re2 1.17.1 → 1.17.4

package/README.md

@@ -13,6 +13,10 @@ at his [Implementing Regular Expressions](http://swtch.com/~rsc/regexp/) page.
 (see [Syntax](https://github.com/google/re2/wiki/Syntax)),
 but it lacks two features: backreferences and lookahead assertions. See below for more details.
 
+`RE2` always works in the [Unicode mode](https://developer.mozilla.org/en-US/docs/Web/JavaScript/Reference/Global_Objects/RegExp/unicode), which means that all matches that use character codes are interpret as Unicode code points, not as binary values of UTF-16.
+See `RE2.unicodeWarningLevel` below for more details.
+
+
 `RE2` object emulates standard `RegExp` making it a practical drop-in replacement in most cases.
 `RE2` is extended to provide `String`-based regular expression methods as well. To help to convert
 `RegExp` objects to `RE2` its constructor can take `RegExp` directly honoring all properties.
@@ -343,6 +347,9 @@ console.log('re2_res : ' + re2_res); // prints: re2_res : abc,a,b,c
 
 ## Release history
 
+- 1.17.4 *Updated deps.*
+- 1.17.3 *Fixed bug with zero-length replacements.*
+- 1.17.2 *Added support for the enhanced local mirroring by updating [install-artifact-from-github](https://github.com/uhop/install-artifact-from-github).*
 - 1.17.1 *Fix for `lastIndex` for U+10000 - U+10FFFF UTF characters. Thx, [omg](https://github.com/omg).*
 - 1.17.0 *Updated GYP, added support for Node 17, updated deps.*
 - 1.16.0 *Updated the compiler (thx, [Sergei Dyshel](https://github.com/sergei-dyshel)), updated GYP, removed support for Node 10, added support for Node 16, updated TS bindings (thx, [BannerBomb](https://github.com/BannerBomb)).*

package/lib/replace.cc

@@ -228,7 +228,9 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 				{
 					size_t s = getUtf8CharSize(data[lastIndex]);
 					lastIndex += s;
-					if (s == 4 && n >= 2) --n; // this utf8 character will take two utf16 characters
+					if (s == 4 && n >= 2) {
+						--n; // this utf8 character will take two utf16 characters
+					}
 					// the decrement above is protected to avoid an overflow of an unsigned integer
 				}
 			}
@@ -245,28 +247,30 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 	while (lastIndex <= size && re2->regexp.Match(str, lastIndex, size, anchor, &groups[0], groups.size()))
 	{
 		noMatch = false;
+		auto offset = match.data() - data;
 		if (!re2->global && re2->sticky)
 		{
-			re2->lastIndex += replacee.isBuffer ? match.data() - data + match.size() - lastIndex : getUtf16Length(data + lastIndex, match.data() + match.size());
+			re2->lastIndex += replacee.isBuffer ? offset + match.size() - lastIndex : getUtf16Length(data + lastIndex, match.data() + match.size());
+		}
+		if (match.data() == data || offset > static_cast<long>(lastIndex))
+		{
+			result += std::string(data + lastIndex, offset - lastIndex);
 		}
+		result += replace(replacer, replacer_size, groups, str, namedGroups);
 		if (match.size())
 		{
-			if (match.data() == data || match.data() - data > static_cast<long>(lastIndex))
-			{
-				result += std::string(data + lastIndex, match.data() - data - lastIndex);
-			}
-			result += replace(replacer, replacer_size, groups, str, namedGroups);
-			lastIndex = match.data() - data + match.size();
+			lastIndex = offset + match.size();
+		}
+		else if (offset < size)
+		{
+			auto sym_size = getUtf8CharSize(data[offset]);
+			result.append(data + offset, sym_size);
+			lastIndex = offset + sym_size;
 		}
 		else
 		{
-			result += replace(replacer, replacer_size, groups, str, namedGroups);
-			size_t sym_size = getUtf8CharSize(data[lastIndex]);
-			if (lastIndex < size)
-			{
-				result.append(data + lastIndex, sym_size);
-			}
-			lastIndex += sym_size;
+			lastIndex = size;
+			break;
 		}
 		if (!re2->global)
 		{
@@ -295,7 +299,7 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 
 inline Nan::Maybe<std::string> replace(const Nan::Callback *replacer, const std::vector<re2::StringPiece> &groups, const re2::StringPiece &str, const v8::Local<v8::Value> &input, bool useBuffers, const std::map<std::string, int> &namedGroups)
 {
-	std::vector<v8::Local<v8::Value>> argv;
+	std::vector<v8::Local<v8::Value> > argv;
 
 	auto context = Nan::GetCurrentContext();
 
@@ -377,7 +381,9 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 				{
 					size_t s = getUtf8CharSize(data[lastIndex]);
 					lastIndex += s;
-					if (s == 4 && n >= 2) --n; // this utf8 character will take two utf16 characters
+					if (s == 4 && n >= 2) {
+						--n; // this utf8 character will take two utf16 characters
+					}
 					// the decrement above is protected to avoid an overflow of an unsigned integer
 				}
 			}
@@ -396,38 +402,35 @@ static Nan::Maybe<std::string> replace(WrappedRE2 *re2, const StrVal &replacee,
 	while (lastIndex <= size && re2->regexp.Match(str, lastIndex, size, anchor, &groups[0], groups.size()))
 	{
 		noMatch = false;
+		auto offset = match.data() - data;
 		if (!re2->global && re2->sticky)
 		{
-			re2->lastIndex += replacee.isBuffer ? match.data() - data + match.size() - lastIndex : getUtf16Length(data + lastIndex, match.data() + match.size());
+			re2->lastIndex += replacee.isBuffer ? offset + match.size() - lastIndex : getUtf16Length(data + lastIndex, match.data() + match.size());
+		}
+		if (match.data() == data || offset > static_cast<long>(lastIndex))
+		{
+			result += std::string(data + lastIndex, offset - lastIndex);
 		}
+		const auto part = replace(replacer, groups, str, input, useBuffers, namedGroups);
+		if (part.IsNothing())
+		{
+			return part;
+		}
+		result += part.FromJust();
 		if (match.size())
 		{
-			if (match.data() == data || match.data() - data > static_cast<long>(lastIndex))
-			{
-				result += std::string(data + lastIndex, match.data() - data - lastIndex);
-			}
-			const auto part = replace(replacer, groups, str, input, useBuffers, namedGroups);
-			if (part.IsNothing())
-			{
-				return part;
-			}
-			result += part.FromJust();
-			lastIndex = match.data() - data + match.size();
+			lastIndex = offset + match.size();
+		}
+		else if (offset < size)
+		{
+			auto sym_size = getUtf8CharSize(data[offset]);
+			result.append(data + offset, sym_size);
+			lastIndex = offset + sym_size;
 		}
 		else
 		{
-			const auto part = replace(replacer, groups, str, input, useBuffers, namedGroups);
-			if (part.IsNothing())
-			{
-				return part;
-			}
-			result += part.FromJust();
-			size_t sym_size = getUtf8CharSize(data[lastIndex]);
-			if (lastIndex < size)
-			{
-				result.append(data + lastIndex, sym_size);
-			}
-			lastIndex += sym_size;
+			lastIndex = size;
+			break;
 		}
 		if (!re2->global)
 		{

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "re2",
-  "version": "1.17.1",
+  "version": "1.17.4",
   "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",
@@ -10,7 +10,7 @@
     "test": "tests"
   },
   "dependencies": {
-    "install-artifact-from-github": "^1.2.0",
+    "install-artifact-from-github": "^1.3.0",
     "nan": "^2.15.0",
     "node-gyp": "^8.4.1"
   },
@@ -20,7 +20,7 @@
   "scripts": {
     "test": "node tests/tests.js",
     "save-to-github": "save-to-github-cache --artifact build/Release/re2.node",
-    "install": "install-from-cache --artifact build/Release/re2.node --host-var RE2_DOWNLOAD_MIRROR || npm run rebuild",
+    "install": "install-from-cache --artifact build/Release/re2.node --host-var RE2_DOWNLOAD_MIRROR --skip-path-var RE2_DOWNLOAD_SKIP_PATH --skip-ver-var RE2_DOWNLOAD_SKIP_VER || npm run rebuild",
     "verify-build": "node scripts/verify-build.js",
     "rebuild": "node-gyp rebuild"
   },
@@ -35,6 +35,6 @@
     "text processing",
     "PCRE alternative"
   ],
-  "author": "Eugene Lazutkin <eugene.lazutkin@gmail.com> (http://lazutkin.com/)",
+  "author": "Eugene Lazutkin <eugene.lazutkin@gmail.com> (https://lazutkin.com/)",
   "license": "BSD-3-Clause"
 }