Move word split to regex pattern class

novelwriter/constants.py

@@ -23,8 +23,6 @@ along with this program. If not, see <https://www.gnu.org/licenses/>.
 """
 from __future__ import annotations
 
-from re import UNICODE, compile
-
 from PyQt5.QtCore import QT_TRANSLATE_NOOP, QCoreApplication
 
 from novelwriter.enum import (
@@ -62,16 +60,13 @@ class nwConst:
 
 class nwRegEx:
 
+    WORDS  = r"\b[^\s\-\+\/–—\[\]:]+\b"
     FMT_EI = r"(?<![\w\\])(_)(?![\s_])(.+?)(?<![\s\\])(\1)(?!\w)"
     FMT_EB = r"(?<![\w\\])(\*{2})(?![\s\*])(.+?)(?<![\s\\])(\1)(?!\w)"
     FMT_ST = r"(?<![\w\\])(~{2})(?![\s~])(.+?)(?<![\s\\])(\1)(?!\w)"
     FMT_SC = r"(?i)(?<!\\)(\[[\/\!]?(?:b|i|s|u|m|sup|sub)\])"
     FMT_SV = r"(?i)(?<!\\)(\[(?:footnote):)(.+?)(?<!\\)(\])"
 
-    RX_WORDS = compile(r"\b[^\s\-\+\/–—\[\]:]+\b", UNICODE)
-    RX_FMT_SC = compile(r"(?i)(?<!\\)(\[[\/\!]?(?:b|i|s|u|m|sup|sub)\])", UNICODE)
-    RX_FMT_SV = compile(r"(?i)(?<!\\)(\[(?:footnote):)(.+?)(?<!\\)(\])", UNICODE)
-
 
 class nwShortcode:
 

novelwriter/gui/dochighlight.py

@@ -37,13 +37,17 @@ from PyQt5.QtGui import (
 
 from novelwriter import CONFIG, SHARED
 from novelwriter.common import checkInt
-from novelwriter.constants import nwHeaders, nwRegEx, nwUnicode
+from novelwriter.constants import nwHeaders, nwUnicode
 from novelwriter.core.index import processComment
 from novelwriter.enum import nwComment
 from novelwriter.text.patterns import REGEX_PATTERNS
 
 logger = logging.getLogger(__name__)
 
+RX_WORDS = REGEX_PATTERNS.wordSplit
+RX_FMT_SC = REGEX_PATTERNS.shortcodePlain
+RX_FMT_SV = REGEX_PATTERNS.shortcodeValue
+
 BLOCK_NONE  = 0
 BLOCK_TEXT  = 1
 BLOCK_META  = 2
@@ -479,7 +483,7 @@ class TextBlockData(QTextBlockUserData):
         """
         if "[" in text:
             # Strip shortcodes
-            for rX in [nwRegEx.RX_FMT_SC, nwRegEx.RX_FMT_SV]:
+            for rX in [RX_FMT_SC, RX_FMT_SV]:
                 for match in re.finditer(rX, text[offset:]):
                     iS = match.start(0) + offset
                     iE = match.end(0) + offset
@@ -488,7 +492,7 @@ class TextBlockData(QTextBlockUserData):
 
         self._spellErrors = []
         checker = SHARED.spelling
-        for match in re.finditer(nwRegEx.RX_WORDS, text[offset:].replace("_", " ")):
+        for match in re.finditer(RX_WORDS, text[offset:].replace("_", " ")):
             if (
                 (word := match.group(0))
                 and not (word.isnumeric() or word.isupper() or checker.checkWord(word))

novelwriter/text/patterns.py

@@ -32,12 +32,18 @@ from novelwriter.constants import nwRegEx
 class RegExPatterns:
 
     # Static RegExes
+    _rxWords   = re.compile(nwRegEx.WORDS, re.UNICODE)
     _rxItalic  = re.compile(nwRegEx.FMT_EI, re.UNICODE)
     _rxBold    = re.compile(nwRegEx.FMT_EB, re.UNICODE)
     _rxStrike  = re.compile(nwRegEx.FMT_ST, re.UNICODE)
     _rxSCPlain = re.compile(nwRegEx.FMT_SC, re.UNICODE)
     _rxSCValue = re.compile(nwRegEx.FMT_SV, re.UNICODE)
 
+    @property
+    def wordSplit(self) -> re.Pattern:
+        """Split text into words."""
+        return self._rxWords
+
     @property
     def markdownItalic(self) -> re.Pattern:
         """Markdown italic style."""

tests/test_text/test_text_patterns.py

@@ -40,6 +40,52 @@ def allMatches(regEx: re.Pattern, text: str) -> list[list[str]]:
     return result
 
 
+@pytest.mark.core
+def testTextPatterns_Words():
+    """Test the word split regex."""
+    regEx = REGEX_PATTERNS.wordSplit
+
+    # Spaces
+    assert allMatches(regEx, "one two three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Hyphens
+    assert allMatches(regEx, "one-two-three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Em Dashes
+    assert allMatches(regEx, "one\u2014two\u2014three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Em Dashes
+    assert allMatches(regEx, "one\u2014two\u2014three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Plus
+    assert allMatches(regEx, "one+two+three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Slash
+    assert allMatches(regEx, "one/two/three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Brackets
+    assert allMatches(regEx, "one[two]three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+    # Colon
+    assert allMatches(regEx, "one:two:three") == [
+        [("one", 0, 3)], [("two", 4, 7)], [("three", 8, 13)]
+    ]
+
+
 @pytest.mark.core
 def testTextPatterns_Markdown():
     """Test the markdown pattern regexes."""