novelWriter 2.5.1__py3-none-any.whl → 2.6b1__py3-none-any.whl

novelwriter/{core → formats}/tokenizer.py

@@ -24,41 +24,56 @@ along with this program. If not, see <https://www.gnu.org/licenses/>.
 """
 from __future__ import annotations
 
-import json
 import logging
 import re
 
 from abc import ABC, abstractmethod
-from functools import partial
 from pathlib import Path
-from time import time
+from typing import NamedTuple
 
-from PyQt5.QtCore import QCoreApplication, QRegularExpression
-from PyQt5.QtGui import QFont
+from PyQt5.QtCore import QLocale
+from PyQt5.QtGui import QColor, QFont
 
 from novelwriter import CONFIG
-from novelwriter.common import checkInt, formatTimeStamp, numberToRoman
-from novelwriter.constants import nwHeadFmt, nwKeyWords, nwLabels, nwShortcode, nwUnicode, trConst
+from novelwriter.common import checkInt, numberToRoman
+from novelwriter.constants import (
+    nwHeadFmt, nwKeyWords, nwLabels, nwShortcode, nwStats, nwStyles, nwUnicode,
+    trConst
+)
 from novelwriter.core.index import processComment
 from novelwriter.core.project import NWProject
 from novelwriter.enum import nwComment, nwItemLayout
-from novelwriter.text.patterns import REGEX_PATTERNS
+from novelwriter.formats.shared import (
+    BlockFmt, BlockTyp, T_Block, T_Formats, T_Note, TextDocumentTheme, TextFmt
+)
+from novelwriter.text.patterns import REGEX_PATTERNS, DialogParser
 
 logger = logging.getLogger(__name__)
 
-ESCAPES = {r"\*": "*", r"\~": "~", r"\_": "_", r"\[": "[", r"\]": "]", r"\ ": ""}
-RX_ESC = re.compile("|".join([re.escape(k) for k in ESCAPES.keys()]), flags=re.DOTALL)
 
-T_Formats = list[tuple[int, int, str]]
-T_Comment = tuple[str, T_Formats]
-T_Token = tuple[int, int, str, T_Formats, int]
+class ComStyle(NamedTuple):
 
+    label: str = ""
+    labelClass: str = ""
+    textClass: str = ""
 
-def stripEscape(text: str) -> str:
-    """Strip escaped Markdown characters from paragraph text."""
-    if "\\" in text:
-        return RX_ESC.sub(lambda x: ESCAPES[x.group(0)], text)
-    return text
+
+COMMENT_STYLE = {
+    nwComment.PLAIN:    ComStyle("Comment", "comment", "comment"),
+    nwComment.IGNORE:   ComStyle(),
+    nwComment.SYNOPSIS: ComStyle("Synopsis", "modifier", "synopsis"),
+    nwComment.SHORT:    ComStyle("Short Description", "modifier", "synopsis"),
+    nwComment.NOTE:     ComStyle("Note", "modifier", "note"),
+    nwComment.FOOTNOTE: ComStyle("", "modifier", "note"),
+    nwComment.COMMENT:  ComStyle(),
+    nwComment.STORY:    ComStyle("", "modifier", "note"),
+}
+HEADINGS = [BlockTyp.TITLE, BlockTyp.HEAD1, BlockTyp.HEAD2, BlockTyp.HEAD3, BlockTyp.HEAD4]
+SKIP_INDENT = [
+    BlockTyp.TITLE, BlockTyp.HEAD1, BlockTyp.HEAD2, BlockTyp.HEAD2, BlockTyp.HEAD3,
+    BlockTyp.HEAD4, BlockTyp.SEP, BlockTyp.SKIP,
+]
+B_EMPTY: T_Block = (BlockTyp.EMPTY, "", "", [], BlockFmt.NONE)
 
 
 class Tokenizer(ABC):
@@ -70,119 +85,71 @@ class Tokenizer(ABC):
     subclasses.
     """
 
-    # In-Text Format
-    FMT_B_B   = 1   # Begin bold
-    FMT_B_E   = 2   # End bold
-    FMT_I_B   = 3   # Begin italics
-    FMT_I_E   = 4   # End italics
-    FMT_D_B   = 5   # Begin strikeout
-    FMT_D_E   = 6   # End strikeout
-    FMT_U_B   = 7   # Begin underline
-    FMT_U_E   = 8   # End underline
-    FMT_M_B   = 9   # Begin mark
-    FMT_M_E   = 10  # End mark
-    FMT_SUP_B = 11  # Begin superscript
-    FMT_SUP_E = 12  # End superscript
-    FMT_SUB_B = 13  # Begin subscript
-    FMT_SUB_E = 14  # End subscript
-    FMT_DL_B  = 15  # Begin dialogue
-    FMT_DL_E  = 16  # End dialogue
-    FMT_ADL_B = 17  # Begin alt dialogue
-    FMT_ADL_E = 18  # End alt dialogue
-    FMT_FNOTE = 19  # Footnote marker
-    FMT_STRIP = 20  # Strip the format code
-
-    # Block Type
-    T_EMPTY    = 1   # Empty line (new paragraph)
-    T_SYNOPSIS = 2   # Synopsis comment
-    T_SHORT    = 3   # Short description comment
-    T_COMMENT  = 4   # Comment line
-    T_KEYWORD  = 5   # Command line
-    T_TITLE    = 6   # Title
-    T_HEAD1    = 7   # Heading 1
-    T_HEAD2    = 8   # Heading 2
-    T_HEAD3    = 9   # Heading 3
-    T_HEAD4    = 10  # Heading 4
-    T_TEXT     = 11  # Text line
-    T_SEP      = 12  # Scene separator
-    T_SKIP     = 13  # Paragraph break
-
-    # Block Style
-    A_NONE     = 0x0000  # No special style
-    A_LEFT     = 0x0001  # Left aligned
-    A_RIGHT    = 0x0002  # Right aligned
-    A_CENTRE   = 0x0004  # Centred
-    A_JUSTIFY  = 0x0008  # Justified
-    A_PBB      = 0x0010  # Page break before
-    A_PBA      = 0x0020  # Page break after
-    A_Z_TOPMRG = 0x0040  # Zero top margin
-    A_Z_BTMMRG = 0x0080  # Zero bottom margin
-    A_IND_L    = 0x0100  # Left indentation
-    A_IND_R    = 0x0200  # Right indentation
-    A_IND_T    = 0x0400  # Text indentation
-
-    # Masks
-    M_ALIGNED = A_LEFT | A_RIGHT | A_CENTRE | A_JUSTIFY
-
-    # Lookups
-    L_HEADINGS = [T_TITLE, T_HEAD1, T_HEAD2, T_HEAD3, T_HEAD4]
-    L_SKIP_INDENT = [T_TITLE, T_HEAD1, T_HEAD2, T_HEAD2, T_HEAD3, T_HEAD4, T_SEP, T_SKIP]
-    L_SUMMARY = [T_SYNOPSIS, T_SHORT]
-
     def __init__(self, project: NWProject) -> None:
 
         self._project = project
 
         # Data Variables
-        self._text   = ""     # The raw text to be tokenized
-        self._handle = None   # The item handle currently being processed
-        self._result = ""     # The result of the last document
-        self._keepMD = False  # Whether to keep the markdown text
-
-        # Tokens and Meta Data (Per Document)
-        self._tokens: list[T_Token] = []
-        self._footnotes: dict[str, T_Comment] = {}
-
-        # Tokens and Meta Data (Per Instance)
+        self._text     = ""     # The raw text to be tokenized
+        self._handle   = None   # The item handle currently being processed
+        self._keepRaw  = False  # Whether to keep the raw text, used by ToRaw
+        self._noTokens = False  # Disable tokenization if they're not needed
+
+        # Blocks and Meta Data (Per Document)
+        self._blocks: list[T_Block] = []
+        self._footnotes: dict[str, T_Note] = {}
+
+        # Blocks and Meta Data (Per Instance)
+        self._raw: list[str] = []
+        self._pages: list[str] = []
         self._counts: dict[str, int] = {}
         self._outline: dict[str, str] = {}
-        self._markdown: list[str] = []
 
         # User Settings
+        self._dLocale      = CONFIG.locale  # The document locale
         self._textFont     = QFont("Serif", 11)  # Output text font
-        self._lineHeight   = 1.15     # Line height in units of em
-        self._blockIndent  = 4.00     # Block indent in units of em
-        self._firstIndent  = False    # Enable first line indent
-        self._firstWidth   = 1.40     # First line indent in units of em
-        self._indentFirst  = False    # Indent first paragraph
-        self._doJustify    = False    # Justify text
-        self._doBodyText   = True     # Include body text
-        self._doSynopsis   = False    # Also process synopsis comments
-        self._doComments   = False    # Also process comments
-        self._doKeywords   = False    # Also process keywords like tags and references
-        self._skipKeywords = set()    # Keywords to ignore
-        self._keepBreaks   = True     # Keep line breaks in paragraphs
+        self._lineHeight   = 1.15    # Line height in units of em
+        self._colorHeads   = True    # Colourise headings
+        self._scaleHeads   = True    # Scale headings to larger font size
+        self._boldHeads    = True    # Bold headings
+        self._blockIndent  = 4.00    # Block indent in units of em
+        self._firstIndent  = False   # Enable first line indent
+        self._firstWidth   = 1.40    # First line indent in units of em
+        self._indentFirst  = False   # Indent first paragraph
+        self._doJustify    = False   # Justify text
+        self._doBodyText   = True    # Include body text
+        self._doSynopsis   = False   # Also process synopsis comments
+        self._doComments   = False   # Also process comments
+        self._doKeywords   = False   # Also process keywords like tags and references
+        self._keepBreaks   = True    # Keep line breaks in paragraphs
+        self._defaultAlign = "left"  # The default text alignment
+
+        self._skipKeywords: set[str] = set()  # Keywords to ignore
+
+        # Other Setting
+        self._theme = TextDocumentTheme()
+        self._classes: dict[str, QColor] = {}
 
         # Margins
-        self._marginTitle = (1.417, 0.500)
-        self._marginHead1 = (1.417, 0.500)
-        self._marginHead2 = (1.668, 0.500)
-        self._marginHead3 = (1.168, 0.500)
-        self._marginHead4 = (1.168, 0.500)
-        self._marginText  = (0.000, 0.584)
-        self._marginMeta  = (0.000, 0.584)
-        self._marginFoot  = (1.417, 0.467)
-        self._marginSep   = (1.168, 1.168)
+        self._marginTitle = nwStyles.T_MARGIN["H0"]
+        self._marginHead1 = nwStyles.T_MARGIN["H1"]
+        self._marginHead2 = nwStyles.T_MARGIN["H2"]
+        self._marginHead3 = nwStyles.T_MARGIN["H3"]
+        self._marginHead4 = nwStyles.T_MARGIN["H4"]
+        self._marginText  = nwStyles.T_MARGIN["TT"]
+        self._marginMeta  = nwStyles.T_MARGIN["MT"]
+        self._marginFoot  = nwStyles.T_MARGIN["FT"]
+        self._marginSep   = nwStyles.T_MARGIN["SP"]
 
         # Title Formats
-        self._fmtTitle   = nwHeadFmt.TITLE  # Formatting for titles
+        self._fmtPart    = nwHeadFmt.TITLE  # Formatting for partitions
         self._fmtChapter = nwHeadFmt.TITLE  # Formatting for numbered chapters
         self._fmtUnNum   = nwHeadFmt.TITLE  # Formatting for unnumbered chapters
         self._fmtScene   = nwHeadFmt.TITLE  # Formatting for scenes
         self._fmtHScene  = nwHeadFmt.TITLE  # Formatting for hard scenes
         self._fmtSection = nwHeadFmt.TITLE  # Formatting for sections
 
-        self._hideTitle   = False  # Do not include title headings
+        self._hidePart    = False  # Do not include partition headings
         self._hideChapter = False  # Do not include chapter headings
         self._hideUnNum   = False  # Do not include unnumbered headings
         self._hideScene   = False  # Do not include scene headings
@@ -191,15 +158,16 @@ class Tokenizer(ABC):
 
         self._linkHeadings = False  # Add an anchor before headings
 
-        self._titleStyle   = self.A_CENTRE | self.A_PBB
-        self._chapterStyle = self.A_PBB
-        self._sceneStyle   = self.A_NONE
+        self._titleStyle   = BlockFmt.CENTRE | BlockFmt.PBB
+        self._partStyle    = BlockFmt.CENTRE | BlockFmt.PBB
+        self._chapterStyle = BlockFmt.PBB
+        self._sceneStyle   = BlockFmt.NONE
 
         # Instance Variables
         self._hFormatter = HeadingFormatter(self._project)
         self._noSep      = True   # Flag to indicate that we don't want a scene separator
         self._noIndent   = False  # Flag to disable text indent on next paragraph
-        self._showDialog = False  # Flag for dialogue highlighting
+        self._breakNext  = False  # Add a page break on next token
 
         # This File
         self._isNovel = False  # Document is a novel document
@@ -210,31 +178,33 @@ class Tokenizer(ABC):
 
         # Function Mapping
         self._localLookup = self._project.localLookup
-        self.tr = partial(QCoreApplication.translate, "Tokenizer")
 
         # Format RegEx
         self._rxMarkdown = [
-            (REGEX_PATTERNS.markdownItalic, [0, self.FMT_I_B, 0, self.FMT_I_E]),
-            (REGEX_PATTERNS.markdownBold,   [0, self.FMT_B_B, 0, self.FMT_B_E]),
-            (REGEX_PATTERNS.markdownStrike, [0, self.FMT_D_B, 0, self.FMT_D_E]),
+            (REGEX_PATTERNS.markdownItalic, [0, TextFmt.I_B, 0, TextFmt.I_E]),
+            (REGEX_PATTERNS.markdownBold,   [0, TextFmt.B_B, 0, TextFmt.B_E]),
+            (REGEX_PATTERNS.markdownStrike, [0, TextFmt.D_B, 0, TextFmt.D_E]),
         ]
-        self._rxShortCodes = REGEX_PATTERNS.shortcodePlain
-        self._rxShortCodeVals = REGEX_PATTERNS.shortcodeValue
 
         self._shortCodeFmt = {
-            nwShortcode.ITALIC_O: self.FMT_I_B,   nwShortcode.ITALIC_C: self.FMT_I_E,
-            nwShortcode.BOLD_O:   self.FMT_B_B,   nwShortcode.BOLD_C:   self.FMT_B_E,
-            nwShortcode.STRIKE_O: self.FMT_D_B,   nwShortcode.STRIKE_C: self.FMT_D_E,
-            nwShortcode.ULINE_O:  self.FMT_U_B,   nwShortcode.ULINE_C:  self.FMT_U_E,
-            nwShortcode.MARK_O:   self.FMT_M_B,   nwShortcode.MARK_C:   self.FMT_M_E,
-            nwShortcode.SUP_O:    self.FMT_SUP_B, nwShortcode.SUP_C:    self.FMT_SUP_E,
-            nwShortcode.SUB_O:    self.FMT_SUB_B, nwShortcode.SUB_C:    self.FMT_SUB_E,
+            nwShortcode.ITALIC_O: TextFmt.I_B,   nwShortcode.ITALIC_C: TextFmt.I_E,
+            nwShortcode.BOLD_O:   TextFmt.B_B,   nwShortcode.BOLD_C:   TextFmt.B_E,
+            nwShortcode.STRIKE_O: TextFmt.D_B,   nwShortcode.STRIKE_C: TextFmt.D_E,
+            nwShortcode.ULINE_O:  TextFmt.U_B,   nwShortcode.ULINE_C:  TextFmt.U_E,
+            nwShortcode.MARK_O:   TextFmt.M_B,   nwShortcode.MARK_C:   TextFmt.M_E,
+            nwShortcode.SUP_O:    TextFmt.SUP_B, nwShortcode.SUP_C:    TextFmt.SUP_E,
+            nwShortcode.SUB_O:    TextFmt.SUB_B, nwShortcode.SUB_C:    TextFmt.SUB_E,
         }
         self._shortCodeVals = {
-            nwShortcode.FOOTNOTE_B: self.FMT_FNOTE,
+            nwShortcode.FOOTNOTE_B: TextFmt.FNOTE,
+            nwShortcode.FIELD_B:    TextFmt.FIELD,
         }
 
-        self._rxDialogue: list[tuple[QRegularExpression, int, int]] = []
+        # Dialogue
+        self._hlightDialog = False
+        self._rxAltDialog = REGEX_PATTERNS.altDialogStyle
+        self._dialogParser = DialogParser()
+        self._dialogParser.initParser()
 
         return
 
@@ -242,16 +212,6 @@ class Tokenizer(ABC):
     #  Properties
     ##
 
-    @property
-    def result(self) -> str:
-        """The result of the build process."""
-        return self._result
-
-    @property
-    def allMarkdown(self) -> list[str]:
-        """The combined novelWriter Markdown text."""
-        return self._markdown
-
     @property
     def textStats(self) -> dict[str, int]:
         """The collected stats about the text."""
@@ -271,10 +231,21 @@ class Tokenizer(ABC):
     #  Setters
     ##
 
-    def setTitleFormat(self, hFormat: str, hide: bool = False) -> None:
-        """Set the title format pattern."""
-        self._fmtTitle = hFormat.strip()
-        self._hideTitle = hide
+    def setLanguage(self, language: str | None) -> None:
+        """Set language for the document."""
+        if language:
+            self._dLocale = QLocale(language)
+        return
+
+    def setTheme(self, theme: TextDocumentTheme) -> None:
+        """Set the document colour theme."""
+        self._theme = theme
+        return
+
+    def setPartitionFormat(self, hFormat: str, hide: bool = False) -> None:
+        """Set the partition format pattern."""
+        self._fmtPart = hFormat.strip()
+        self._hidePart = hide
         return
 
     def setChapterFormat(self, hFormat: str, hide: bool = False) -> None:
@@ -309,23 +280,26 @@ class Tokenizer(ABC):
 
     def setTitleStyle(self, center: bool, pageBreak: bool) -> None:
         """Set the title heading style."""
-        self._titleStyle = (
-            (self.A_CENTRE if center else self.A_NONE) | (self.A_PBB if pageBreak else self.A_NONE)
-        )
+        self._titleStyle = BlockFmt.CENTRE if center else BlockFmt.NONE
+        self._titleStyle |= BlockFmt.PBB if pageBreak else BlockFmt.NONE
+        return
+
+    def setPartitionStyle(self, center: bool, pageBreak: bool) -> None:
+        """Set the partition heading style."""
+        self._partStyle = BlockFmt.CENTRE if center else BlockFmt.NONE
+        self._partStyle |= BlockFmt.PBB if pageBreak else BlockFmt.NONE
         return
 
     def setChapterStyle(self, center: bool, pageBreak: bool) -> None:
         """Set the chapter heading style."""
-        self._chapterStyle = (
-            (self.A_CENTRE if center else self.A_NONE) | (self.A_PBB if pageBreak else self.A_NONE)
-        )
+        self._chapterStyle = BlockFmt.CENTRE if center else BlockFmt.NONE
+        self._chapterStyle |= BlockFmt.PBB if pageBreak else BlockFmt.NONE
         return
 
     def setSceneStyle(self, center: bool, pageBreak: bool) -> None:
         """Set the scene heading style."""
-        self._sceneStyle = (
-            (self.A_CENTRE if center else self.A_NONE) | (self.A_PBB if pageBreak else self.A_NONE)
-        )
+        self._sceneStyle = BlockFmt.CENTRE if center else BlockFmt.NONE
+        self._sceneStyle |= BlockFmt.PBB if pageBreak else BlockFmt.NONE
         return
 
     def setFont(self, font: QFont) -> None:
@@ -338,6 +312,13 @@ class Tokenizer(ABC):
         self._lineHeight = min(max(float(height), 0.5), 5.0)
         return
 
+    def setHeadingStyles(self, color: bool, scale: bool, bold: bool) -> None:
+        """Set text style for headings."""
+        self._colorHeads = color
+        self._scaleHeads = scale
+        self._boldHeads = bold
+        return
+
     def setBlockIndent(self, indent: float) -> None:
         """Set the block indent between 0.0 and 10.0."""
         self._blockIndent = min(max(float(indent), 0.0), 10.0)
@@ -357,27 +338,9 @@ class Tokenizer(ABC):
         self._doJustify = state
         return
 
-    def setDialogueHighlight(self, state: bool) -> None:
+    def setDialogHighlight(self, state: bool) -> None:
         """Enable or disable dialogue highlighting."""
-        self._rxDialogue = []
-        self._showDialog = state
-        if state:
-            if CONFIG.dialogStyle > 0:
-                self._rxDialogue.append((
-                    REGEX_PATTERNS.dialogStyle, self.FMT_DL_B, self.FMT_DL_E
-                ))
-            if CONFIG.dialogLine:
-                self._rxDialogue.append((
-                    REGEX_PATTERNS.dialogLine, self.FMT_DL_B, self.FMT_DL_E
-                ))
-            if CONFIG.narratorBreak:
-                self._rxDialogue.append((
-                    REGEX_PATTERNS.narratorBreak, self.FMT_DL_E, self.FMT_DL_B
-                ))
-            if CONFIG.altDialogOpen and CONFIG.altDialogClose:
-                self._rxDialogue.append((
-                    REGEX_PATTERNS.altDialogStyle, self.FMT_ADL_B, self.FMT_ADL_E
-                ))
+        self._hlightDialog = state
         return
 
     def setTitleMargins(self, upper: float, lower: float) -> None:
@@ -455,11 +418,6 @@ class Tokenizer(ABC):
         self._keepBreaks = state
         return
 
-    def setKeepMarkdown(self, state: bool) -> None:
-        """Keep original markdown during build."""
-        self._keepMD = state
-        return
-
     ##
     #  Class Methods
     ##
@@ -468,27 +426,54 @@ class Tokenizer(ABC):
     def doConvert(self) -> None:
         raise NotImplementedError
 
+    @abstractmethod
+    def closeDocument(self) -> None:
+        raise NotImplementedError
+
+    @abstractmethod
+    def saveDocument(self, path: Path) -> None:
+        raise NotImplementedError
+
+    def initDocument(self) -> None:
+        """Initialise data after settings."""
+        self._classes["modifier"] = self._theme.modifier
+        self._classes["synopsis"] = self._theme.note
+        self._classes["comment"] = self._theme.comment
+        self._classes["dialog"] = self._theme.dialog
+        self._classes["altdialog"] = self._theme.altdialog
+        self._classes["tag"] = self._theme.tag
+        self._classes["keyword"] = self._theme.keyword
+        self._classes["optional"] = self._theme.optional
+        return
+
+    def setBreakNext(self) -> None:
+        """Set a page break for next block."""
+        self._breakNext = True
+        return
+
     def addRootHeading(self, tHandle: str) -> None:
         """Add a heading at the start of a new root folder."""
         self._text = ""
         self._handle = None
 
-        if (tItem := self._project.tree[tHandle]) and tItem.isRootType():
+        if (item := self._project.tree[tHandle]) and item.isRootType():
             self._handle = tHandle
+            style = BlockFmt.CENTRE
             if self._isFirst:
-                textAlign = self.A_CENTRE
                 self._isFirst = False
             else:
-                textAlign = self.A_PBB | self.A_CENTRE
+                style |= BlockFmt.PBB
 
-            trNotes = self._localLookup("Notes")
-            title = f"{trNotes}: {tItem.itemName}"
-            self._tokens = []
-            self._tokens.append((
-                self.T_TITLE, 1, title, [], textAlign
-            ))
-            if self._keepMD:
-                self._markdown.append(f"#! {title}\n\n")
+            title = item.itemName
+            if not item.isNovelLike():
+                notes = self._localLookup("Notes")
+                title = f"{notes}: {title}"
+
+            self._blocks = [(
+                BlockTyp.TITLE, f"{self._handle}:T0001", title, [], style
+            )]
+            if self._keepRaw:
+                self._raw.append(f"#! {title}\n\n")
 
         return
 
@@ -515,9 +500,11 @@ class Tokenizer(ABC):
             xRep = re.compile("|".join([re.escape(k) for k in repDict.keys()]), flags=re.DOTALL)
             self._text = xRep.sub(lambda x: repDict[x.group(0)], self._text)
 
-        # Process the character translation map
-        trDict = {nwUnicode.U_MAPOS: nwUnicode.U_RSQUO}
-        self._text = self._text.translate(str.maketrans(trDict))
+        # Process the translation map for placeholder characters
+        self._text = self._text.translate(str.maketrans({
+            nwUnicode.U_MAPOS: nwUnicode.U_RSQUO,
+            nwUnicode.U_HBAR: nwUnicode.U_EMDASH,
+        }))
 
         return
 
@@ -526,42 +513,50 @@ class Tokenizer(ABC):
         characters that indicate headings, comments, commands etc, or
         just contain plain text. In the case of plain text, apply the
         same RegExes that the syntax highlighter uses and save the
-        locations of these formatting tags into the token array.
+        locations of these formatting tags into the blocks list.
 
-        The format of the token list is an entry with a five-tuple for
+        The format of the blocs list is an entry with a five-tuple for
         each line in the file. The tuple is as follows:
-          1: The type of the block, self.T_*
+          1: The type of the block, BlockType.*
           2: The heading number under which the text is placed
           3: The text content of the block, without leading tags
-          4: The internal formatting map of the text, self.FMT_*
-          5: The style of the block, self.A_*
+          4: The internal formatting map of the text, TxtFmt.*
+          5: The formats of the block, BlockFmt.*
         """
+        if self._keepRaw:
+            self._raw.append(f"{self._text.rstrip()}\n\n")
+        if self._noTokens:
+            return
         if self._isNovel:
             self._hFormatter.setHandle(self._handle)
 
+        # Cache Flags
+        isNovel = self._isNovel
+        doJustify = self._doJustify
+        keepBreaks = self._keepBreaks
+        indentFirst = self._indentFirst
+        firstIndent = self._firstIndent
+
+        # Replace all instances of [br] with a placeholder character
+        text = REGEX_PATTERNS.lineBreak.sub("\uffff", self._text)
+
         nHead = 0
-        breakNext = False
-        tmpMarkdown = []
         tHandle = self._handle or ""
-        tokens: list[T_Token] = []
-        for aLine in self._text.splitlines():
+        tBlocks: list[T_Block] = [B_EMPTY]
+        for bLine in text.splitlines():
+            aLine = bLine.replace("\uffff", "")  # Remove placeholder characters
             sLine = aLine.strip().lower()
 
             # Check for blank lines
-            if len(sLine) == 0:
-                tokens.append((
-                    self.T_EMPTY, nHead, "", [], self.A_NONE
-                ))
-                if self._keepMD:
-                    tmpMarkdown.append("\n")
-
+            if not sLine:
+                tBlocks.append(B_EMPTY)
                 continue
 
-            if breakNext:
-                sAlign = self.A_PBB
-                breakNext = False
+            if self._breakNext:
+                tStyle = BlockFmt.PBB
+                self._breakNext = False
             else:
-                sAlign = self.A_NONE
+                tStyle = BlockFmt.NONE
 
             # Check Line Format
             # =================
@@ -574,24 +569,24 @@ class Tokenizer(ABC):
                 # therefore proceed to check other formats.
 
                 if sLine in ("[newpage]", "[new page]"):
-                    breakNext = True
+                    self._breakNext = True
                     continue
 
                 elif sLine == "[vspace]":
-                    tokens.append(
-                        (self.T_SKIP, nHead, "", [], sAlign)
+                    tBlocks.append(
+                        (BlockTyp.SKIP, "", "", [], tStyle)
                     )
                     continue
 
                 elif sLine.startswith("[vspace:") and sLine.endswith("]"):
                     nSkip = checkInt(sLine[8:-1], 0)
                     if nSkip >= 1:
-                        tokens.append(
-                            (self.T_SKIP, nHead, "", [], sAlign)
+                        tBlocks.append(
+                            (BlockTyp.SKIP, "", "", [], tStyle)
                         )
                     if nSkip > 1:
-                        tokens += (nSkip - 1) * [
-                            (self.T_SKIP, nHead, "", [], self.A_NONE)
+                        tBlocks += (nSkip - 1) * [
+                            (BlockTyp.SKIP, "", "", [], BlockFmt.NONE)
                         ]
                     continue
 
@@ -605,32 +600,24 @@ class Tokenizer(ABC):
                     continue
 
                 cStyle, cKey, cText, _, _ = processComment(aLine)
-                if cStyle == nwComment.SYNOPSIS:
-                    tLine, tFmt = self._extractFormats(cText)
-                    tokens.append((
-                        self.T_SYNOPSIS, nHead, tLine, tFmt, sAlign
-                    ))
-                    if self._doSynopsis and self._keepMD:
-                        tmpMarkdown.append(f"{aLine}\n")
-                elif cStyle == nwComment.SHORT:
-                    tLine, tFmt = self._extractFormats(cText)
-                    tokens.append((
-                        self.T_SHORT, nHead, tLine, tFmt, sAlign
+                if cStyle in (nwComment.SYNOPSIS, nwComment.SHORT) and not self._doSynopsis:
+                    continue
+                if cStyle == nwComment.PLAIN and not self._doComments:
+                    continue
+
+                if doJustify and not tStyle & BlockFmt.ALIGNED:
+                    tStyle |= BlockFmt.JUSTIFY
+
+                if cStyle in (nwComment.SYNOPSIS, nwComment.SHORT, nwComment.PLAIN):
+                    bStyle = COMMENT_STYLE[cStyle]
+                    tLine, tFmt = self._formatComment(bStyle, cKey, cText)
+                    tBlocks.append((
+                        BlockTyp.COMMENT, "", tLine, tFmt, tStyle
                     ))
-                    if self._doSynopsis and self._keepMD:
-                        tmpMarkdown.append(f"{aLine}\n")
+
                 elif cStyle == nwComment.FOOTNOTE:
-                    tLine, tFmt = self._extractFormats(cText, skip=self.FMT_FNOTE)
+                    tLine, tFmt = self._extractFormats(cText, skip=TextFmt.FNOTE)
                     self._footnotes[f"{tHandle}:{cKey}"] = (tLine, tFmt)
-                    if self._keepMD:
-                        tmpMarkdown.append(f"{aLine}\n")
-                else:
-                    tLine, tFmt = self._extractFormats(cText)
-                    tokens.append((
-                        self.T_COMMENT, nHead, tLine, tFmt, sAlign
-                    ))
-                    if self._doComments and self._keepMD:
-                        tmpMarkdown.append(f"{aLine}\n")
 
             elif aLine.startswith("@"):
                 # Keywords
@@ -638,16 +625,12 @@ class Tokenizer(ABC):
                 # Only valid keyword lines are parsed, and any ignored keywords
                 # are automatically skipped.
 
-                valid, bits, _ = self._project.index.scanThis(aLine)
-                if (
-                    valid and bits and bits[0] in nwLabels.KEY_NAME
-                    and bits[0] not in self._skipKeywords
-                ):
-                    tokens.append((
-                        self.T_KEYWORD, nHead, aLine[1:].strip(), [], sAlign
-                    ))
-                    if self._doKeywords and self._keepMD:
-                        tmpMarkdown.append(f"{aLine}\n")
+                if self._doKeywords:
+                    tTag, tLine, tFmt = self._formatMeta(aLine)
+                    if tLine:
+                        tBlocks.append((
+                            BlockTyp.KEYWORD, tTag[1:], tLine, tFmt, tStyle
+                        ))
 
             elif aLine.startswith(("# ", "#! ")):
                 # Title or Partition Headings
@@ -662,28 +645,26 @@ class Tokenizer(ABC):
 
                 nHead += 1
                 tText = aLine[2:].strip()
-                tType = self.T_HEAD1 if isPlain else self.T_TITLE
-                tStyle = self.A_NONE if isPlain else (self.A_PBB | self.A_CENTRE)
-                sHide = self._hideTitle if isPlain else False
-                if self._isNovel:
+                tType = BlockTyp.HEAD1 if isPlain else BlockTyp.TITLE
+                sHide = self._hidePart if isPlain else False
+                if not (isPlain or isNovel and sHide):
+                    tStyle |= self._titleStyle
+                if isNovel:
                     if sHide:
                         tText = ""
-                        tType = self.T_EMPTY
-                        tStyle = self.A_NONE
+                        tType = BlockTyp.EMPTY
                     elif isPlain:
-                        tText = self._hFormatter.apply(self._fmtTitle, tText, nHead)
-                        tStyle = self._titleStyle
+                        tText = self._hFormatter.apply(self._fmtPart, tText, nHead)
+                        tStyle |= self._partStyle
                     if isPlain:
                         self._hFormatter.resetScene()
                     else:
                         self._hFormatter.resetAll()
                     self._noSep = True
 
-                tokens.append((
-                    tType, nHead, tText, [], tStyle
+                tBlocks.append((
+                    tType, f"{tHandle}:T{nHead:04d}", tText, [], tStyle
                 ))
-                if self._keepMD:
-                    tmpMarkdown.append(f"{aLine}\n")
 
             elif aLine.startswith(("## ", "##! ")):
                 # (Unnumbered) Chapter Headings
@@ -698,27 +679,24 @@ class Tokenizer(ABC):
 
                 nHead += 1
                 tText = aLine[3:].strip()
-                tType = self.T_HEAD2
-                tStyle = self.A_NONE
+                tType = BlockTyp.HEAD2
                 sHide = self._hideChapter if isPlain else self._hideUnNum
                 tFormat = self._fmtChapter if isPlain else self._fmtUnNum
-                if self._isNovel:
+                if isNovel:
                     if isPlain:
                         self._hFormatter.incChapter()
                     if sHide:
                         tText = ""
-                        tType = self.T_EMPTY
+                        tType = BlockTyp.EMPTY
                     else:
                         tText = self._hFormatter.apply(tFormat, tText, nHead)
-                        tStyle = self._chapterStyle
+                        tStyle |= self._chapterStyle
                     self._hFormatter.resetScene()
                     self._noSep = True
 
-                tokens.append((
-                    tType, nHead, tText, [], tStyle
+                tBlocks.append((
+                    tType, f"{tHandle}:T{nHead:04d}", tText, [], tStyle
                 ))
-                if self._keepMD:
-                    tmpMarkdown.append(f"{aLine}\n")
 
             elif aLine.startswith(("### ", "###! ")):
                 # (Alternative) Scene Headings
@@ -735,31 +713,28 @@ class Tokenizer(ABC):
 
                 nHead += 1
                 tText = aLine[4:].strip()
-                tType = self.T_HEAD3
-                tStyle = self.A_NONE
+                tType = BlockTyp.HEAD3
                 sHide = self._hideScene if isPlain else self._hideHScene
                 tFormat = self._fmtScene if isPlain else self._fmtHScene
-                if self._isNovel:
+                if isNovel:
                     self._hFormatter.incScene()
                     if sHide:
                         tText = ""
-                        tType = self.T_EMPTY
+                        tType = BlockTyp.EMPTY
                     else:
                         tText = self._hFormatter.apply(tFormat, tText, nHead)
-                        tStyle = self._sceneStyle
+                        tStyle |= self._sceneStyle
                         if tText == "":  # Empty Format
-                            tType = self.T_EMPTY if self._noSep else self.T_SKIP
+                            tType = BlockTyp.EMPTY if self._noSep else BlockTyp.SKIP
                         elif tText == tFormat:  # Static Format
                             tText = "" if self._noSep else tText
-                            tType = self.T_EMPTY if self._noSep else self.T_SEP
-                            tStyle = self.A_NONE if self._noSep else self.A_CENTRE
+                            tType = BlockTyp.EMPTY if self._noSep else BlockTyp.SEP
+                            tStyle |= BlockFmt.NONE if self._noSep else BlockFmt.CENTRE
                     self._noSep = False
 
-                tokens.append((
-                    tType, nHead, tText, [], tStyle
+                tBlocks.append((
+                    tType, f"{tHandle}:T{nHead:04d}", tText, [], tStyle
                 ))
-                if self._keepMD:
-                    tmpMarkdown.append(f"{aLine}\n")
 
             elif aLine.startswith("#### "):
                 # Section Headings
@@ -771,25 +746,22 @@ class Tokenizer(ABC):
 
                 nHead += 1
                 tText = aLine[5:].strip()
-                tType = self.T_HEAD4
-                tStyle = self.A_NONE
-                if self._isNovel:
+                tType = BlockTyp.HEAD4
+                if isNovel:
                     if self._hideSection:
                         tText = ""
-                        tType = self.T_EMPTY
+                        tType = BlockTyp.EMPTY
                     else:
                         tText = self._hFormatter.apply(self._fmtSection, tText, nHead)
                         if tText == "":  # Empty Format
-                            tType = self.T_SKIP
+                            tType = BlockTyp.SKIP
                         elif tText == self._fmtSection:  # Static Format
-                            tType = self.T_SEP
-                            tStyle = self.A_CENTRE
+                            tType = BlockTyp.SEP
+                            tStyle |= BlockFmt.CENTRE
 
-                tokens.append((
-                    tType, nHead, tText, [], tStyle
+                tBlocks.append((
+                    tType, f"{tHandle}:T{nHead:04d}", tText, [], tStyle
                 ))
-                if self._keepMD:
-                    tmpMarkdown.append(f"{aLine}\n")
 
             else:
                 # Text Lines
@@ -805,135 +777,133 @@ class Tokenizer(ABC):
                 alnRight = False
                 indLeft = False
                 indRight = False
-                if aLine.startswith(">>"):
+                if bLine.startswith(">>"):
                     alnRight = True
-                    aLine = aLine[2:].lstrip(" ")
-                elif aLine.startswith(">"):
+                    bLine = bLine[2:].lstrip(" ")
+                elif bLine.startswith(">"):
                     indLeft = True
-                    aLine = aLine[1:].lstrip(" ")
+                    bLine = bLine[1:].lstrip(" ")
 
-                if aLine.endswith("<<"):
+                if bLine.endswith("<<"):
                     alnLeft = True
-                    aLine = aLine[:-2].rstrip(" ")
-                elif aLine.endswith("<"):
+                    bLine = bLine[:-2].rstrip(" ")
+                elif bLine.endswith("<"):
                     indRight = True
-                    aLine = aLine[:-1].rstrip(" ")
+                    bLine = bLine[:-1].rstrip(" ")
 
                 if alnLeft and alnRight:
-                    sAlign |= self.A_CENTRE
+                    tStyle |= BlockFmt.CENTRE
                 elif alnLeft:
-                    sAlign |= self.A_LEFT
+                    tStyle |= BlockFmt.LEFT
                 elif alnRight:
-                    sAlign |= self.A_RIGHT
+                    tStyle |= BlockFmt.RIGHT
 
                 if indLeft:
-                    sAlign |= self.A_IND_L
+                    tStyle |= BlockFmt.IND_L
                 if indRight:
-                    sAlign |= self.A_IND_R
+                    tStyle |= BlockFmt.IND_R
 
                 # Process formats
-                tLine, tFmt = self._extractFormats(aLine)
-                tokens.append((
-                    self.T_TEXT, nHead, tLine, tFmt, sAlign
+                tLine, tFmt = self._extractFormats(bLine, hDialog=isNovel)
+                tBlocks.append((
+                    BlockTyp.TEXT, "", tLine, tFmt, tStyle
                 ))
-                if self._keepMD:
-                    tmpMarkdown.append(f"{aLine}\n")
 
         # If we have content, turn off the first page flag
-        if self._isFirst and tokens:
+        if self._isFirst and len(tBlocks) > 1:
             self._isFirst = False  # First document has been processed
 
-            # Make sure the token array doesn't start with a page break
-            # on the very first page, adding a blank first page.
-            if tokens[0][4] & self.A_PBB:
-                cToken = tokens[0]
-                tokens[0] = (
-                    cToken[0], cToken[1], cToken[2], cToken[3], cToken[4] & ~self.A_PBB
-                )
+            # Make sure the blocks array doesn't start with a page break
+            # on the very first block, adding a blank first page.
+            for n, cBlock in enumerate(tBlocks):
+                if cBlock[0] != BlockTyp.EMPTY:
+                    if cBlock[4] & BlockFmt.PBB:
+                        tBlocks[n] = (
+                            cBlock[0], cBlock[1], cBlock[2], cBlock[3], cBlock[4] & ~BlockFmt.PBB
+                        )
+                    break
 
         # Always add an empty line at the end of the file
-        tokens.append((
-            self.T_EMPTY, nHead, "", [], self.A_NONE
-        ))
-        if self._keepMD:
-            tmpMarkdown.append("\n")
-            self._markdown.append("".join(tmpMarkdown))
+        tBlocks.append(B_EMPTY)
 
         # Second Pass
         # ===========
         # This second pass strips away consecutive blank lines, and
         # combines consecutive text lines into the same paragraph.
         # It also ensures that there isn't paragraph spacing between
-        # meta data lines for formats that has spacing.
-
-        self._tokens = []
-        pToken: T_Token = (self.T_EMPTY, 0, "", [], self.A_NONE)
-        nToken: T_Token = (self.T_EMPTY, 0, "", [], self.A_NONE)
+        # meta data lines for formats that have spacing.
 
-        lineSep = "\n" if self._keepBreaks else " "
-        pLines: list[T_Token] = []
+        lineSep = "\n" if keepBreaks else " "
 
-        tCount = len(tokens)
-        for n, cToken in enumerate(tokens):
+        pLines: list[T_Block] = []
+        sBlocks: list[T_Block] = []
+        for n, cBlock in enumerate(tBlocks[1:-1], 1):
 
-            if n > 0:
-                pToken = tokens[n-1]  # Look behind
-            if n < tCount - 1:
-                nToken = tokens[n+1]  # Look ahead
+            pBlock = tBlocks[n-1]  # Look behind
+            nBlock = tBlocks[n+1]  # Look ahead
 
-            if cToken[0] in self.L_SKIP_INDENT and not self._indentFirst:
+            if cBlock[0] in SKIP_INDENT and not indentFirst:
                 # Unless the indentFirst flag is set, we set up the next
                 # paragraph to not be indented if we see a block of a
                 # specific type
                 self._noIndent = True
 
-            if cToken[0] == self.T_EMPTY:
+            if cBlock[0] == BlockTyp.EMPTY:
                 # We don't need to keep the empty lines after this pass
                 pass
 
-            elif cToken[0] == self.T_KEYWORD:
+            elif cBlock[0] == BlockTyp.KEYWORD:
                 # Adjust margins for lines in a list of keyword lines
-                aStyle = cToken[4]
-                if pToken[0] == self.T_KEYWORD:
-                    aStyle |= self.A_Z_TOPMRG
-                if nToken[0] == self.T_KEYWORD:
-                    aStyle |= self.A_Z_BTMMRG
-                self._tokens.append((
-                    cToken[0], cToken[1], cToken[2], cToken[3], aStyle
+                aStyle = cBlock[4]
+                if pBlock[0] == BlockTyp.KEYWORD:
+                    aStyle |= BlockFmt.Z_TOP
+                if nBlock[0] == BlockTyp.KEYWORD:
+                    aStyle |= BlockFmt.Z_BTM
+                sBlocks.append((
+                    cBlock[0], cBlock[1], cBlock[2], cBlock[3], aStyle
                 ))
 
-            elif cToken[0] == self.T_TEXT:
+            elif cBlock[0] == BlockTyp.TEXT:
                 # Combine lines from the same paragraph
-                pLines.append(cToken)
+                pLines.append(cBlock)
 
-                if nToken[0] != self.T_TEXT:
-                    # Next token is not text, so we add the buffer to tokens
+                if nBlock[0] != BlockTyp.TEXT:
+                    # Next block is not text, so we add the buffer to blocks
                     nLines = len(pLines)
                     cStyle = pLines[0][4]
-                    if self._firstIndent and not (self._noIndent or cStyle & self.M_ALIGNED):
+                    if firstIndent and not (self._noIndent or cStyle & BlockFmt.ALIGNED):
                         # If paragraph indentation is enabled, not temporarily
                         # turned off, and the block is not aligned, we add the
                         # text indentation flag
-                        cStyle |= self.A_IND_T
+                        cStyle |= BlockFmt.IND_T
 
                     if nLines == 1:
-                        # The paragraph contains a single line, so we just
-                        # save that directly to the token list
-                        self._tokens.append((
-                            self.T_TEXT, pLines[0][1], pLines[0][2], pLines[0][3], cStyle
+                        # The paragraph contains a single line, so we just save
+                        # that directly to the blocks list. If justify is
+                        # enabled, and there is no alignment, we apply it.
+                        if doJustify and not cStyle & BlockFmt.ALIGNED:
+                            cStyle |= BlockFmt.JUSTIFY
+
+                        pTxt = pLines[0][2].replace("\uffff", "\n")
+                        sBlocks.append((
+                            BlockTyp.TEXT, pLines[0][1], pTxt, pLines[0][3], cStyle
                         ))
+
                     elif nLines > 1:
                         # The paragraph contains multiple lines, so we need to
                         # join them according to the line break policy, and
                         # recompute all the formatting markers
                         tTxt = ""
                         tFmt: T_Formats = []
-                        for aToken in pLines:
+                        for aBlock in pLines:
                             tLen = len(tTxt)
-                            tTxt += f"{aToken[2]}{lineSep}"
-                            tFmt.extend((p+tLen, fmt, key) for p, fmt, key in aToken[3])
-                        self._tokens.append((
-                            self.T_TEXT, pLines[0][1], tTxt[:-1], tFmt, cStyle
+                            tTxt += f"{aBlock[2]}{lineSep}"
+                            tFmt.extend((p+tLen, fmt, key) for p, fmt, key in aBlock[3])
+                            cStyle |= aBlock[4]
+
+                        pTxt = tTxt[:-1].replace("\uffff", "\n")
+                        sBlocks.append((
+                            BlockTyp.TEXT, pLines[0][1], pTxt, tFmt, cStyle
                         ))
 
                     # Reset buffer and make sure text indent is on for next pass
@@ -941,50 +911,50 @@ class Tokenizer(ABC):
                     self._noIndent = False
 
             else:
-                self._tokens.append(cToken)
+                sBlocks.append(cBlock)
+
+        self._blocks = sBlocks
 
         return
 
     def buildOutline(self) -> None:
         """Build an outline of the text up to level 3 headings."""
-        tHandle = self._handle or ""
         isNovel = self._isNovel
-        for tType, nHead, tText, _, _ in self._tokens:
-            if tType == self.T_TITLE:
+        for tType, tKey, tText, _, _ in self._blocks:
+            if tType == BlockTyp.TITLE:
                 prefix = "TT"
-            elif tType == self.T_HEAD1:
+            elif tType == BlockTyp.HEAD1:
                 prefix = "PT" if isNovel else "H1"
-            elif tType == self.T_HEAD2:
+            elif tType == BlockTyp.HEAD2:
                 prefix = "CH" if isNovel else "H2"
-            elif tType == self.T_HEAD3:
+            elif tType == BlockTyp.HEAD3:
                 prefix = "SC" if isNovel else "H3"
             else:
                 continue
 
-            key = f"{tHandle}:T{nHead:04d}"
             text = tText.replace(nwHeadFmt.BR, " ").replace("&amp;", "&")
-            self._outline[key] = f"{prefix}|{text}"
+            self._outline[tKey] = f"{prefix}|{text}"
 
         return
 
     def countStats(self) -> None:
         """Count stats on the tokenized text."""
-        titleCount = self._counts.get("titleCount", 0)
-        paragraphCount = self._counts.get("paragraphCount", 0)
+        titleCount = self._counts.get(nwStats.TITLES, 0)
+        paragraphCount = self._counts.get(nwStats.PARAGRAPHS, 0)
 
-        allWords = self._counts.get("allWords", 0)
-        textWords = self._counts.get("textWords", 0)
-        titleWords = self._counts.get("titleWords", 0)
+        allWords = self._counts.get(nwStats.WORDS_ALL, 0)
+        textWords = self._counts.get(nwStats.WORDS_TEXT, 0)
+        titleWords = self._counts.get(nwStats.WORDS_TITLE, 0)
 
-        allChars = self._counts.get("allChars", 0)
-        textChars = self._counts.get("textChars", 0)
-        titleChars = self._counts.get("titleChars", 0)
+        allChars = self._counts.get(nwStats.CHARS_ALL, 0)
+        textChars = self._counts.get(nwStats.CHARS_TEXT, 0)
+        titleChars = self._counts.get(nwStats.CHARS_TITLE, 0)
 
-        allWordChars = self._counts.get("allWordChars", 0)
-        textWordChars = self._counts.get("textWordChars", 0)
-        titleWordChars = self._counts.get("titleWordChars", 0)
+        allWordChars = self._counts.get(nwStats.WCHARS_ALL, 0)
+        textWordChars = self._counts.get(nwStats.WCHARS_TEXT, 0)
+        titleWordChars = self._counts.get(nwStats.WCHARS_TITLE, 0)
 
-        for tType, _, tText, _, _ in self._tokens:
+        for tType, _, tText, _, _ in self._blocks:
             tText = tText.replace(nwUnicode.U_ENDASH, " ")
             tText = tText.replace(nwUnicode.U_EMDASH, " ")
 
@@ -993,7 +963,7 @@ class Tokenizer(ABC):
             nChars = len(tText)
             nWChars = len("".join(tWords))
 
-            if tType == self.T_TEXT:
+            if tType == BlockTyp.TEXT:
                 tPWords = tText.split()
                 nPWords = len(tPWords)
                 nPChars = len(tText)
@@ -1007,7 +977,7 @@ class Tokenizer(ABC):
                 allWordChars += nPWChars
                 textWordChars += nPWChars
 
-            elif tType in self.L_HEADINGS:
+            elif tType in HEADINGS:
                 titleCount += 1
                 allWords += nWords
                 titleWords += nWords
@@ -1016,143 +986,170 @@ class Tokenizer(ABC):
                 titleChars += nChars
                 titleWordChars += nWChars
 
-            elif tType == self.T_SEP:
+            elif tType == BlockTyp.SEP:
                 allWords += nWords
                 allChars += nChars
                 allWordChars += nWChars
 
-            elif tType == self.T_SYNOPSIS and self._doSynopsis:
-                text = "{0}: {1}".format(self._localLookup("Synopsis"), tText)
-                words = text.split()
-                allWords += len(words)
-                allChars += len(text)
-                allWordChars += len("".join(words))
-
-            elif tType == self.T_SHORT and self._doSynopsis:
-                text = "{0}: {1}".format(self._localLookup("Short Description"), tText)
-                words = text.split()
+            elif tType in (BlockTyp.COMMENT, BlockTyp.KEYWORD):
+                words = tText.split()
                 allWords += len(words)
-                allChars += len(text)
+                allChars += len(tText)
                 allWordChars += len("".join(words))
 
-            elif tType == self.T_COMMENT and self._doComments:
-                text = "{0}: {1}".format(self._localLookup("Comment"), tText)
-                words = text.split()
-                allWords += len(words)
-                allChars += len(text)
-                allWordChars += len("".join(words))
-
-            elif tType == self.T_KEYWORD and self._doKeywords:
-                valid, bits, _ = self._project.index.scanThis("@"+tText)
-                if valid and bits:
-                    key = self._localLookup(nwLabels.KEY_NAME[bits[0]])
-                    text = "{0}: {1}".format(key, ", ".join(bits[1:]))
-                    words = text.split()
-                    allWords += len(words)
-                    allChars += len(text)
-                    allWordChars += len("".join(words))
-
-        self._counts["titleCount"] = titleCount
-        self._counts["paragraphCount"] = paragraphCount
+        self._counts[nwStats.TITLES] = titleCount
+        self._counts[nwStats.PARAGRAPHS] = paragraphCount
 
-        self._counts["allWords"] = allWords
-        self._counts["textWords"] = textWords
-        self._counts["titleWords"] = titleWords
+        self._counts[nwStats.WORDS_ALL] = allWords
+        self._counts[nwStats.WORDS_TEXT] = textWords
+        self._counts[nwStats.WORDS_TITLE] = titleWords
 
-        self._counts["allChars"] = allChars
-        self._counts["textChars"] = textChars
-        self._counts["titleChars"] = titleChars
+        self._counts[nwStats.CHARS_ALL] = allChars
+        self._counts[nwStats.CHARS_TEXT] = textChars
+        self._counts[nwStats.CHARS_TITLE] = titleChars
 
-        self._counts["allWordChars"] = allWordChars
-        self._counts["textWordChars"] = textWordChars
-        self._counts["titleWordChars"] = titleWordChars
+        self._counts[nwStats.WCHARS_ALL] = allWordChars
+        self._counts[nwStats.WCHARS_TEXT] = textWordChars
+        self._counts[nwStats.WCHARS_TITLE] = titleWordChars
 
         return
 
-    def saveRawMarkdown(self, path: str | Path) -> None:
-        """Save the raw text to a plain text file."""
-        with open(path, mode="w", encoding="utf-8") as outFile:
-            for nwdPage in self._markdown:
-                outFile.write(nwdPage)
-        return
-
-    def saveRawMarkdownJSON(self, path: str | Path) -> None:
-        """Save the raw text to a JSON file."""
-        timeStamp = time()
-        data = {
-            "meta": {
-                "projectName": self._project.data.name,
-                "novelAuthor": self._project.data.author,
-                "buildTime": int(timeStamp),
-                "buildTimeStr": formatTimeStamp(timeStamp),
-            },
-            "text": {
-                "nwd": [page.rstrip("\n").split("\n") for page in self._markdown],
-            }
-        }
-        with open(path, mode="w", encoding="utf-8") as fObj:
-            json.dump(data, fObj, indent=2)
-        return
-
     ##
     #  Internal Functions
     ##
 
-    def _extractFormats(self, text: str, skip: int = 0) -> tuple[str, T_Formats]:
-        """Extract format markers from a text paragraph."""
+    def _formatInt(self, value: int) -> str:
+        """Return a localised integer."""
+        return self._dLocale.toString(value)
+
+    def _formatComment(self, style: ComStyle, key: str, text: str) -> tuple[str, T_Formats]:
+        """Apply formatting to comments and notes."""
+        tTxt, tFmt = self._extractFormats(text)
+        tFmt.insert(0, (0, TextFmt.COL_B, style.textClass))
+        tFmt.append((len(tTxt), TextFmt.COL_E, ""))
+        if label := (self._localLookup(style.label) + (f" ({key})" if key else "")).strip():
+            shift = len(label) + 2
+            tTxt = f"{label}: {tTxt}"
+            rFmt = [(0, TextFmt.B_B, ""), (shift - 1, TextFmt.B_E, "")]
+            if style.labelClass:
+                rFmt.insert(1, (0, TextFmt.COL_B, style.labelClass))
+                rFmt.insert(2, (shift - 1, TextFmt.COL_E, ""))
+            rFmt.extend((p + shift, f, d) for p, f, d in tFmt)
+        return tTxt, rFmt
+
+    def _formatMeta(self, text: str) -> tuple[str, str, T_Formats]:
+        """Apply formatting to a meta data line."""
+        tag = ""
+        txt = []
+        fmt = []
+        valid, bits, _ = self._project.index.scanThis(text)
+        if valid and bits and bits[0] in nwLabels.KEY_NAME and bits[0] not in self._skipKeywords:
+            tag = bits[0]
+            pos = 0
+            lbl = f"{self._localLookup(nwLabels.KEY_NAME[tag])}:"
+            end = len(lbl)
+            fmt = [
+                (pos, TextFmt.B_B, ""),
+                (pos, TextFmt.COL_B, "keyword"),
+                (end, TextFmt.COL_E, ""),
+                (end, TextFmt.B_E, ""),
+            ]
+            txt = [lbl, " "]
+            pos = end + 1
+
+            if (num := len(bits)) > 1:
+                if bits[0] == nwKeyWords.TAG_KEY:
+                    one, two = self._project.index.parseValue(bits[1])
+                    end = pos + len(one)
+                    fmt.append((pos, TextFmt.COL_B, "tag"))
+                    fmt.append((pos, TextFmt.ANM_B, f"tag_{one}".lower()))
+                    fmt.append((end, TextFmt.ANM_E, ""))
+                    fmt.append((end, TextFmt.COL_E, ""))
+                    txt.append(one)
+                    pos = end
+                    if two:
+                        txt.append(" | ")
+                        pos += 3
+                        end = pos + len(two)
+                        fmt.append((pos, TextFmt.COL_B, "optional"))
+                        fmt.append((end, TextFmt.COL_E, ""))
+                        txt.append(two)
+                        pos = end
+                else:
+                    for n, bit in enumerate(bits[1:], 2):
+                        end = pos + len(bit)
+                        fmt.append((pos, TextFmt.COL_B, "tag"))
+                        fmt.append((pos, TextFmt.ARF_B, f"#tag_{bit}".lower()))
+                        fmt.append((end, TextFmt.ARF_E, ""))
+                        fmt.append((end, TextFmt.COL_E, ""))
+                        txt.append(bit)
+                        pos = end
+                        if n < num:
+                            txt.append(", ")
+                            pos += 2
+
+        return tag, "".join(txt), fmt
+
+    def _extractFormats(
+        self, text: str, skip: int = 0, hDialog: bool = False
+    ) -> tuple[str, T_Formats]:
+        """Extract format markers from a text paragraph. In order to
+        also process dialogue highlighting, the hDialog flag must be set
+        to True. See issues #2011 and #2013.
+        """
         temp: list[tuple[int, int, int, str]] = []
 
         # Match Markdown
         for regEx, fmts in self._rxMarkdown:
-            rxItt = regEx.globalMatch(text, 0)
-            while rxItt.hasNext():
-                rxMatch = rxItt.next()
+            for res in regEx.finditer(text):
                 temp.extend(
-                    (rxMatch.capturedStart(n), rxMatch.capturedLength(n), fmt, "")
+                    (res.start(n), res.end(n), fmt, "")
                     for n, fmt in enumerate(fmts) if fmt > 0
                 )
 
+        # Match URLs
+        for res in REGEX_PATTERNS.url.finditer(text):
+            temp.append((res.start(0), 0, TextFmt.HRF_B, res.group(0)))
+            temp.append((res.end(0), 0, TextFmt.HRF_E, ""))
+
         # Match Shortcodes
-        rxItt = self._rxShortCodes.globalMatch(text, 0)
-        while rxItt.hasNext():
-            rxMatch = rxItt.next()
+        for res in REGEX_PATTERNS.shortcodePlain.finditer(text):
             temp.append((
-                rxMatch.capturedStart(1),
-                rxMatch.capturedLength(1),
-                self._shortCodeFmt.get(rxMatch.captured(1).lower(), 0),
+                res.start(1), res.end(1),
+                self._shortCodeFmt.get(res.group(1).lower(), 0),
                 "",
             ))
 
         # Match Shortcode w/Values
-        rxItt = self._rxShortCodeVals.globalMatch(text, 0)
         tHandle = self._handle or ""
-        while rxItt.hasNext():
-            rxMatch = rxItt.next()
-            kind = self._shortCodeVals.get(rxMatch.captured(1).lower(), 0)
+        for res in REGEX_PATTERNS.shortcodeValue.finditer(text):
+            kind = self._shortCodeVals.get(res.group(1).lower(), 0)
             temp.append((
-                rxMatch.capturedStart(0),
-                rxMatch.capturedLength(0),
-                self.FMT_STRIP if kind == skip else kind,
-                f"{tHandle}:{rxMatch.captured(2)}",
+                res.start(0), res.end(0),
+                TextFmt.STRIP if kind == skip else kind,
+                f"{tHandle}:{res.group(2)}",
             ))
 
         # Match Dialogue
-        if self._rxDialogue:
-            for regEx, fmtB, fmtE in self._rxDialogue:
-                rxItt = regEx.globalMatch(text, 0)
-                while rxItt.hasNext():
-                    rxMatch = rxItt.next()
-                    temp.append((rxMatch.capturedStart(0), 0, fmtB, ""))
-                    temp.append((rxMatch.capturedEnd(0), 0, fmtE, ""))
+        if self._hlightDialog and hDialog:
+            if self._dialogParser.enabled:
+                for pos, end in self._dialogParser(text):
+                    temp.append((pos, 0, TextFmt.COL_B, "dialog"))
+                    temp.append((end, 0, TextFmt.COL_E, ""))
+            if self._rxAltDialog:
+                for res in self._rxAltDialog.finditer(text):
+                    temp.append((res.start(0), 0, TextFmt.COL_B, "altdialog"))
+                    temp.append((res.end(0), 0, TextFmt.COL_E, ""))
 
         # Post-process text and format
         result = text
         formats = []
-        for pos, n, fmt, key in reversed(sorted(temp, key=lambda x: x[0])):
+        for pos, end, fmt, meta in reversed(sorted(temp, key=lambda x: x[0])):
             if fmt > 0:
-                result = result[:pos] + result[pos+n:]
-                formats = [(p-n, f, k) for p, f, k in formats]
-                formats.insert(0, (pos, fmt, key))
+                if end > pos:
+                    result = result[:pos] + result[end:]
+                    formats = [(p+pos-end if p > pos else p, f, m) for p, f, m in formats]
+                formats.insert(0, (pos, fmt, meta))
 
         return result, formats
 
@@ -1198,6 +1195,7 @@ class HeadingFormatter:
     def apply(self, hFormat: str, text: str, nHead: int) -> str:
         """Apply formatting to a specific heading."""
         hFormat = hFormat.replace(nwHeadFmt.TITLE, text)
+        hFormat = hFormat.replace(nwHeadFmt.BR, "\n")
         hFormat = hFormat.replace(nwHeadFmt.CH_NUM, str(self._chCount))
         hFormat = hFormat.replace(nwHeadFmt.SC_NUM, str(self._scChCount))
         hFormat = hFormat.replace(nwHeadFmt.SC_ABS, str(self._scAbsCount))