private void AddWord(ITextToken tok, WordAndPunct wap)
{
TextTokenSubstring tts = new TextTokenSubstring(tok, wap.Offset, wap.Word.Length);
if (desiredKey == "" || desiredKey == tts.InventoryText)
result.Add(tts);
}
private void CheckWordAndPunct(WordAndPunct wordAndPunct, string word, string punct, int offset)
{
Assert.AreEqual(word, wordAndPunct.Word, "The word is not correct");
Assert.AreEqual(punct, wordAndPunct.Punct, "The punctuation is not correct");
Assert.AreEqual(offset, wordAndPunct.Offset, "The offset is not correct");
}
private void ProcessWord(ITextToken tok, WordAndPunct wap)
{
if (wap.Word == "")
return;
string nextWord = wap.Word.ToLower();
if (prevWord == nextWord)
AddWord(tok, wap);
prevWord = nextWord;
// If there are characters (such as quotes) between words,
// then two words are not considered repeating, even if they are identical
foreach (char cc in wap.Punct)
{
if (!char.IsWhiteSpace(cc))
{
Reset();
break;
}
}
}
//public string PunctuationCharacters { get { return punctuationCharacters; } }
public virtual List<WordAndPunct> WordAndPuncts(string text)
{
char cc;
int punctOffset;
List<WordAndPunct> waps = new List<WordAndPunct>();
for (int i = 0; i < text.Length; )
{
WordAndPunct wap = new WordAndPunct();
// Ignore any initial separator characters
while (i < text.Length && char.IsSeparator(text[i]))
i++;
wap.Offset = i;
bool isFirstCharacterInWord = true;
while (i < text.Length)
{
cc = text[i];
if (IsSingleCharacterWord(cc))
{
if (isFirstCharacterInWord)
{
// Single Character key is the first character in the key.
// It forms a key all by itself.
i = i + 1;
}
else
{
// Single Character key is NOT the first character in the key.
// It ends the key currently being formed.
// 'i' is not incremented
}
break;
}
else if (IsWordMedialPunctuation(cc))
{
// Word medial punctuation only counts if this is not the first
// character in a key AND the next character is key forming.
//! can we have multiple key medial punctuation?
if (isFirstCharacterInWord)
break;
if (i + 1 >= text.Length || !IsWordFormingCharacter(text[i + 1]))
break;
}
else if (char.IsDigit(cc))
{
// allow digits in words
}
else if (!IsWordFormingCharacter(cc))
break;
i = i + 1;
isFirstCharacterInWord = false;
}
wap.Word = text.Substring(wap.Offset, i - wap.Offset);
punctOffset = i;
while (i < text.Length)
{
cc = text[i];
if (IsWordFormingCharacter(cc) || char.IsDigit(cc))
break;
i = i + 1;
}
wap.Punct = text.Substring(punctOffset, i - punctOffset);
waps.Add(wap);
}
return waps;
}
/// ------------------------------------------------------------------------------------
/// <summary>
/// Gets words and punctuation from text.
/// </summary>
/// <param name="text">The text.</param>
/// <returns>a collection of words and punctuation</returns>
/// ------------------------------------------------------------------------------------
public override List <WordAndPunct> WordAndPuncts(string text)
{
char cc;
int punctOffset;
List <WordAndPunct> waps = new List <WordAndPunct>();
for (int i = 0; i < text.Length;)
{
WordAndPunct wap = new WordAndPunct();
// Ignore any initial separator characters
while (i < text.Length && m_charPropEngine.get_IsSeparator(text[i]))
{
i++;
}
if (i == text.Length)
{
return(waps);
}
wap.Offset = i;
bool isFirstCharacterInWord = true;
while (i < text.Length)
{
cc = text[i];
if (IsSingleCharacterWord(cc))
{
if (isFirstCharacterInWord)
{
// Single Character key is the first character in the key.
// It forms a key all by itself.
i = i + 1;
}
else
{
// Single Character key is NOT the first character in the key.
// It ends the key currently being formed.
// 'i' is not incremented
}
break;
}
else if (m_charPropEngine.get_IsNumber(cc))
{
// allow digits in words
}
else if (!m_validChars.IsWordForming(cc))
{
break;
}
i = i + 1;
isFirstCharacterInWord = false;
}
wap.Word = text.Substring(wap.Offset, i - wap.Offset);
punctOffset = i;
while (i < text.Length)
{
cc = text[i];
if (m_validChars.IsWordForming(cc) || m_charPropEngine.get_IsNumber(cc))
{
break;
}
i = i + 1;
}
wap.Punct = text.Substring(punctOffset, i - punctOffset);
waps.Add(wap);
}
return(waps);
}
/// ------------------------------------------------------------------------------------
/// <summary>
///
/// </summary>
/// <param name="tok"></param>
/// <param name="wap"></param>
/// <param name="desiredKey"></param>
/// ------------------------------------------------------------------------------------
private void AddWord(ITextToken tok, WordAndPunct wap, string desiredKey)
{
TextTokenSubstring tts = new TextTokenSubstring(tok, wap.Offset, wap.Word.Length);
if (String.IsNullOrEmpty(desiredKey) || desiredKey == tts.InventoryText)
m_result.Add(tts);
}
/// ------------------------------------------------------------------------------------
/// <summary>
///
/// </summary>
/// <param name="tok"></param>
/// <param name="wap"></param>
/// <param name="desiredKey"></param>
/// ------------------------------------------------------------------------------------
public void ProcessWord(ITextToken tok, WordAndPunct wap, string desiredKey)
{
AWord word = new AWord(wap.Word, m_categorizer);
if (word.Prefix == string.Empty && word.Suffix == string.Empty)
return;
if (m_uncapitalizedPrefixes.Contains(word.Prefix))
return;
if (m_uncapitalizedPrefixes.Contains("*" + word.Prefix[word.Prefix.Length - 1]))
return;
if (m_uncapitalizedPrefixes.Contains("*"))
return;
if (m_capitalizedSuffixes.Contains(word.Suffix))
return;
if (m_capitalizedPrefixes.Contains(word.Prefix))
return;
AddWord(tok, wap, desiredKey);
}
/// ------------------------------------------------------------------------------------
/// <summary>
/// Gets words and punctuation from text.
/// </summary>
/// <param name="text">The text.</param>
/// <returns>a collection of words and punctuation</returns>
/// ------------------------------------------------------------------------------------
public override List<WordAndPunct> WordAndPuncts(string text)
{
char cc;
int punctOffset;
List<WordAndPunct> waps = new List<WordAndPunct>();
for (int i = 0; i < text.Length; )
{
WordAndPunct wap = new WordAndPunct();
// Ignore any initial separator characters
while (i < text.Length && m_charPropEngine.get_IsSeparator(text[i]))
i++;
if (i == text.Length)
return waps;
wap.Offset = i;
bool isFirstCharacterInWord = true;
while (i < text.Length)
{
cc = text[i];
if (IsSingleCharacterWord(cc))
{
if (isFirstCharacterInWord)
{
// Single Character key is the first character in the key.
// It forms a key all by itself.
i = i + 1;
}
else
{
// Single Character key is NOT the first character in the key.
// It ends the key currently being formed.
// 'i' is not incremented
}
break;
}
else if (m_charPropEngine.get_IsNumber(cc))
{
// allow digits in words
}
else if (!m_validChars.IsWordForming(cc))
break;
i = i + 1;
isFirstCharacterInWord = false;
}
wap.Word = text.Substring(wap.Offset, i - wap.Offset);
punctOffset = i;
while (i < text.Length)
{
cc = text[i];
if (m_validChars.IsWordForming(cc) || m_charPropEngine.get_IsNumber(cc))
break;
i = i + 1;
}
wap.Punct = text.Substring(punctOffset, i - punctOffset);
waps.Add(wap);
}
return waps;
}
//public string PunctuationCharacters { get { return punctuationCharacters; } }
public virtual List <WordAndPunct> WordAndPuncts(string text)
{
char cc;
int punctOffset;
List <WordAndPunct> waps = new List <WordAndPunct>();
for (int i = 0; i < text.Length;)
{
WordAndPunct wap = new WordAndPunct();
// Ignore any initial separator characters
while (i < text.Length && char.IsSeparator(text[i]))
{
i++;
}
wap.Offset = i;
bool isFirstCharacterInWord = true;
while (i < text.Length)
{
cc = text[i];
if (IsSingleCharacterWord(cc))
{
if (isFirstCharacterInWord)
{
// Single Character key is the first character in the key.
// It forms a key all by itself.
i = i + 1;
}
else
{
// Single Character key is NOT the first character in the key.
// It ends the key currently being formed.
// 'i' is not incremented
}
break;
}
else if (IsWordMedialPunctuation(cc))
{
// Word medial punctuation only counts if this is not the first
// character in a key AND the next character is key forming.
//! can we have multiple key medial punctuation?
if (isFirstCharacterInWord)
{
break;
}
if (i + 1 >= text.Length || !IsWordFormingCharacter(text[i + 1]))
{
break;
}
}
else if (char.IsDigit(cc))
{
// allow digits in words
}
else if (!IsWordFormingCharacter(cc))
{
break;
}
i = i + 1;
isFirstCharacterInWord = false;
}
wap.Word = text.Substring(wap.Offset, i - wap.Offset);
punctOffset = i;
while (i < text.Length)
{
cc = text[i];
if (IsWordFormingCharacter(cc) || char.IsDigit(cc))
{
break;
}
i = i + 1;
}
wap.Punct = text.Substring(punctOffset, i - punctOffset);
waps.Add(wap);
}
return(waps);
}
private void CheckWordAndPunct(WordAndPunct wordAndPunct, string word, string punct, int offset)
{
Assert.AreEqual(word, wordAndPunct.Word, "The word is not correct");
Assert.AreEqual(punct, wordAndPunct.Punct, "The punctuation is not correct");
Assert.AreEqual(offset, wordAndPunct.Offset, "The offset is not correct");
}
