src: do not ignore IDNA conversion error

Old behavior can be restored using a special `lenient` mode, as used in
the legacy URL parser.

PR-URL: https://github.com/nodejs/node/pull/11549
Reviewed-By: Anna Henningsen <anna@addaleax.net>
Reviewed-By: Ben Noordhuis <info@bnoordhuis.nl>
Reviewed-By: James M Snell <jasnell@gmail.com>
Reviewed-By: Joyee Cheung <joyeec9h3@gmail.com>

3 files changed, 22 insertions, 9 deletions

diff --git a/lib/url.js b/lib/url.js
index 5d59d3c10f..2db780c7eb 100644
--- a/lib/url.js
+++ b/lib/url.js
@@ -319,7 +319,10 @@ Url.prototype.parse = function(url, parseQueryString, slashesDenoteHost) {
       // It only converts parts of the domain name that
       // have non-ASCII characters, i.e. it doesn't matter if
       // you call it with a domain that already is ASCII-only.
-      this.hostname = toASCII(this.hostname);
+
+      // Use lenient mode (`true`) to try to support even non-compliant
+      // URLs.
+      this.hostname = toASCII(this.hostname, true);
     }
 
     var p = this.port ? ':' + this.port : '';
diff --git a/src/node_i18n.cc b/src/node_i18n.cc
index ae14aed7c6..b337456c63 100644
--- a/src/node_i18n.cc
+++ b/src/node_i18n.cc
@@ -410,7 +410,8 @@ bool InitializeICUDirectory(const std::string& path) {
 
 int32_t ToUnicode(MaybeStackBuffer<char>* buf,
                   const char* input,
-                  size_t length) {
+                  size_t length,
+                  bool lenient) {
   UErrorCode status = U_ZERO_ERROR;
   uint32_t options = UIDNA_DEFAULT;
   options |= UIDNA_NONTRANSITIONAL_TO_UNICODE;
@@ -435,7 +436,7 @@ int32_t ToUnicode(MaybeStackBuffer<char>* buf,
                                   &status);
   }
 
-  if (U_FAILURE(status)) {
+  if (U_FAILURE(status) || (!lenient && info.errors != 0)) {
     len = -1;
     buf->SetLength(0);
   } else {
@@ -448,7 +449,8 @@ int32_t ToUnicode(MaybeStackBuffer<char>* buf,
 
 int32_t ToASCII(MaybeStackBuffer<char>* buf,
                 const char* input,
-                size_t length) {
+                size_t length,
+                bool lenient) {
   UErrorCode status = U_ZERO_ERROR;
   uint32_t options = UIDNA_DEFAULT;
   options |= UIDNA_NONTRANSITIONAL_TO_ASCII;
@@ -473,7 +475,7 @@ int32_t ToASCII(MaybeStackBuffer<char>* buf,
                                  &status);
   }
 
-  if (U_FAILURE(status)) {
+  if (U_FAILURE(status) || (!lenient && info.errors != 0)) {
     len = -1;
     buf->SetLength(0);
   } else {
@@ -489,8 +491,11 @@ static void ToUnicode(const FunctionCallbackInfo<Value>& args) {
   CHECK_GE(args.Length(), 1);
   CHECK(args[0]->IsString());
   Utf8Value val(env->isolate(), args[0]);
+  // optional arg
+  bool lenient = args[1]->BooleanValue(env->context()).FromJust();
+
   MaybeStackBuffer<char> buf;
-  int32_t len = ToUnicode(&buf, *val, val.length());
+  int32_t len = ToUnicode(&buf, *val, val.length(), lenient);
 
   if (len < 0) {
     return env->ThrowError("Cannot convert name to Unicode");
@@ -508,8 +513,11 @@ static void ToASCII(const FunctionCallbackInfo<Value>& args) {
   CHECK_GE(args.Length(), 1);
   CHECK(args[0]->IsString());
   Utf8Value val(env->isolate(), args[0]);
+  // optional arg
+  bool lenient = args[1]->BooleanValue(env->context()).FromJust();
+
   MaybeStackBuffer<char> buf;
-  int32_t len = ToASCII(&buf, *val, val.length());
+  int32_t len = ToASCII(&buf, *val, val.length(), lenient);
 
   if (len < 0) {
     return env->ThrowError("Cannot convert name to ASCII");
diff --git a/src/node_i18n.h b/src/node_i18n.h
index 21567eeb3e..270eb1f3d1 100644
--- a/src/node_i18n.h
+++ b/src/node_i18n.h
@@ -18,10 +18,12 @@ bool InitializeICUDirectory(const std::string& path);
 
 int32_t ToASCII(MaybeStackBuffer<char>* buf,
                 const char* input,
-                size_t length);
+                size_t length,
+                bool lenient = false);
 int32_t ToUnicode(MaybeStackBuffer<char>* buf,
                   const char* input,
-                  size_t length);
+                  size_t length,
+                  bool lenient = false);
 
 }  // namespace i18n
 }  // namespace node