[gtksourceview] def.lang: Add Unicode identifier category regular expressions

[Christian Hergert <chergert src gnome org>]

commit 3abe31a02169cd3883b6a08dfddf6984177e47ff
Author: Jeffery To <jeffery to gmail com>
Date:   Mon Nov 4 05:52:00 2019 +0800

    def.lang: Add Unicode identifier category regular expressions
    
    This adds regular expressions for the Unicode ID_Start and ID_Continue
    categories to def.lang, so that they may be reused in different language
    definitions.
    
    (These regular expressions are necessary because GRegex does not yet
    support referencing these categories directly, e.g. \p{ID_Start}.)
    
    This also updates python3.lang to use these new regular expressions.

 data/language-specs/def.lang     | 11 +++++++++++
 data/language-specs/python3.lang | 19 ++++---------------
 2 files changed, 15 insertions(+), 15 deletions(-)
---
diff --git a/data/language-specs/def.lang b/data/language-specs/def.lang
index 4bd0c0ca..f95c1f67 100644
--- a/data/language-specs/def.lang
+++ b/data/language-specs/def.lang
@@ -124,6 +124,17 @@
     <!-- $^ never matches. -->
     <define-regex id="never-match">$^</define-regex>
 
+    <!-- Unicode ID_Start and ID_Continue characters
+         Descriptions from https://www.unicode.org/reports/tr31/#Table_Lexical_Classes_for_Identifiers
+         Other_ID_Start and Other_ID_Continue code points from 
https://www.unicode.org/Public/12.1.0/ucd/PropList.txt
+    -->
+    <define-regex id="unicode-id-start" extended="true">
+      [\p{L}\p{Nl}\x{1885}-\x{1886}\x{2118}\x{212E}\x{309B}-\x{309C}]
+    </define-regex>
+    <define-regex id="unicode-id-continue" extended="true">
+      
[\p{L}\p{Nl}\x{1885}-\x{1886}\x{2118}\x{212E}\x{309B}-\x{309C}\p{Mn}\p{Mc}\p{Nd}\p{Pc}\x{00B7}\x{0387}\x{1369}-\x{1371}\x{19DA}]
+    </define-regex>
+
     <define-regex id="decimal" extended="true">
       (?&lt;![\w\.]) ([1-9][0-9]* | 0) (?![\w\.])
     </define-regex>
diff --git a/data/language-specs/python3.lang b/data/language-specs/python3.lang
index 3d3d0f44..dca60eea 100644
--- a/data/language-specs/python3.lang
+++ b/data/language-specs/python3.lang
@@ -50,22 +50,11 @@
   </styles>
 
   <definitions>
-    <!-- https://docs.python.org/3/reference/lexical_analysis.html#identifiers -->
-    <define-regex id="id-start" extended="true">
-      (?:
-        [\p{Lu}\p{Ll}\p{Lt}\p{Lm}\p{Lo}\p{Nl}_] |
-        [\x{1885}-\x{1886}\x{2118}\x{212E}\x{309B}-\x{309C}]  # Other_ID_Start (Unicode 12.1.0)
-      )
-    </define-regex>
-    <define-regex id="id-continue" extended="true">
-      (?:
-        \%{id-start} |
-        [\p{Mn}\p{Mc}\p{Nd}\p{Pc}] |
-        [\x{00B7}\x{0387}\x{1369}-\x{1371}\x{19DA}]  # Other_ID_Continue (Unicode 12.1.0)
-      )
-    </define-regex>
+    <!-- https://docs.python.org/3/reference/lexical_analysis.html#identifiers
+         Underscore ('LOW LINE' U+005F) is a member of \p{Pc}, and so included in def:unicode-id-continue
+    -->
     <define-regex id="identifier" extended="true">
-      (?&gt; \%{id-start} \%{id-continue}* )
+      (?&gt; (?: _ | \%{def:unicode-id-start} ) \%{def:unicode-id-continue}* )
     </define-regex>
     <define-regex id="number">[1-9][0-9]*</define-regex>