Improved regex, highlighting and HTML conversion of bold and italic text. Reverting to old syntax for italic.

nw/constants/constants.py

@@ -37,10 +37,9 @@ class nwConst():
 
 class nwRegEx():
 
-    FMT_B  = r"(?<![\w|\*|_|\\])([\*|_]{2})(?!\s|\*|_)(.+?)(?<![\s|\\])(\1)(?!\w)"
+    FMT_I  = r"(?<![\w\\])(_)(?![\s_])(.+?)(?<![\s\\])(\1)(?!\w)"
-    FMT_I  = r"(?<![\w|\*|_|\\])([\*|_])(?!\s|\*|_)(.+?)(?<![\s|\\])(\1)(?!\w)"
+    FMT_B  = r"(?<![\w\\])([\*]{2})(?![\s\*])(.+?)(?<![\s\\])(\1)(?!\w)"
-    FMT_BI = r"(?<![\w|\*|\\])([\*]{3})(?!\s|\*)(.+?)(?<![\s|\\])(\1)(?!\w)"
+    FMT_ST = r"(?<![\w\\])([~]{2})(?![\s~])(.+?)(?<![\s\\])(\1)(?!\w)"
-    FMT_ST = r"(?<![\w|~|\\])([~]{2})(?!\s|~)(.+?)(?<![\s|\\])(\1)(?!\w)"
 
 # END Class nwRegEx
 

nw/core/tohtml.py

@@ -107,6 +107,7 @@ class ToHtml(Tokenizer):
         """Reverse the html entities replacement on the markdown text.
         Otherwise, all the &something; bits will also be in there.
         """
+        Tokenizer.doPostProcessing(self)
         if self.genMode == self.M_PREVIEW:
             # Doesn't matter for preview as we don't use the markdown
             return
@@ -125,8 +126,6 @@ class ToHtml(Tokenizer):
                 self.FMT_B_E : "</b>",
                 self.FMT_I_B : "<i>",
                 self.FMT_I_E : "</i>",
-                self.FMT_S_B : "<b><i>",
-                self.FMT_S_E : "</i></b>",
                 self.FMT_D_B : "<span style='text-decoration: line-through;'>",
                 self.FMT_D_E : "</span>",
             }
@@ -136,8 +135,6 @@ class ToHtml(Tokenizer):
                 self.FMT_B_E : "</strong>",
                 self.FMT_I_B : "<em>",
                 self.FMT_I_E : "</em>",
-                self.FMT_S_B : "<strong><em>",
-                self.FMT_S_E : "</em></strong>",
                 self.FMT_D_B : "<del>",
                 self.FMT_D_E : "</del>",
             }

nw/core/tokenizer.py

@@ -44,10 +44,8 @@ class Tokenizer():
     FMT_B_E    =  2 # End bold
     FMT_I_B    =  3 # Begin italics
     FMT_I_E    =  4 # End italics
-    FMT_S_B    =  5 # Begin bold italic
+    FMT_D_B    =  5 # Begin strikeout
-    FMT_S_E    =  6 # End bold italic
+    FMT_D_E    =  6 # End strikeout
-    FMT_D_B    =  7 # Begin strikeout
-    FMT_D_E    =  8 # End strikeout
 
     T_EMPTY    =  1 # Empty line (new paragraph)
     T_SYNOPSIS =  2 # Synopsis comment
@@ -269,7 +267,6 @@ class Tokenizer():
     def doAutoReplace(self):
         """Run through the user's auto-replace dictionary.
         """
-
         if len(self.theProject.autoReplace) > 0:
             repDict = {}
             for aKey, aVal in self.theProject.autoReplace.items():
@@ -280,8 +277,20 @@ class Tokenizer():
         return
 
     def doPostProcessing(self):
-        """Do some postprocessing. Overloaded by subclasses.
+        """Do some postprocessing. Overloaded by subclasses. This just
+        does the standard escaped characters.
         """
+        escapeDict = {
+            "\*" : "*",
+            "\~" : "~",
+            "\_" : "_",
+        }
+        escReplace = re.compile(
+            "|".join([re.escape(k) for k in escapeDict.keys()]), flags=re.DOTALL
+        )
+        self.theResult = escReplace.sub(
+            lambda x: escapeDict[x.group(0)], self.theResult
+        )
         return
 
     def tokenizeText(self):
@@ -304,7 +313,6 @@ class Tokenizer():
         rxFormats = [
             (QRegularExpression(nwRegEx.FMT_I),  [None, self.FMT_I_B, None, self.FMT_I_E]),
             (QRegularExpression(nwRegEx.FMT_B),  [None, self.FMT_B_B, None, self.FMT_B_E]),
-            (QRegularExpression(nwRegEx.FMT_BI), [None, self.FMT_S_B, None, self.FMT_S_E]),
             (QRegularExpression(nwRegEx.FMT_ST), [None, self.FMT_D_B, None, self.FMT_D_E]),
         ]
 

nw/core/tools.py

@@ -34,6 +34,10 @@ from os import path, unlink, rmdir
 
 logger = logging.getLogger(__name__)
 
+# =========================================================================== #
+#  Simple Word Counter
+# =========================================================================== #
+
 def countWords(theText):
     """Count words in a piece of text, skipping special syntax and
     comments.
@@ -80,6 +84,10 @@ def countWords(theText):
 
     return charCount, wordCount, paraCount
 
+# =========================================================================== #
+#  Convert an Integer to a Word Number
+# =========================================================================== #
+
 def numberToWord(numVal, theLanguage):
     """Wrapper for converting numbers to words for chapter headings.
     """

nw/gui/dochighlight.py

@@ -107,7 +107,6 @@ class GuiDocHighlighter(QSyntaxHighlighter):
             "header4h"   : self._makeFormat(self.colHeadH, "bold", 1.2),
             "bold"       : self._makeFormat(self.colEmph,  "bold"),
             "italic"     : self._makeFormat(self.colEmph,  "italic"),
-            "bolditalic" : self._makeFormat(self.colEmph, ("bold","italic")),
             "strike"     : self._makeFormat(self.colEmph,  "strike"),
             "trailing"   : self._makeFormat(self.colTrail, "background"),
             "nobreak"    : self._makeFormat(self.colTrail, "background"),
@@ -137,36 +136,6 @@ class GuiDocHighlighter(QSyntaxHighlighter):
             }
         ))
 
-        # Markdown
-        self.hRules.append((
-            nwRegEx.FMT_I, {
-                1 : self.hStyles["hidden"],
-                2 : self.hStyles["italic"],
-                3 : self.hStyles["hidden"],
-            }
-        ))
-        self.hRules.append((
-            nwRegEx.FMT_B, {
-                1 : self.hStyles["hidden"],
-                2 : self.hStyles["bold"],
-                3 : self.hStyles["hidden"],
-            }
-        ))
-        self.hRules.append((
-            nwRegEx.FMT_BI, {
-                1 : self.hStyles["hidden"],
-                2 : self.hStyles["bolditalic"],
-                3 : self.hStyles["hidden"],
-            }
-        ))
-        self.hRules.append((
-            nwRegEx.FMT_ST, {
-                1 : self.hStyles["hidden"],
-                2 : self.hStyles["strike"],
-                3 : self.hStyles["hidden"],
-            }
-        ))
-
         # Quoted Strings
         if self.mainConf.highlightQuotes:
             self.hRules.append((
@@ -185,6 +154,29 @@ class GuiDocHighlighter(QSyntaxHighlighter):
                 }
             ))
 
+        # Markdown
+        self.hRules.append((
+            nwRegEx.FMT_I, {
+                1 : self.hStyles["hidden"],
+                2 : self.hStyles["italic"],
+                3 : self.hStyles["hidden"],
+            }
+        ))
+        self.hRules.append((
+            nwRegEx.FMT_B, {
+                1 : self.hStyles["hidden"],
+                2 : self.hStyles["bold"],
+                3 : self.hStyles["hidden"],
+            }
+        ))
+        self.hRules.append((
+            nwRegEx.FMT_ST, {
+                1 : self.hStyles["hidden"],
+                2 : self.hStyles["strike"],
+                3 : self.hStyles["hidden"],
+            }
+        ))
+
         # Auto-Replace Tags
         self.hRules.append((
             r"<(\S+?)>", {
@@ -302,10 +294,14 @@ class GuiDocHighlighter(QSyntaxHighlighter):
                 rxItt = rX.globalMatch(theText, 0)
                 while rxItt.hasNext():
                     rxMatch = rxItt.next()
-                    for xM in xFmt.keys():
+                    for xM in xFmt:
                         xPos = rxMatch.capturedStart(xM)
                         xLen = rxMatch.capturedLength(xM)
-                        self.setFormat(xPos, xLen, xFmt[xM])
+                        for x in range(xPos, xPos+xLen):
+                            spFmt = self.format(x)
+                            if spFmt != self.hStyles["hidden"]:
+                                spFmt.merge(xFmt[xM])
+                                self.setFormat(x, 1, spFmt)
 
         if self.theDict is None or not self.spellCheck:
             return
@@ -318,11 +314,11 @@ class GuiDocHighlighter(QSyntaxHighlighter):
                     continue
                 xPos = rxMatch.capturedStart(0)
                 xLen = rxMatch.capturedLength(0)
-                for x in range(xLen):
+                for x in range(xPos, xPos+xLen):
-                    spFmt = self.format(xPos+x)
+                    spFmt = self.format(x)
                     spFmt.setUnderlineColor(self.colSpell)
                     spFmt.setUnderlineStyle(QTextCharFormat.SpellCheckUnderline)
-                    self.setFormat(xPos+x, 1, spFmt)
+                    self.setFormat(x, 1, spFmt)
 
         return