private ParagraphIterator SeekLastLeaf()
{
DocumentObject obj = Current;
if (!(obj is ParagraphElements))
{
return(this);
}
List <int> indices = new List <int>(_positionIndices);
while (obj is ParagraphElements)
{
ParagraphElements parEls = (ParagraphElements)obj;
if (((ParagraphElements)obj).Count == 0)
{
return(new ParagraphIterator(_rootNode, obj, indices));
}
int idx = ((ParagraphElements)obj).Count - 1;
indices.Add(idx);
obj = GetNodeObject(parEls[idx]);
}
return(new ParagraphIterator(_rootNode, obj, indices));
}
private static void ProcessParagraphElements(ParagraphElements pe, XmlWriter xmlw)
{
xmlw.WriteStartElement("ParagraphElements");
xmlw.WriteAttributeString("Count", pe.Count.ToStringSafe());
foreach (ParagraphElement p in pe)
{
ProcessParagraphElement(p, xmlw);
}
xmlw.WriteEndElement();
}
private static void ProcessParagraphElements(ParagraphElements pe, Utf8JsonWriter jsonw)
{
jsonw.WritePropertyName("ParagraphElements");
jsonw.WriteStartObject();
jsonw.WriteString("Count", pe.Count.ToStringSafe());
jsonw.WritePropertyName("Items");
jsonw.WriteStartArray();
foreach (ParagraphElement p in pe)
{
ProcessParagraphElement(p, jsonw);
}
jsonw.WriteEndArray();
jsonw.WriteEndObject();
}
/// <summary>
/// Returns the previous iterator to a leaf in the document object tree pointing.
/// </summary>
/// <returns>The previous leaf, null if none exists.</returns>
internal ParagraphIterator GetPreviousLeaf()
{
//Move up to appropriate parent element
ParagraphIterator parIterator = GetParentIterator();
if (parIterator == null)
{
return(null);
}
int elementIndex = LastIndex;
ParagraphElements parEls = (ParagraphElements)parIterator._current;
while (elementIndex == 0)
{
elementIndex = parIterator.LastIndex;
parIterator = parIterator.GetParentIterator();
if (parIterator == null)
{
break;
}
parEls = (ParagraphElements)parIterator._current;
}
if (parIterator == null)
{
return(null);
}
int newIndex = elementIndex - 1;
if (newIndex < 0)
{
return(null);
}
List <int> indices = new List <int>(parIterator._positionIndices);//(Array_List)parIterator.positionIndices.Clone();
indices.Add(newIndex);
DocumentObject obj = GetNodeObject(parEls[newIndex]);
ParagraphIterator iterator = new ParagraphIterator(_rootNode, obj, indices);
return(iterator.SeekLastLeaf());
}
/// <summary>
/// Returns the next iterator in the tree pointing to a leaf.
/// </summary>
/// <remarks>This function is intended to receive the renderable objects of a paragraph.
/// Thus, empty ParagraphElement objects (which are collections) don't count as leafs.</remarks>
internal ParagraphIterator GetNextLeaf()
{
//Move up to appropriate parent element
ParagraphIterator parIterator = GetParentIterator();
if (parIterator == null)
{
return(null);
}
int elementIndex = this.LastIndex;
ParagraphElements parEls = (ParagraphElements)parIterator.current;
while (elementIndex == parEls.Count - 1)
{
elementIndex = parIterator.LastIndex;
parIterator = parIterator.GetParentIterator();
if (parIterator == null)
{
break;
}
parEls = (ParagraphElements)parIterator.current;
}
if (parIterator == null)
{
return(null);
}
int newIndex = elementIndex + 1;
if (newIndex >= parEls.Count)
{
return(null);
}
ArrayList indices = (ArrayList)parIterator.positionIndices.Clone();
indices.Add(newIndex);
DocumentObject obj = GetNodeObject(parEls[newIndex]);
ParagraphIterator iterator = new ParagraphIterator(this.rootNode, obj, indices);
return(iterator.SeekFirstLeaf());
}
/// <summary>
/// Return all text elements in the paragraph - including the text elements
/// inside the FormattedText elements.
/// </summary>
/// <param name="paragraph">The paragraph.</param>
internal static IEnumerable <string> GetTextElements(this ParagraphElements paragraph)
{
if (paragraph == null)
{
throw new ArgumentNullException(nameof(paragraph));
}
foreach (DocumentObject obj in paragraph)
{
if (obj is Text text)
{
yield return(text.Content);
}
else if (obj is Character character)
{
if (character.SymbolName == SymbolName.LineBreak)
{
yield return(Environment.NewLine);
}
else
{
// This probably doesn't work.
yield return(new string(character.Char, character.Count));
}
}
else if (obj is FormattedText formattedText)
{
foreach (string subtext in formattedText.Elements.GetTextElements())
{
yield return(subtext);
}
}
else if (obj is Hyperlink link)
{
foreach (string subtext in link.Elements.GetTextElements())
{
yield return(subtext);
}
}
}
}
/// <summary>
/// Gets the leftmost leaf within the hierarchy.
/// </summary>
/// <returns>The searched leaf.</returns>
ParagraphIterator SeekFirstLeaf()
{
DocumentObject obj = this.Current;
if (!(obj is ParagraphElements))
{
return(this);
}
ArrayList indices = (ArrayList)this.positionIndices.Clone();
while (obj is ParagraphElements)
{
ParagraphElements parEls = (ParagraphElements)obj;
if (parEls.Count == 0)
{
return(new ParagraphIterator(this.rootNode, obj, indices));
}
indices.Add(0);
obj = GetNodeObject(parEls[0]);
}
return(new ParagraphIterator(this.rootNode, obj, indices));
}
/// <summary>
/// Gets the leftmost leaf within the hierarchy.
/// </summary>
/// <returns>The searched leaf.</returns>
ParagraphIterator SeekFirstLeaf()
{
DocumentObject obj = Current;
if (!(obj is ParagraphElements))
{
return(this);
}
List <int> indices = new List <int>(_positionIndices);
while (obj is ParagraphElements)
{
ParagraphElements parEls = (ParagraphElements)obj;
if (parEls.Count == 0)
{
return(new ParagraphIterator(_rootNode, obj, indices));
}
indices.Add(0);
obj = GetNodeObject(parEls[0]);
}
return(new ParagraphIterator(_rootNode, obj, indices));
}
/// <summary>
/// Parses the keyword «\symbol» resp. «\(».
/// </summary>
private void ParseSymbol(ParagraphElements elements)
{
AssertSymbol(Symbol.Symbol);
ReadCode(); // read '('
AssertSymbol(Symbol.ParenLeft);
const char ch = (char)0;
SymbolName symtype = 0;
int count = 1;
ReadCode(); // read name
if (TokenType == TokenType.Identifier)
{
try
{
if (Enum.IsDefined(typeof(SymbolName), Token))
{
AssertCondition(IsSymbolType(Token), DomMsgID.InvalidSymbolType, Token);
symtype = (SymbolName)Enum.Parse(typeof(SymbolName), Token, true);
}
}
catch (Exception ex)
{
ThrowParserException(ex, DomMsgID.InvalidEnum, Token);
}
}
else
{
ThrowParserException(DomMsgID.UnexpectedSymbol, Token);
}
ReadCode(); // read integer or identifier
if (Symbol == Symbol.Comma)
{
ReadCode(); // read integer
if (TokenType == TokenType.IntegerLiteral)
count = _scanner.GetTokenValueAsInt();
ReadCode();
}
AssertSymbol(Symbol.ParenRight);
if (symtype != 0)
elements.AddCharacter(symtype, count);
else
elements.AddCharacter(ch, count);
}
/// <summary>
/// Parses the keyword «\chr».
/// </summary>
private void ParseChr(ParagraphElements elements)
{
AssertSymbol(Symbol.Chr);
ReadCode(); // read '('
AssertSymbol(Symbol.ParenLeft);
char ch = (char)0;
SymbolName symtype = 0;
int count = 1;
ReadCode(); // read integer
if (TokenType == TokenType.IntegerLiteral)
{
int val = _scanner.GetTokenValueAsInt();
if (val >= 1 && val < 256)
ch = (char)val;
else
ThrowParserException(DomMsgID.OutOfRange, "1 - 255");
}
else
{
ThrowParserException(DomMsgID.UnexpectedSymbol, Token);
}
ReadCode(); // read integer or identifier
if (Symbol == Symbol.Comma)
{
ReadCode(); // read integer
if (TokenType == TokenType.IntegerLiteral)
count = _scanner.GetTokenValueAsInt();
ReadCode();
}
AssertSymbol(Symbol.ParenRight);
if (symtype != 0)
elements.AddCharacter(symtype, count);
else
elements.AddCharacter(ch, count);
}
/// <summary>
/// Removes the last blank from the text. Used before a tab, a linebreak or a space will be
/// added to the text.
/// </summary>
private void RemoveTrailingBlank(ParagraphElements elements)
{
DocumentObject dom = elements.LastObject;
Text text = dom as Text;
if (text != null)
{
if (text.Content.EndsWith(" "))
text.Content = text.Content.Remove(text.Content.Length - 1, 1);
}
}
/// <summary>
/// Parses the inner text of a paragraph. Parsing ends if '}' is reached or an empty
/// line occurs on nesting level 0.
/// </summary>
private void ParseFormattedText(ParagraphElements elements, int nestingLevel)
{
MoveToParagraphContent();
bool loop = true;
bool rootLevel = nestingLevel == 0;
ReadText(rootLevel);
while (loop)
{
switch (Symbol)
{
case Symbol.Eof:
ThrowParserException(DomMsgID.UnexpectedEndOfFile);
break;
case Symbol.EmptyLine:
elements.AddCharacter(SymbolName.ParaBreak);
ReadText(rootLevel);
break;
case Symbol.BraceRight:
loop = false;
break;
case Symbol.Comment:
// Ignore comments.
ReadText(rootLevel);
break;
case Symbol.Text:
elements.AddText(Token);
ReadText(rootLevel);
break;
case Symbol.Tab:
RemoveTrailingBlank(elements);
elements.AddTab();
_scanner.MoveToNonWhiteSpaceOrEol();
ReadText(rootLevel);
break;
case Symbol.LineBreak:
RemoveTrailingBlank(elements);
elements.AddLineBreak();
_scanner.MoveToNonWhiteSpaceOrEol();
ReadText(rootLevel);
break;
case Symbol.Bold:
ParseBoldItalicEtc(elements.AddFormattedText(TextFormat.Bold), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Italic:
ParseBoldItalicEtc(elements.AddFormattedText(TextFormat.Italic), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Underline:
ParseBoldItalicEtc(elements.AddFormattedText(TextFormat.Underline), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Font:
ParseFont(elements.AddFormattedText(), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.FontSize:
ParseFontSize(elements.AddFormattedText(), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.FontColor:
ParseFontColor(elements.AddFormattedText(), nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Image:
ParseImage(elements.AddImage(""), true);
ReadText(rootLevel);
break;
case Symbol.Field:
ParseField(elements, nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Footnote:
ParseFootnote(elements, nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Hyperlink:
ParseHyperlink(elements, nestingLevel + 1);
ReadText(rootLevel);
break;
case Symbol.Space:
RemoveTrailingBlank(elements);
ParseSpace(elements, nestingLevel + 1);
_scanner.MoveToNonWhiteSpaceOrEol();
ReadText(rootLevel);
break;
case Symbol.Symbol:
ParseSymbol(elements);
ReadText(rootLevel);
break;
case Symbol.Chr:
ParseChr(elements);
ReadText(rootLevel);
break;
default:
ThrowParserException(DomMsgID.UnexpectedSymbol, Token);
break;
}
}
}
/// <summary>
/// Removes the last blank from the text. Used before a tab, a linebreak or a space will be
/// added to the text.
/// </summary>
private void RemoveTrailingBlank(ParagraphElements elements)
{
DocumentObject dom = elements.LastObject;
if (dom is Text)
{
Text text = (Text)dom;
if (text.Content.EndsWith(" "))
text.Content = text.Content.Remove(text.Content.Length - 1, 1);
}
}
/// <summary>
/// Initializes a paragraph iterator pointing on the given paragraph elements object.
/// Paragraph iterators received from this paragraph iterator relate to this root node.
/// </summary>
/// <param name="rootNode">The root node for the paragraph iterator.</param>
internal ParagraphIterator(ParagraphElements rootNode)
{
this.rootNode = rootNode;
this.current = rootNode;
this.positionIndices = new ArrayList();
}
/// <summary>
/// Initializes a paragraph iterator given the root node, its position in the object tree and the current object
/// </summary>
/// <param name="rootNode">The node the position indices relate to.</param>
/// <param name="current">The element the iterator shall point to.</param>
/// <param name="indices">The position of the paragraph iterator in terms of element indices.</param>
private ParagraphIterator(ParagraphElements rootNode, DocumentObject current, ArrayList indices)
{
this.rootNode = rootNode;
this.positionIndices = indices;
this.current = current;
}
/// <summary>
/// Parses the keyword «\hyperlink».
/// </summary>
private void ParseHyperlink(ParagraphElements elements, int nestingLevel)
{
AssertSymbol(Symbol.Hyperlink);
ReadCode();
Hyperlink hyperlink = elements.AddHyperlink("");
//NYI: Without name and type the hyperlink is senseless, so attributes need to be checked
if (Symbol == Symbol.BracketLeft)
ParseAttributes(hyperlink);
AssertSymbol(Symbol.BraceLeft);
ParseFormattedText(hyperlink.Elements, nestingLevel);
AssertSymbol(Symbol.BraceRight);
}
/// <summary>
/// Parses the keyword «\field».
/// </summary>
private void ParseField(ParagraphElements elements, int nestingLevel)
{
AssertSymbol(Symbol.Field);
ReadCode(); // read '('
AssertSymbol(Symbol.ParenLeft);
ReadCode(); // read identifier
AssertSymbol(Symbol.Identifier);
string fieldType = Token.ToLower();
ReadCode(); // read ')'
AssertSymbol(Symbol.ParenRight);
DocumentObject field = null;
switch (fieldType)
{
case "date":
field = elements.AddDateField();
break;
case "page":
field = elements.AddPageField();
break;
case "numpages":
field = elements.AddNumPagesField();
break;
case "info":
field = elements.AddInfoField(0);
break;
case "sectionpages":
field = elements.AddSectionPagesField();
break;
case "section":
field = elements.AddSectionField();
break;
case "bookmark":
field = elements.AddBookmark("");
break;
case "pageref":
field = elements.AddPageRefField("");
break;
}
AssertCondition(field != null, DomMsgID.InvalidFieldType, Token);
if (_scanner.PeekSymbol() == Symbol.BracketLeft)
{
ReadCode(); // read '['
ParseAttributes(field, false);
}
}
/// <summary>
/// Parses the keyword «\footnote».
/// </summary>
private void ParseFootnote(ParagraphElements elements, int nestingLevel)
{
AssertSymbol(Symbol.Footnote);
ReadCode();
Footnote footnote = elements.AddFootnote();
if (Symbol == Symbol.BracketLeft)
ParseAttributes(footnote);
AssertSymbol(Symbol.BraceLeft);
// The keyword «\paragraph» is typically ommitted.
if (IsParagraphContent())
{
Paragraph paragraph = footnote.Elements.AddParagraph();
ParseParagraphContent(footnote.Elements, paragraph);
}
else
{
ReadCode(); // read beyond '{'
ParseDocumentElements(footnote.Elements, Symbol.Footnote);
}
AssertSymbol(Symbol.BraceRight);
}
/// <summary>
/// Initializes a paragraph iterator pointing on the given paragraph elements object.
/// Paragraph iterators received from this paragraph iterator relate to this root node.
/// </summary>
/// <param name="rootNode">The root node for the paragraph iterator.</param>
internal ParagraphIterator(ParagraphElements rootNode)
{
_rootNode = rootNode;
_current = rootNode;
_positionIndices = new List <int>();
}
/// <summary>
/// Parses the keyword «\space».
/// </summary>
private void ParseSpace(ParagraphElements elements, int nestingLevel)
{
// Samples
// \space
// \space(5)
// \space(em)
// \space(em,5)
AssertSymbol(Symbol.Space);
Character space = elements.AddSpace(1);
// «\space» can stand alone
if (_scanner.PeekSymbol() == Symbol.ParenLeft)
{
ReadCode(); // read '('
AssertSymbol(Symbol.ParenLeft);
ReadCode(); // read beyond '('
if (Symbol == Symbol.Identifier)
{
string type = Token;
if (!IsSpaceType(type))
ThrowParserException(DomMsgID.InvalidEnum, type);
space.SymbolName = (SymbolName)Enum.Parse(typeof(SymbolName), type, true);
ReadCode(); // read ',' or ')'
if (Symbol == Symbol.Comma)
{
ReadCode(); // read integer
AssertSymbol(Symbol.IntegerLiteral);
space.Count = _scanner.GetTokenValueAsInt();
ReadCode(); // read ')'
}
}
else if (Symbol == Symbol.IntegerLiteral)
{
space.Count = _scanner.GetTokenValueAsInt();
ReadCode();
}
AssertSymbol(Symbol.ParenRight);
}
}
/// <summary>
/// Initializes a paragraph iterator given the root node, its position in the object tree and the current object
/// </summary>
/// <param name="rootNode">The node the position indices relate to.</param>
/// <param name="current">The element the iterator shall point to.</param>
/// <param name="indices">The position of the paragraph iterator in terms of element indices.</param>
private ParagraphIterator(ParagraphElements rootNode, DocumentObject current, List <int> indices)
{
_rootNode = rootNode;
_positionIndices = indices;
_current = current;
}
