re2 1.21.5 → 1.22.0

package/LICENSE

@@ -7,7 +7,7 @@ The text of the BSD license is reproduced below.
 The "New" BSD License:
 **********************
 
-Copyright (c) 2005-2024, Eugene Lazutkin
+Copyright (c) 2005-2025, Eugene Lazutkin
 All rights reserved.
 
 Redistribution and use in source and binary forms, with or without

package/README.md

@@ -351,9 +351,18 @@ console.log('re2_res : ' + re2_res); // prints: re2_res : abc,a,b,c
 
 `RE2` always works in the Unicode mode. See `RE2.unicodeWarningLevel` above for more details on how to control warnings about this feature.
 
+#### Unicode classes
+
+`RE2` supports a subset of Unicode classes as defined in [RE2 Syntax](https://github.com/google/re2/wiki/Syntax). Native Google RE2 supports only short names, e.g., `L` for `Letter`, `N` for `Number`, etc. Like `RegExp`, `RE2` supports both short and long names, e.g., `Letter` for `L`, by translating them to short names.
+
+The extended form `\p{name=value}` is not supported. Only form `\p{name}` is supported.
+
+The same applies to `\P{name}`.
+
 ## Release history
 
-- 1.21.5 *Updated all dependencies and the list of pre-compiled targets. Fixed minor bugs. C++ style fix (thx, [Benjamin Brienen](https://github.com/BenjaminBrienen)). Added Windows 11 ARM build runner [Kagami Sascha Rosylight](https://github.com/saschanaz).*
+- 1.22.0 *Added support for translation of Unicode classes (thx, [John Livingston](https://github.com/JohnXLivingston)). Added [attestations](https://github.com/uhop/node-re2/attestations).*
+- 1.21.5 *Updated all dependencies and the list of pre-compiled targets. Fixed minor bugs. C++ style fix (thx, [Benjamin Brienen](https://github.com/BenjaminBrienen)). Added Windows 11 ARM build runner (thx, [Kagami Sascha Rosylight](https://github.com/saschanaz)).*
 - 1.21.4 *Fixed a regression reported by [caroline-matsec](https://github.com/caroline-matsec), thx! Added pre-compilation targets for Alpine Linux on ARM. Updated deps.*
 - 1.21.3 *Fixed an empty string regression reported by [Rhys Arkins](https://github.com/rarkins), thx! Updated deps.*
 - 1.21.2 *Fixed another memory regression reported by [matthewvalentine](https://github.com/matthewvalentine), thx! Updated deps. Added more tests and benchmarks.*

package/lib/new.cc

@@ -1,6 +1,7 @@
 #include "./wrapped_re2.h"
 #include "./util.h"
 
+#include <map>
 #include <memory>
 #include <string>
 #include <unordered_set>
@@ -18,6 +19,47 @@ inline bool isHexadecimal(char ch)
 	return ('0' <= ch && ch <= '9') || ('A' <= ch && ch <= 'Z') || ('a' <= ch && ch <= 'z');
 }
 
+static std::map<std::string, std::string> unicodeClasses = {
+	{"Uppercase_Letter", "Lu"},
+	{"Lowercase_Letter", "Ll"},
+	{"Titlecase_Letter", "Lt"},
+	{"Cased_Letter", "LC"},
+	{"Modifier_Letter", "Lm"},
+	{"Other_Letter", "Lo"},
+	{"Letter", "L"},
+	{"Nonspacing_Mark", "Mn"},
+	{"Spacing_Mark", "Mc"},
+	{"Enclosing_Mark", "Me"},
+	{"Mark", "M"},
+	{"Decimal_Number", "Nd"},
+	{"Letter_Number", "Nl"},
+	{"Other_Number", "No"},
+	{"Number", "N"},
+	{"Connector_Punctuation", "Pc"},
+	{"Dash_Punctuation", "Pd"},
+	{"Open_Punctuation", "Ps"},
+	{"Close_Punctuation", "Pe"},
+	{"Initial_Punctuation", "Pi"},
+	{"Final_Punctuation", "Pf"},
+	{"Other_Punctuation", "Po"},
+	{"Punctuation", "P"},
+	{"Math_Symbol", "Sm"},
+	{"Currency_Symbol", "Sc"},
+	{"Modifier_Symbol", "Sk"},
+	{"Other_Symbol", "So"},
+	{"Symbol", "S"},
+	{"Space_Separator", "Zs"},
+	{"Line_Separator", "Zl"},
+	{"Paragraph_Separator", "Zp"},
+	{"Separator", "Z"},
+	{"Control", "Cc"},
+	{"Format", "Cf"},
+	{"Surrogate", "Cs"},
+	{"Private_Use", "Co"},
+	{"Unassigned", "Cn"},
+	{"Other", "C"},
+};
+
 static bool translateRegExp(const char *data, size_t size, bool multiline, std::vector<char> &buffer)
 {
 	std::string result;
@@ -98,6 +140,36 @@ static bool translateRegExp(const char *data, size_t size, bool multiline, std::
 					result += "\\u";
 					i += 2;
 					continue;
+				case 'p':
+				case 'P':
+					if (i + 2 < size) {
+						if (data[i + 2] == '{') {
+							size_t j = i + 3;
+							while (j < size && data[j] != '}') ++j;
+							if (j < size) {
+								result += "\\";
+								result += data[i + 1];
+								std::string name(data + i + 3, j - i - 3);
+								if (unicodeClasses.find(name) != unicodeClasses.end()) {
+									name = unicodeClasses[name];
+								}
+								if (name.size() == 1) {
+									result += name;
+								} else {
+									result += "{";
+									result += name;
+									result += "}";
+								}
+								i = j + 1;
+								changed = true;
+								continue;
+							}
+						}
+					}
+					result += "\\";
+					result += data[i + 1];
+					i += 2;
+					continue;
 				default:
 					result += "\\";
 					size_t sym_size = getUtf8CharSize(ch);

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "re2",
-  "version": "1.21.5",
+  "version": "1.22.0",
   "description": "Bindings for RE2: fast, safe alternative to backtracking regular expression engines.",
   "homepage": "https://github.com/uhop/node-re2",
   "bugs": "https://github.com/uhop/node-re2/issues",
@@ -15,7 +15,7 @@
     "vendor"
   ],
   "dependencies": {
-    "install-artifact-from-github": "^1.3.5",
+    "install-artifact-from-github": "^1.4.0",
     "nan": "^2.22.2",
     "node-gyp": "^11.2.0"
   },