void Analyze(FlowDocument document)
{
before.Document = document;
string text = new TextRange(document.ContentStart, document.ContentEnd).Text;
//Tutaj jakaś deobfuskacja
MostCommonWords mostCommonWords = new MostCommonWords();
var res = mostCommonWords.Find(text, 3).GetValueOrDefault("and", 0);
// Jeśli and jest pośród 3 najczęstrzych wyrazów to zamień , and na kropki
if (res > 0)
{
text = string.Join(". ", text.Split(", and ").Select(words => Vocabulary.capitalize(words)));
}
BasicAnalizer basicAnalizer = new BasicAnalizer();
var baRes = basicAnalizer.AnalyzeSingleText(text);
var sentenceCount = baRes.GetValueOrDefault("sentenceCount", 0);
var generalNounCount = baRes.GetValueOrDefault("generalNounCount", 0);
// Jeśli generalizacji jest co najwyżej 10 razy mniej niż zdań to spróbuj je usunąć.
if (generalNounCount / sentenceCount > 0.1)
{
List <string> sentences = new List <string>(text.Split("."));
List <string> outputSentences = new List <string>();
for (var i = 0; i < sentences.Count - 1; i++)
{
var sentence = sentences[i].Trim();
List <string> splitSentence = new List <string>(sentence.Split(" "));
// Jeśli zdanie posiada generalizacje
if (splitSentence.Where(e => Vocabulary.generalNouns.Contains(e)).Count() > 0)
{
var nextSentence = sentences[i + 1].Trim();
List <string> splitNextSentence = new List <string>(nextSentence.Split(" "));
if (splitNextSentence.Count < 5)
{
outputSentences.Add(sentence);
continue;
}
bool hasIs = splitNextSentence[2] == "was" || splitNextSentence[2] == "is";
bool hasA = splitNextSentence[3] == "a" || splitNextSentence[3] == "an";
var unGeneralized = splitSentence.Where(e => Vocabulary.generalNouns.Contains(e)).Where(e => {
if (Vocabulary.generalizations.TryGetValue(splitNextSentence[4], out (string, string)value))
{
return(value.Item2 == e);
}
return(false);
});
if (splitNextSentence.Count == 5 && hasIs && hasA && unGeneralized.Count() > 0)
{
var newSentence = sentence;
foreach (var a in unGeneralized)
{
newSentence = newSentence.Replace(a, splitNextSentence[4]);
}
outputSentences.Add(newSentence);
i++;
}
else
{
outputSentences.Add(sentence);
}
}
else
{
outputSentences.Add(sentence);
}
}
var texts = outputSentences.Select(s => s.Trim()).Where(s => s != "").Select(s => Vocabulary.capitalize(s)).ToArray();
text = string.Join(". ", texts) + ".";
}
// Zapisz deobfuskowany tekst
deobfDoc = @"<FlowDocument xmlns=""http://schemas.microsoft.com/winfx/2006/xaml/presentation""
xmlns:local=""clr-namespace:AnalyzerObfuscator.test_documents""
ColumnWidth=""400"" FontSize=""14"" FontFamily=""Georgia"" ColumnGap=""20"" PagePadding=""20"">
<Paragraph>
" + text +
@"
</Paragraph>
</FlowDocument>";
FlowDocument content = XamlReader.Parse(deobfDoc) as FlowDocument;
reader.Document = content;
}
void Analyze(FlowDocument document, Dictionary <string, double> obfsNames)
{
string text = new TextRange(document.ContentStart, document.ContentEnd).Text;
text = text.Trim().Substring(0, text.Trim().Length - 1);
List <IObfuscator> obfs = new List <IObfuscator>();
if (obfsNames.ContainsKey("gen"))
{
obfs.Add(new GeneralizationObfuscator(obfsNames.GetValueOrDefault("gen")));
}
if (obfsNames.ContainsKey("syn"))
{
obfs.Add(new SynonymObfuscator(obfsNames.GetValueOrDefault("syn")));
}
if (obfsNames.ContainsKey("pas"))
{
obfs.Add(new PassiveObfuscator());
}
if (obfsNames.ContainsKey("and"))
{
obfs.Add(new AndObfuscator(obfsNames.GetValueOrDefault("and")));
}
string[] results = IObfuscator.JoinObf(obfs, text.Replace(". ", ".")).Split(".").Select(s => s.Trim()).Where(s => s != "").Select(s => Vocabulary.capitalize(s)).ToArray();
string result = string.Join(". ", results) + ".";
obfDoc = @"<FlowDocument xmlns=""http://schemas.microsoft.com/winfx/2006/xaml/presentation""
xmlns:local=""clr-namespace:AnalyzerObfuscator.test_documents""
ColumnWidth=""400"" FontSize=""14"" FontFamily=""Georgia"" ColumnGap=""20"" PagePadding=""20"">
<Paragraph>
" + result +
@"
</Paragraph>
</FlowDocument>";
FlowDocument content = XamlReader.Parse(obfDoc) as FlowDocument;
before.Document = document;
reader.Document = content;
}
string MakePassive(string sentence)
{
int verbIdx = -1;
Verb verb = null;
List <string> splitSentence = new List <string>(sentence.Split(" "));
for (int i = splitSentence.Count - 1; i >= 0; i--)
{
if (splitSentence[i].Length == 0 || splitSentence[i].Equals(" "))
{
splitSentence.RemoveAt(i);
}
}
sentence = string.Join(" ", splitSentence);
for (int i = 0; i < splitSentence.Count; i++)
{
var verbOpt = Verb.ParseVerb(splitSentence[i]);
if (verbOpt != null)
{
verbIdx = i;
verb = verbOpt;
}
}
if (verb == null || !verb.IsPassive)
{
return(sentence);
}
int n1Idx = -1;
Noun n1 = null;
for (int i = verbIdx - 1; i >= 0; i--)
{
var nounOpt = Noun.ParseNoun(splitSentence[i]);
if (nounOpt != null)
{
n1 = nounOpt;
n1Idx = i;
break;
}
}
int n2Idx = -1;
Noun n2 = null;
for (int i = verbIdx + 1; i < splitSentence.Count; i++)
{
var nounOpt = Noun.ParseNoun(splitSentence[i]);
if (nounOpt != null)
{
n2 = nounOpt;
n2Idx = i;
break;
}
}
if (n1 == null || n2 == null)
{
return(sentence);
}
try
{
int p1Idx = -1;
int p2Idx = -1;
if (n1Idx >= 1 && IsParticle(splitSentence[n1Idx - 1]))
{
p1Idx = n1Idx - 1;
}
else if (n1Idx >= 2 && IsParticle(splitSentence[n1Idx - 2]))
{
p1Idx = n1Idx - 2;
}
if (n2Idx >= 1 && IsParticle(splitSentence[n2Idx - 1]))
{
p2Idx = n2Idx - 1;
}
else if (n2Idx >= 2 && IsParticle(splitSentence[n2Idx - 2]))
{
p2Idx = n2Idx - 2;
}
splitSentence[n1Idx] = n2.Text;
splitSentence[n2Idx] = n1.Text;
splitSentence[verbIdx] = "being " + verb.Text + "en by";
if (p1Idx > -1 && p2Idx > -1)
{
splitSentence[p1Idx] = n2.Particle;
splitSentence[p2Idx] = n1.Particle;
if (p1Idx == n1Idx - 2)
{
splitSentence[p2Idx] += " " + splitSentence[n1Idx - 1];
splitSentence[n1Idx - 1] = "";
}
if (p2Idx == n2Idx - 2)
{
splitSentence[p1Idx] += " " + splitSentence[n2Idx - 1];
splitSentence[n2Idx - 1] = "";
}
}
splitSentence[0] = Vocabulary.capitalize(splitSentence[0]);
return(string.Join(" ", splitSentence));
}
catch (Exception)
{
return(sentence);
}
}
public string ObfuscateText(string text)
{
string[] sentences = text.Split(".");
List <string> outputSentences = new List <string>();
for (int i = 0; i < sentences.Length; i++)
{
if (rand.NextDouble() >= _probability)
{
outputSentences.Add(sentences[i]);
continue;
}
string givenSubject = null;
string sentence = Vocabulary.lower(sentences[i]);
List <string> splitSentence = new List <string>(sentence.Split(" "));
splitSentence[0] = Vocabulary.lower(splitSentence[0]);
List <string> tmpSentences = new List <string>();
foreach (KeyValuePair <string, (string, string)> subject in Vocabulary.generalizations)
{
if (splitSentence.Contains(subject.Key))
{
int idxOfSub = splitSentence.IndexOf(subject.Key);
if (idxOfSub >= 1 && Vocabulary.IsParticle(splitSentence[idxOfSub - 1]))
{
splitSentence[idxOfSub - 1] = subject.Value.Item1;
}
else if (idxOfSub >= 2 && Vocabulary.IsParticle(splitSentence[idxOfSub - 2]))
{
splitSentence[idxOfSub - 2] = subject.Value.Item1;
}
splitSentence[idxOfSub] = subject.Value.Item2;
givenSubject = subject.Key;
Vocabulary.subjects.TryGetValue(subject.Key, out Noun noun);
if (noun == null)
{
break;
}
tmpSentences.Add("The " + subject.Value.Item2 + " was " + noun.Particle + " " + subject.Key);
break;
}
}
if (givenSubject == null)
{
foreach (KeyValuePair <string, Noun> subject in Vocabulary.subjects)
{
if (splitSentence.Contains(subject.Key))
{
givenSubject = subject.Key;
break;
}
}
}
if (givenSubject != null)
{
foreach (KeyValuePair <string, string> adjective in Vocabulary.adjectives)
{
if (splitSentence.Contains(adjective.Key))
{
int index = splitSentence.IndexOf(adjective.Key);
if (index > 1 && index < splitSentence.Count - 1 && (splitSentence[index - 1].ToLower() == "a" || splitSentence[index - 1].ToLower() == "an") && Vocabulary.subjects.TryGetValue(splitSentence[index + 1], out Noun noun))
{
splitSentence[index - 1] = noun.Particle;
}
splitSentence.Remove(adjective.Key);
tmpSentences.Add("The " + givenSubject + " was " + adjective.Key);
}
}
sentence = String.Join(" ", splitSentence);
}
outputSentences.Add(Vocabulary.capitalize(sentence.Trim()));
outputSentences.AddRange(tmpSentences.Select(s => Vocabulary.capitalize(s.Trim())));
}
return(string.Join(".", outputSentences.ToArray()));
}
