| public Add ( string match, string replacement ) : void | ||
| match | string | input String to be replaced |
| replacement | string | output String |
| return | void | |
public virtual void TestFinalOffsetSpecialCase()
{
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("t", "");
// even though this below rule has no effect, the test passes if you remove it!!
builder.Add("tmakdbl", "c");
NormalizeCharMap map = builder.Build();
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper2(this, map);
string text = "gzw f quaxot";
CheckAnalysisConsistency(Random(), analyzer, false, text);
}
public virtual void Test()
{
CharArraySet cas = new CharArraySet(TEST_VERSION_CURRENT, 3, false);
cas.add("jjp");
cas.add("wlmwoknt");
cas.add("tcgyreo");
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("mtqlpi", "");
builder.Add("mwoknt", "jjp");
builder.Add("tcgyreo", "zpfpajyws");
NormalizeCharMap map = builder.Build();
Analyzer a = new AnalyzerAnonymousInnerClassHelper(this, cas, map);
CheckAnalysisConsistency(Random(), a, false, "wmgddzunizdomqyj");
}
public virtual void TestFinalOffsetSpecialCase()
{
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("t", "");
// even though this below rule has no effect, the test passes if you remove it!!
builder.Add("tmakdbl", "c");
NormalizeCharMap map = builder.Build();
Analyzer analyzer = Analyzer.NewAnonymous(createComponents: (fieldName, reader) =>
{
Tokenizer tokenizer = new MockTokenizer(reader, MockTokenizer.WHITESPACE, false);
return(new TokenStreamComponents(tokenizer, tokenizer));
}, initReader: (fieldName, reader) => new MappingCharFilter(map, reader));
string text = "gzw f quaxot";
CheckAnalysisConsistency(Random, analyzer, false, text);
}
protected virtual void ParseRules(IList <string> rules, NormalizeCharMap.Builder builder)
{
foreach (string rule in rules)
{
Match m = p.Match(rule);
if (!m.Success)
{
throw new System.ArgumentException("Invalid Mapping Rule : [" + rule + "], file = " + mapping);
}
builder.Add(ParseString(m.Groups[1].Value), ParseString(m.Groups[2].Value));
}
}
public override void SetUp()
{
base.SetUp();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("aa", "a");
builder.Add("bbb", "b");
builder.Add("cccc", "cc");
builder.Add("h", "i");
builder.Add("j", "jj");
builder.Add("k", "kkk");
builder.Add("ll", "llll");
builder.Add("empty", "");
// BMP (surrogate pair):
builder.Add(UnicodeUtil.NewString(new int[] { 0x1D122 }, 0, 1), "fclef");
builder.Add("\uff01", "full-width-exclamation");
normMap = builder.Build();
}
public override void SetUp()
{
base.SetUp();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("aa", "a");
builder.Add("bbb", "b");
builder.Add("cccc", "cc");
builder.Add("h", "i");
builder.Add("j", "jj");
builder.Add("k", "kkk");
builder.Add("ll", "llll");
builder.Add("empty", "");
// BMP (surrogate pair):
builder.Add(UnicodeUtil.NewString(new int[] { 0x1D122 }, 0, 1), "fclef");
builder.Add("\uff01", "full-width-exclamation");
normMap = builder.Build();
}
public virtual void TestOffsetCorrection()
{
const string INPUT = "Günther Günther is here";
// create MappingCharFilter
IList<string> mappingRules = new List<string>();
mappingRules.Add("\"ü\" => \"ü\"");
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("ü", "ü");
NormalizeCharMap normMap = builder.Build();
CharFilter charStream = new MappingCharFilter(normMap, new StringReader(INPUT));
// create PatternTokenizer
TokenStream stream = new PatternTokenizer(charStream, new Regex("[,;/\\s]+", RegexOptions.Compiled), -1);
AssertTokenStreamContents(stream, new string[] { "Günther", "Günther", "is", "here" }, new int[] { 0, 13, 26, 29 }, new int[] { 12, 25, 28, 33 }, INPUT.Length);
charStream = new MappingCharFilter(normMap, new StringReader(INPUT));
stream = new PatternTokenizer(charStream, new Regex("Günther", RegexOptions.Compiled), 0);
AssertTokenStreamContents(stream, new string[] { "Günther", "Günther" }, new int[] { 0, 13 }, new int[] { 12, 25 }, INPUT.Length);
}
private NormalizeCharMap RandomMap()
{
Random random = Random();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
// we can't add duplicate keys, or NormalizeCharMap gets angry
ISet <string> keys = new HashSet <string>();
int num = random.Next(5);
//System.out.println("NormalizeCharMap=");
for (int i = 0; i < num; i++)
{
string key = TestUtil.RandomSimpleString(random);
if (!keys.Contains(key) && key.Length != 0)
{
string value = TestUtil.RandomSimpleString(random);
builder.Add(key, value);
keys.Add(key);
//System.out.println("mapping: '" + key + "' => '" + value + "'");
}
}
return(builder.Build());
}
public virtual void TestRandomMaps2()
{
Random random = Random();
int numIterations = AtLeast(3);
for (int iter = 0; iter < numIterations; iter++)
{
if (VERBOSE)
{
Console.WriteLine("\nTEST iter=" + iter);
}
char endLetter = (char)TestUtil.NextInt(random, 'b', 'z');
IDictionary <string, string> map = new Dictionary <string, string>();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
int numMappings = AtLeast(5);
if (VERBOSE)
{
Console.WriteLine(" mappings:");
}
while (map.Count < numMappings)
{
string key = TestUtil.RandomSimpleStringRange(random, 'a', endLetter, 7);
if (key.Length != 0 && !map.ContainsKey(key))
{
string value = TestUtil.RandomSimpleString(random);
map[key] = value;
builder.Add(key, value);
if (VERBOSE)
{
Console.WriteLine(" " + key + " -> " + value);
}
}
}
NormalizeCharMap charMap = builder.Build();
if (VERBOSE)
{
Console.WriteLine(" test random documents...");
}
for (int iter2 = 0; iter2 < 100; iter2++)
{
string content = TestUtil.RandomSimpleStringRange(random, 'a', endLetter, AtLeast(1000));
if (VERBOSE)
{
Console.WriteLine(" content=" + content);
}
// Do stupid dog-slow mapping:
// Output string:
StringBuilder output = new StringBuilder();
// Maps output offset to input offset:
IList <int?> inputOffsets = new List <int?>();
int cumDiff = 0;
int charIdx = 0;
while (charIdx < content.Length)
{
int matchLen = -1;
string matchRepl = null;
foreach (KeyValuePair <string, string> ent in map)
{
string match = ent.Key;
if (charIdx + match.Length <= content.Length)
{
int limit = charIdx + match.Length;
bool matches = true;
for (int charIdx2 = charIdx; charIdx2 < limit; charIdx2++)
{
if (match[charIdx2 - charIdx] != content[charIdx2])
{
matches = false;
break;
}
}
if (matches)
{
string repl = ent.Value;
if (match.Length > matchLen)
{
// Greedy: longer match wins
matchLen = match.Length;
matchRepl = repl;
}
}
}
}
if (matchLen != -1)
{
// We found a match here!
if (VERBOSE)
{
Console.WriteLine(" match=" + content.Substring(charIdx, matchLen) + " @ off=" + charIdx + " repl=" + matchRepl);
}
output.Append(matchRepl);
int minLen = Math.Min(matchLen, matchRepl.Length);
// Common part, directly maps back to input
// offset:
for (int outIdx = 0; outIdx < minLen; outIdx++)
{
inputOffsets.Add(output.Length - matchRepl.Length + outIdx + cumDiff);
}
cumDiff += matchLen - matchRepl.Length;
charIdx += matchLen;
if (matchRepl.Length < matchLen)
{
// Replacement string is shorter than matched
// input: nothing to do
}
else if (matchRepl.Length > matchLen)
{
// Replacement string is longer than matched
// input: for all the "extra" chars we map
// back to a single input offset:
for (int outIdx = matchLen; outIdx < matchRepl.Length; outIdx++)
{
inputOffsets.Add(output.Length + cumDiff - 1);
}
}
else
{
// Same length: no change to offset
}
Debug.Assert(inputOffsets.Count == output.Length, "inputOffsets.size()=" + inputOffsets.Count + " vs output.length()=" + output.Length);
}
else
{
inputOffsets.Add(output.Length + cumDiff);
output.Append(content[charIdx]);
charIdx++;
}
}
string expected = output.ToString();
if (VERBOSE)
{
Console.Write(" expected:");
for (int charIdx2 = 0; charIdx2 < expected.Length; charIdx2++)
{
Console.Write(" " + expected[charIdx2] + "/" + inputOffsets[charIdx2]);
}
Console.WriteLine();
}
MappingCharFilter mapFilter = new MappingCharFilter(charMap, new StringReader(content));
StringBuilder actualBuilder = new StringBuilder();
IList <int?> actualInputOffsets = new List <int?>();
// Now consume the actual mapFilter, somewhat randomly:
while (true)
{
if (random.Next(0, 1) == 1)
{
int ch = mapFilter.Read();
if (ch == -1)
{
break;
}
actualBuilder.Append((char)ch);
}
else
{
char[] buffer = new char[TestUtil.NextInt(random, 1, 100)];
int off = buffer.Length == 1 ? 0 : random.Next(buffer.Length - 1);
int count = mapFilter.Read(buffer, off, buffer.Length - off);
if (count == -1)
{
break;
}
else
{
actualBuilder.Append(buffer, off, count);
}
}
if (random.Next(10) == 7)
{
// Map offsets
while (actualInputOffsets.Count < actualBuilder.Length)
{
actualInputOffsets.Add(mapFilter.CorrectOffset(actualInputOffsets.Count));
}
}
}
// Finish mappping offsets
while (actualInputOffsets.Count < actualBuilder.Length)
{
actualInputOffsets.Add(mapFilter.CorrectOffset(actualInputOffsets.Count));
}
string actual = actualBuilder.ToString();
// Verify:
assertEquals(expected, actual);
assertEquals(inputOffsets, actualInputOffsets);
}
}
}
public virtual void TestNormalizeWinDelimToLinuxDelim()
{
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("\\", "/");
NormalizeCharMap normMap = builder.Build();
string path = "c:\\a\\b\\c";
Reader cs = new MappingCharFilter(normMap, new StringReader(path));
PathHierarchyTokenizer t = new PathHierarchyTokenizer(cs);
AssertTokenStreamContents(t, new string[] { "c:", "c:/a", "c:/a/b", "c:/a/b/c" }, new int[] { 0, 0, 0, 0 }, new int[] { 2, 4, 6, 8 }, new int[] { 1, 0, 0, 0 }, path.Length);
}
public virtual void TestInvalidOffsets()
{
CharArraySet dict = makeDictionary("fall");
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("ü", "ue");
NormalizeCharMap normMap = builder.Build();
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper(this, dict, normMap);
AssertAnalyzesTo(analyzer, "banküberfall", new string[] { "bankueberfall", "fall" }, new int[] { 0, 0 }, new int[] { 12, 12 });
}
public virtual void TestRandomMaps2()
{
Random random = Random();
int numIterations = AtLeast(3);
for (int iter = 0; iter < numIterations; iter++)
{
if (VERBOSE)
{
Console.WriteLine("\nTEST iter=" + iter);
}
char endLetter = (char)TestUtil.NextInt(random, 'b', 'z');
IDictionary<string, string> map = new Dictionary<string, string>();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
int numMappings = AtLeast(5);
if (VERBOSE)
{
Console.WriteLine(" mappings:");
}
while (map.Count < numMappings)
{
string key = TestUtil.RandomSimpleStringRange(random, 'a', endLetter, 7);
if (key.Length != 0 && !map.ContainsKey(key))
{
string value = TestUtil.RandomSimpleString(random);
map[key] = value;
builder.Add(key, value);
if (VERBOSE)
{
Console.WriteLine(" " + key + " -> " + value);
}
}
}
NormalizeCharMap charMap = builder.Build();
if (VERBOSE)
{
Console.WriteLine(" test random documents...");
}
for (int iter2 = 0; iter2 < 100; iter2++)
{
string content = TestUtil.RandomSimpleStringRange(random, 'a', endLetter, AtLeast(1000));
if (VERBOSE)
{
Console.WriteLine(" content=" + content);
}
// Do stupid dog-slow mapping:
// Output string:
StringBuilder output = new StringBuilder();
// Maps output offset to input offset:
IList<int?> inputOffsets = new List<int?>();
int cumDiff = 0;
int charIdx = 0;
while (charIdx < content.Length)
{
int matchLen = -1;
string matchRepl = null;
foreach (KeyValuePair<string, string> ent in map)
{
string match = ent.Key;
if (charIdx + match.Length <= content.Length)
{
int limit = charIdx + match.Length;
bool matches = true;
for (int charIdx2 = charIdx; charIdx2 < limit; charIdx2++)
{
if (match[charIdx2 - charIdx] != content[charIdx2])
{
matches = false;
break;
}
}
if (matches)
{
string repl = ent.Value;
if (match.Length > matchLen)
{
// Greedy: longer match wins
matchLen = match.Length;
matchRepl = repl;
}
}
}
}
if (matchLen != -1)
{
// We found a match here!
if (VERBOSE)
{
Console.WriteLine(" match=" + content.Substring(charIdx, matchLen) + " @ off=" + charIdx + " repl=" + matchRepl);
}
output.Append(matchRepl);
int minLen = Math.Min(matchLen, matchRepl.Length);
// Common part, directly maps back to input
// offset:
for (int outIdx = 0; outIdx < minLen; outIdx++)
{
inputOffsets.Add(output.Length - matchRepl.Length + outIdx + cumDiff);
}
cumDiff += matchLen - matchRepl.Length;
charIdx += matchLen;
if (matchRepl.Length < matchLen)
{
// Replacement string is shorter than matched
// input: nothing to do
}
else if (matchRepl.Length > matchLen)
{
// Replacement string is longer than matched
// input: for all the "extra" chars we map
// back to a single input offset:
for (int outIdx = matchLen; outIdx < matchRepl.Length; outIdx++)
{
inputOffsets.Add(output.Length + cumDiff - 1);
}
}
else
{
// Same length: no change to offset
}
Debug.Assert(inputOffsets.Count == output.Length, "inputOffsets.size()=" + inputOffsets.Count + " vs output.length()=" + output.Length);
}
else
{
inputOffsets.Add(output.Length + cumDiff);
output.Append(content[charIdx]);
charIdx++;
}
}
string expected = output.ToString();
if (VERBOSE)
{
Console.Write(" expected:");
for (int charIdx2 = 0; charIdx2 < expected.Length; charIdx2++)
{
Console.Write(" " + expected[charIdx2] + "/" + inputOffsets[charIdx2]);
}
Console.WriteLine();
}
MappingCharFilter mapFilter = new MappingCharFilter(charMap, new StringReader(content));
StringBuilder actualBuilder = new StringBuilder();
IList<int?> actualInputOffsets = new List<int?>();
// Now consume the actual mapFilter, somewhat randomly:
while (true)
{
if (random.Next(0, 1) == 1)
{
int ch = mapFilter.Read();
if (ch == -1)
{
break;
}
actualBuilder.Append((char)ch);
}
else
{
char[] buffer = new char[TestUtil.NextInt(random, 1, 100)];
int off = buffer.Length == 1 ? 0 : random.Next(buffer.Length - 1);
int count = mapFilter.Read(buffer, off, buffer.Length - off);
if (count == -1)
{
break;
}
else
{
actualBuilder.Append(buffer, off, count);
}
}
if (random.Next(10) == 7)
{
// Map offsets
while (actualInputOffsets.Count < actualBuilder.Length)
{
actualInputOffsets.Add(mapFilter.CorrectOffset(actualInputOffsets.Count));
}
}
}
// Finish mappping offsets
while (actualInputOffsets.Count < actualBuilder.Length)
{
actualInputOffsets.Add(mapFilter.CorrectOffset(actualInputOffsets.Count));
}
string actual = actualBuilder.ToString();
// Verify:
assertEquals(expected, actual);
assertEquals(inputOffsets, actualInputOffsets);
}
}
}
private NormalizeCharMap RandomMap()
{
Random random = Random();
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
// we can't add duplicate keys, or NormalizeCharMap gets angry
ISet<string> keys = new HashSet<string>();
int num = random.Next(5);
//System.out.println("NormalizeCharMap=");
for (int i = 0; i < num; i++)
{
string key = TestUtil.RandomSimpleString(random);
if (!keys.Contains(key) && key.Length != 0)
{
string value = TestUtil.RandomSimpleString(random);
builder.Add(key, value);
keys.Add(key);
//System.out.println("mapping: '" + key + "' => '" + value + "'");
}
}
return builder.Build();
}
public virtual void TestFinalOffsetSpecialCase()
{
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("t", "");
// even though this below rule has no effect, the test passes if you remove it!!
builder.Add("tmakdbl", "c");
NormalizeCharMap map = builder.Build();
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper2(this, map);
string text = "gzw f quaxot";
CheckAnalysisConsistency(Random(), analyzer, false, text);
}
public virtual void TestChangedOffsets()
{
NormalizeCharMap.Builder builder = new NormalizeCharMap.Builder();
builder.Add("a", "一二");
builder.Add("b", "二三");
NormalizeCharMap norm = builder.Build();
Analyzer analyzer = new AnalyzerAnonymousInnerClassHelper(this, norm);
AssertAnalyzesTo(analyzer, "ab", new string[] { "一二", "二二", "二三" }, new int[] { 0, 0, 1 }, new int[] { 1, 1, 2 });
// note: offsets are strange since this is how the charfilter maps them...
// before bigramming, the 4 tokens look like:
// { 0, 0, 1, 1 },
// { 0, 1, 1, 2 }
}