public TextTraits(TextTraits Traits)
: this()
{
mWhitespacePatterns.Replace(Traits.mWhitespacePatterns);
mNewLinePatterns.Replace(Traits.mNewLinePatterns);
mDividerPatterns.Replace(Traits.mDividerPatterns);
this.DividerDelimClasses = Traits.DividerDelimClasses;
mCommentToEndPatterns.Replace(Traits.mCommentToEndPatterns);
mQuotePatterns.Replace(Traits.mQuotePatterns);
this.VerbatimLiteralPattern = Traits.VerbatimLiteralPattern;
mExpressionPatterns.Replace(Traits.mExpressionPatterns);
mEndStmtPatterns.Replace(Traits.mEndStmtPatterns);
mOpenNamedBracedPatterns.Replace(Traits.mOpenNamedBracedPatterns);
mOpenContentBracedPatterns.Replace(Traits.mOpenContentBracedPatterns);
mOpenBracedPatterns = null;
mCloseBracedPatterns = null;
mPathSepPatterns.Replace(Traits.mPathSepPatterns);
mNonWordPatterns = null;
mQem = Traits.mQem;
mFormSentencesFromWhitespaceDelimWords =
Traits.mFormSentencesFromWhitespaceDelimWords;
}
public CsvBuilder()
{
mTraits = new TextTraits();
mTraits.DividerPatterns.AddDistinct(",", Enums.DelimClassification.DividerSymbol);
mTraits.OpenNamedBracedPatterns.Replace("(", Enums.DelimClassification.OpenNamedBraced);
mTraits.QuoteEncapsulation = QuoteEncapsulation.Double;
}
public WordValue(string InValue, TextTraits InTraits)
{
CharObjectPair results = Stringer.CalcWordClassification(InValue, InTraits);
CommonConstruct(
InValue, (WordClassification)results.b, InTraits, results.a);
}
// ------------------------ ScanFirstWord -------------------------
public static WordCursor ScanFirstWord(string InString, TextTraits InTraits)
{
BoundedString boundedString = new BoundedString(InString);
WordCursor res = ScanFirstWord(boundedString, InTraits);
return(res);
}
public CsvString( )
{
Empty( );
mTraits = new TextTraits( );
mTraits.DelimChars = new char[] { ',' };
mTraits.BraceChars = new char[] { '(' };
}
// ----------------------- ScanWord_IsolatedWord_Braced -----------------------
public static void ScanWord_IsolateWord_Braced(
string Text,
int WordBx,
ScanPatternResults NonWordResults,
ref WordCursor Results,
TextTraits Traits)
{
string wordText;
int Lx, Ix;
int braceIx = NonWordResults.FoundPos;
char braceChar = NonWordResults.FoundPat.LeadChar;
if (Traits.BracedTreatment == ScannerBracedTreatment.Parts)
{
// a standalone open brace char. the brace char is the word ( and it will
// also be the delim )
if (WordBx == braceIx)
{
Results.SetWord(
NonWordResults.FoundPat.PatternValue,
WordClassification.OpenContentBraced, WordBx, braceChar);
}
else
{
wordText =
Text.Substring(WordBx, braceIx - WordBx);
Results.SetWord(
wordText, WordClassification.OpenNamedBraced, WordBx, braceChar);
}
}
// the whole braced word. braced word runs all the way to the closing brace.
else if (Traits.BracedTreatment == ScannerBracedTreatment.Whole)
{
int remLx = Text.Length - braceIx;
Ix = ScanCloseBrace(
Text,
braceIx, remLx, Traits.QuoteEncapsulation);
if (Ix == -1)
{
throw new ApplicationException(
"Closing brace not found starting at position " +
braceIx + " in " + Text);
}
Lx = Ix - WordBx + 1;
wordText = Text.Substring(WordBx, Lx);
if (WordBx == braceIx)
{
Results.SetWord(
wordText, WordClassification.ContentBraced, WordBx, braceChar);
}
else
{
Results.SetWord(
wordText, WordClassification.NamedBraced, WordBx, braceChar);
}
}
}
// ------------------ ScanWord_IsolatedWord_CommentToEnd -------------------
private static ScanPatternResults ScanWord_IsolateWord_CommentToEnd(
BoundedString InBoundedString,
int InWordBx,
ref WordCursor InOutResults, TextTraits InTraits)
{
string wordText;
ScanPatternResults spr = null;
// look for end of comment. ( either end of line or end of string )
int fx = Scanner.ScanEqual(InBoundedString, InWordBx, Environment.NewLine).ResultPos;
if (fx >= 0)
{
int Lx = fx - InWordBx;
wordText = InBoundedString.Substring(InWordBx, Lx);
ScanPattern pat =
InTraits.NonWordPatterns.FindPatternAtSubstring(InBoundedString, fx);
spr = new ScanPatternResults(fx, pat);
}
else
{
wordText = InBoundedString.Substring(InWordBx);
spr = new ScanPatternResults(-1);
}
// store info on the word found in the return WordCursor argument.
InOutResults.SetWord(wordText, WordClassification.CommentToEnd, InWordBx);
// return value of method contains info on the word delim.
return(spr);
}
// ------------------ ScanWord_IsolatedWord_CommentToEnd -------------------
public static ScanPatternResults ScanWord_IsolateWord_CommentToEnd(
string Text,
int WordBx,
ref WordCursor Results, TextTraits Traits)
{
string wordText;
ScanPatternResults spr = null;
// look for end of comment. ( either end of line or end of string )
int fx = Text.IndexOf(Environment.NewLine, WordBx);
if (fx >= 0)
{
int Lx = fx - WordBx;
wordText = Text.Substring(WordBx, Lx);
ScanPattern pat =
Traits.NonWordPatterns.FindPatternAtSubstring(Text, fx, 2);
spr = new ScanPatternResults(fx, pat);
}
else
{
wordText = Text.Substring(WordBx);
spr = new ScanPatternResults(-1);
}
// store info on the word found in the return WordCursor argument.
Results.SetWord(wordText, WordClassification.CommentToEnd, WordBx);
// return value of method contains info on the word delim.
return(spr);
}
public ScanNonWordxResults(
TextTraits InTextTraits, ScannerNonWord InNonWord, ScanCharResults InCharResults)
{
mTextTraits = InTextTraits;
mNonWord = InNonWord;
mCharResults = InCharResults;
}
// ----------------------- ScanWord_IsolatedWord_Braced -----------------------
private static void ScanWord_IsolateWord_Braced(
BoundedString InBoundedString,
int InWordBx,
ScanPatternResults InNonWordResults,
ref WordCursor InOutResults,
TextTraits InTraits)
{
string wordText;
int Lx, Ix;
int braceIx = InNonWordResults.FoundPos;
char braceChar = InNonWordResults.FoundPat.LeadChar;
if (InTraits.BracedTreatment == ScannerBracedTreatment.Parts)
{
// a standalone open brace char. the brace char is the word ( and it will
// also be the delim )
if (InWordBx == braceIx)
{
InOutResults.SetWord(
InNonWordResults.FoundPat.PatternValue,
WordClassification.OpenContentBraced, InWordBx, braceChar);
}
else
{
wordText =
InBoundedString.String.Substring(InWordBx, braceIx - InWordBx);
InOutResults.SetWord(
wordText, WordClassification.OpenNamedBraced, InWordBx, braceChar);
}
}
// the whole braced word. braced word runs all the way to the closing brace.
else if (InTraits.BracedTreatment == ScannerBracedTreatment.Whole)
{
Ix = ScanCloseBrace(
InBoundedString.String,
braceIx, InBoundedString.Ex, InTraits.QuoteEncapsulation);
if (Ix == -1)
{
throw new ApplicationException(
"Closing brace not found starting at position " +
braceIx + " in " + InBoundedString.String);
}
Lx = Ix - InWordBx + 1;
wordText = InBoundedString.String.Substring(InWordBx, Lx);
if (InWordBx == braceIx)
{
InOutResults.SetWord(
wordText, WordClassification.ContentBraced, InWordBx, braceChar);
}
else
{
InOutResults.SetWord(
wordText, WordClassification.NamedBraced, InWordBx, braceChar);
}
}
}
// ----------------------- PositionBeginWord -----------------------------
// position at begin of string to scan.
public static WordCursor PositionBeginWord(string Text, TextTraits Traits)
{
WordCursor word = new WordCursor( );
word.Position = RelativePosition.Begin;
word.SetTraits(Traits);
return(word);
}
// ----------------------------- constructor ---------------------------
public WordCursor()
{
EmptyWordParts();
mTraits = null;
mString = null;
mStayAtFlag = false;
mVirtualCursor = enumVirtualCursor.None;
this.DelimClass = DelimClassification.NotAssigned;
}
/// <summary>
/// position a cursor before a character location in the string.
/// </summary>
/// <param name="InText"></param>
/// <param name="InIx"></param>
/// <returns></returns>
public static WordCursor PositionBefore(string InText, int InIx, TextTraits InTraits)
{
WordCursor csr = new WordCursor();
csr.Position = RelativePosition.Before;
csr.WordBx = InIx;
csr.TextTraits = InTraits;
return(csr);
}
void ConstructCommon()
{
Empty();
mTraits = new TextTraits();
mTraits.DividerPatterns.AddDistinct(",", DelimClassification.DividerSymbol);
mTraits.OpenNamedBracedPatterns.Replace("(", DelimClassification.OpenNamedBraced);
mTraits.QuoteEncapsulation = QuoteEncapsulation.Double;
EncapsulateValues = false;
}
public WordCursor PositionBefore(int InIx, TextTraits InTraits)
{
WordCursor csr = new WordCursor();
ThrowOutsideBounds(InIx);
csr.Position = RelativePosition.Before;
csr.WordBx = InIx;
csr.TextTraits = InTraits;
return(csr);
}
IsolateNumericLiteral(
ScanStream ScanStream,
TextTraits Traits,
int Bx)
{
// for now, all numeric literals are simple integers.
// have to expand to determine if a float, decimal, what the sign is,
// what the precision is.
LiteralType litType = LiteralType.Integer;
string litText = null;
PatternScanResults nonWord = null;
// step from char to char. Look for a char that is not part of the
// numeric literal.
int ix = Bx;
int litEx = Bx;
while (true)
{
if (ix >= ScanStream.Stream.Length)
{
break;
}
char ch1 = ScanStream.Stream[ix];
if (Char.IsDigit(ch1) == false)
{
break;
}
litEx = ix;
ix += 1;
}
// isolate the numeric literal.
int lx = litEx - Bx + 1;
litText = ScanStream.Substring(Bx, lx);
// isolate the delim that follows that numeric literal.
int bx = litEx + 1;
if (bx < ScanStream.Stream.Length)
{
nonWord = Scanner.ScanEqualAny(ScanStream.Stream, bx, Traits.DelimPatterns);
// foundPat = rv.Item1;
// foundIx = rv.Item2;
// foundLx = rv.Item3;
// nonWord = rv.Item3;
}
return(new Tuple <LiteralType, string, PatternScanResults>(
litType, litText, nonWord));
}
// ------------------------ ScanFirstWord -------------------------
public static WordCursor ScanFirstWord(
BoundedString InBoundedString, TextTraits InTraits)
{
WordCursor csr = new WordCursor();
csr.Position = RelativePosition.Begin;
csr.TextTraits = InTraits;
WordCursor res = ScanNextWord(InBoundedString, csr);
return(res);
}
public WordCursor(WordCursor InWord)
{
mWordValue = InWord.mWordValue;
mWordBx = InWord.WordBx;
mDelimBx = InWord.DelimBx;
mDelim = InWord.mDelim;
mDelimIsWhitespace = InWord.mDelimIsWhitespace;
mRltv = InWord.mRltv;
mTraits = InWord.mTraits;
mString = InWord.mString;
}
public CsvCursor(string InCsvString)
{
mTraits = new TextTraits();
mTraits.DividerPatterns.AddDistinct(",", Enums.DelimClassification.DividerSymbol);
mTraits.OpenNamedBracedPatterns.Replace("[", Enums.DelimClassification.OpenNamedBraced);
mCursor = new WordCursor();
mCursor.TextTraits = mTraits;
mCursor.Position = RelativePosition.Begin;
mCsvString = InCsvString;
}
public static TextLinesWordCursor ScanFirstWord(
TextLines InLines, TextTraits InTraits)
{
TextLinesWordCursor csr = new TextLinesWordCursor();
csr.Position = RelativePosition.Begin;
csr.TextTraits = InTraits;
TextLinesWordCursor res = ScanNextWord(InLines, csr);
return(res);
}
public TextWord(string Value, WordClassification WordClass, TextTraits Traits)
{
char?braceChar = null;
if (WordClass.IsOpenBraced())
{
braceChar = Value[0];
if (Traits.IsOpenBraceChar(braceChar.Value) == false)
{
throw new ApplicationException("not an open brace char");
}
}
CommonConstruct(Value, WordClass, Traits, braceChar);
}
public WordCursor(WordCursor InWord)
{
mWord = InWord.mWord;
mWordBx = InWord.WordBx;
mDelimBx = InWord.DelimBx;
mDelim = InWord.mDelim;
mDelimIsWhitespace = InWord.mDelimIsWhitespace;
this.DelimClass = InWord.mDelimClass;
mWhitespaceFollowsWord = InWord.mWhitespaceFollowsWord;
mWhitespaceFollowsDelim = InWord.mWhitespaceFollowsDelim;
mRltv = InWord.mRltv;
mTraits = InWord.mTraits;
mString = InWord.mString;
mStayAtFlag = InWord.mStayAtFlag;
mVirtualCursor = InWord.mVirtualCursor;
}
/// <summary>
/// calc if the word starting at InBx is part of a path.
/// </summary>
/// <param name="InBoundedString"></param>
/// <param name="InTraits"></param>
/// <param name="InBx"></param>
/// <returns></returns>
private static bool ScanWord_IsolateDelim_IsPathPart(
BoundedString InBoundedString, TextTraits InTraits, int InBx)
{
bool rc = false;
WordCursor csr = InBoundedString.PositionBefore(InBx, InTraits);
csr = ScanNextWord(InBoundedString, csr);
if ((csr.IsPathPart == true))
{
rc = true;
}
return(rc);
}
// --------------------------- ScanWord_IsolateDelim_SetDelimIsWhitespace ----------
private static void ScanWord_IsolateDelim_SetDelimIsWhitespace(
BoundedString InBoundedString, TextTraits InTraits,
WordCursor InOutResults, int InWsIx)
{
// store the actual string of whitespace characters. ( the whitespace can be
// checked later to see if it contains tabs or newlines )
ScanPatternResults spr = ScanNotEqual(
InBoundedString.String, InWsIx, InBoundedString.Ex,
InTraits.WhitespacePatterns);
string delimVlu = spr.ScannedOverString;
InOutResults.SetDelim(
InBoundedString, delimVlu, InWsIx, DelimClassification.Whitespace);
InOutResults.DelimIsWhitespace = true;
}
private void CommonConstruct(
string InValue, WordClassification InClass, TextTraits InTraits, char InBraceChar)
{
mValue = InValue;
mClass = InClass;
mTraits = InTraits;
mBraceChar = InBraceChar;
// depending on word class, brace char must be specified or be blanks.
if ((mClass == WordClassification.Braced) ||
(mClass == WordClassification.NameBraced))
{
if (mBraceChar == ' ')
{
throw(new ApplicationException("Brace character of braced word is missing"));
}
}
}
// -------------------- ScanWord_IsolatexWord ---------------------------
private static TextLinesWordCursor ScanWord_IsolatexWord(
TextLines InLines,
TextLinesCursor InBxCsr,
TextTraits InTraits)
{
TextLinesWordCursor tlwc = null;
TextLinesCursor csr = null;
TextLinesCursor endcsr = null;
ScanPatternResults spr = null;
csr = new TextLinesCursor(InBxCsr);
char ch1 = InBxCsr.CursorChar;
// look ahead to see if this word is braced.
if (IsOpenQuoteChar(ch1) == false)
{
spr =
ScanEqualAny(csr.LineData, csr.LineOx, csr.LineData.Length - 1, InTraits.NonWordPatterns);
}
else
{
spr = new ScanPatternResults(-1);
}
// the rule is only braced words can span multiple lines. so if the word is
// not braced, it can be parsed by the more general purpose IsolatexWord method.
if ((IsOpenQuoteChar(ch1) == true) ||
(InTraits.IsOpenBraceChar(spr.FoundChar.Value) == false))
{
ScanBoundedString bs = new ScanBoundedString(csr.LineData);
WordCursor wc = new WordCursor();
wc.SetTraits(InTraits);
ScanWord_IsolateWord(bs, csr.LineOx, ref wc, InTraits);
endcsr =
new TextLinesCursor(csr.LinesNode, wc.ScanEx, AcRelativePosition.At);
tlwc = new TextLinesWordCursor(wc, csr, endcsr);
}
else
{
}
return(tlwc);
}
// ------------------------ AdvanceNextWord -------------------------
// Scans to the next word in the string. ( a word being the text bounded by the
// delimeter and whitespace characters as spcfd in the TextTraits argument )
// Return null when end of string.
public static void AdvancexNextWord(
string InString,
WordCursor InOutWordCursor,
TextTraits InTraits)
{
int Bx;
BoundedString boundedString = new BoundedString(InString);
ScanPatternResults spr = null;
// calc scan start position
Bx = ScanWord_CalcStartBx(boundedString, InOutWordCursor);
// empty the word parts of the cursor.
InOutWordCursor.EmptyWordParts();
// advance past whitespace
if (Bx <= boundedString.Ex)
{
Bx = ScanNotEqual(
boundedString.String, Bx, boundedString.Ex,
InTraits.WhitespacePatterns).FoundPos;
}
// got the start of something. scan for the delimeter ( could be the current char )
spr = null;
if (Bx <= boundedString.Ex)
{
spr = ScanWord_IsolateWord(boundedString, Bx, ref InOutWordCursor, InTraits);
}
// depending on the word, isolate and store the delim that follows.
ScanWord_IsolateDelim(boundedString, spr, ref InOutWordCursor, InTraits);
// current word position.
if (InOutWordCursor.ScanEx == -1)
{
InOutWordCursor.Position = RelativePosition.End;
}
else
{
InOutWordCursor.Position = RelativePosition.At;
}
}
// ------------------------ ScanNextWord -------------------------
// Scans to the next word in the string. ( a word being the text bounded by the
// delimeter and whitespace characters as spcfd in the TextTraits argument )
// Return null when end of string.
public static WordCursor ScanNextWord(
string InString,
WordCursor InCurrentWord,
TextTraits InTraits)
{
int Bx;
WordCursor results = new WordCursor( )
.SetString(InString)
.SetTraits(InTraits);
// calc scan start position
Bx = ScanWord_CalcStartBx(InString, InCurrentWord);
// advance past whitespace
if (Bx < InString.Length)
{
Bx = ScanNotEqual(InString, Bx, InTraits.WhitespaceChars).ResultPos;
}
// got the start of something. scan for the delimeter ( could be the current char )
if (Bx < InString.Length)
{
ScanWord_IsolateWord(InString, Bx, ref results, InTraits);
}
// depending on the word, isolate and store the delim that follows.
ScanWord_IsolateDelim(InString, Bx, ref results, InTraits);
// current word position.
if (results.ScanEx == -1)
{
results.RelativePosition = AcRelativePosition.End;
}
else
{
results.RelativePosition = AcRelativePosition.At;
}
return(results);
}
public TextWord(string Text, LiteralType LiteralType, TextTraits Traits)
{
mValue = Text;
_LiteralType = LiteralType;
if (this.LiteralType.IsNumeric( ))
{
mClass = WordClassification.Numeric;
}
else if (this.LiteralType.IsQuoted( ))
{
mClass = WordClassification.Quoted;
}
else
{
throw new ApplicationException(
"LiteralType is not supported. " + LiteralType.ToString());
}
mTraits = Traits;
_BraceChar = null;
}
// ------------------------ AdvanceNextWord -------------------------
// Scans to the next word in the string. ( a word being the text bounded by the
// delimeter and whitespace characters as spcfd in the TextTraits argument )
// Return null when end of string.
public static void AdvanceNextWord(
string InString,
WordCursor InOutWordCursor,
TextTraits InTraits)
{
int Bx;
// calc scan start position
Bx = ScanWord_CalcStartBx(InString, InOutWordCursor);
// empty the word parts of the cursor.
InOutWordCursor.EmptyWord( );
// advance past whitespace
if (Bx < InString.Length)
{
Bx = ScanNotEqual(InString, Bx, InTraits.WhitespaceChars).ResultPos;
}
// got the start of something. scan for the delimeter ( could be the current char )
if (Bx < InString.Length)
{
ScanWord_IsolateWord(InString, Bx, ref InOutWordCursor, InTraits);
}
// depending on the word, isolate and store the delim that follows.
ScanWord_IsolateDelim(InString, Bx, ref InOutWordCursor, InTraits);
// current word position.
if (InOutWordCursor.ScanEx == -1)
{
InOutWordCursor.RelativePosition = AcRelativePosition.End;
}
else
{
InOutWordCursor.RelativePosition = AcRelativePosition.At;
}
}
