From 43fdbf94ad87ea91a173c72688cad70a0a5f1ab4 Mon Sep 17 00:00:00 2001
From: Ryo Nihei <nihei.dev@gmail.com>
Date: Fri, 30 Apr 2021 01:54:02 +0900
Subject: Add character property expression (Meet RL1.2 of UTS #18 partially)

\p{property name=property value} matches a character has the property.
When the property name is General_Category, it can be omitted.
That is, \p{Letter} equals \p{General_Category=Letter}.
Currently, only General_Category is supported.
This feature meets RL1.2 of UTS #18 partially.

RL1.2 Properties: https://unicode.org/reports/tr18/#RL1.2
---
 compiler/ucd_table.go.tmpl | 43 +++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 43 insertions(+)
 create mode 100644 compiler/ucd_table.go.tmpl

(limited to 'compiler/ucd_table.go.tmpl')

diff --git a/compiler/ucd_table.go.tmpl b/compiler/ucd_table.go.tmpl
new file mode 100644
index 0000000..a364191
--- /dev/null
+++ b/compiler/ucd_table.go.tmpl
@@ -0,0 +1,43 @@
+// Code generated by {{ .GeneratorName }}; DO NOT EDIT.
+
+package compiler
+
+import "github.com/nihei9/maleeni/ucd"
+
+// https://www.unicode.org/reports/tr44/#GC_Values_Table
+var compositGeneralCategories = map[string][]string{
+	// Cased_Letter
+	"lc": {"lu", "ll", "lt"},
+	// Letter
+	"l": {"lu", "ll", "lt", "lm", "lo"},
+	// Mark
+	"m": {"mm", "mc", "me"},
+	// Number
+	"n": {"nd", "nl", "no"},
+	// Punctuation
+	"p": {"pc", "pd", "ps", "pi", "pe", "pf", "po"},
+	// Symbol
+	"s": {"sm", "sc", "sk", "so"},
+	// Separator
+	"z": {"zs", "zl", "zp"},
+	// Other
+	"c": {"cc", "cf", "cs", "co", "cn"},
+}
+
+// https://www.unicode.org/Public/13.0.0/ucd/PropertyAliases.txt
+var propertyNameAbbs = map[string]string{
+    "generalcategory": "gc",
+    "gc": "gc",
+}
+
+// https://www.unicode.org/Public/13.0.0/ucd/PropertyValueAliases.txt
+var generalCategoryValueAbbs = map[string]string{ {{ range $long, $abb := .PropertyValueAliases.GeneralCategory }}
+    "{{ $long }}": "{{ $abb }}",{{ end }}
+}
+
+// https://www.unicode.org/Public/13.0.0/ucd/UnicodeData.txt
+var generalCategoryCodePoints = map[string][]*ucd.CodePointRange{ {{ range $propName, $codePoints := .UnicodeData.GeneralCategory }}
+	"{{ $propName }}": { {{ range $codePoints }}
+	   &ucd.CodePointRange{From: rune({{ .From }}), To: rune({{ .To }})},{{ end }}
+	},{{ end }}
+}
-- 
cgit v1.2.3