
lst = dic3.Values.ToList();
lst.Sort((x, y) =>
{
return(x.Error.CompareTo(y.Error));
});
Debug.WriteLine("Endings reduced:{0}", lst.Count);
if (File.Exists(path2))
{
File.Delete(path2);
}
var ends = lst.Select(l => l.Ends).Take(endsLimit);
File.WriteAllLines(path2, ends);
}