From 801d4e82651b0d4f68ddbf5e76d3acacbdc77f82 Mon Sep 17 00:00:00 2001
From: "smontagu@smontagu.org" <smontagu@smontagu.org>
Date: Wed, 5 Sep 2007 22:38:21 -0700
Subject: [PATCH] Tests for bug 381412

---
 .../tests/unit/test_bug381412.Big5-HKSCS.js   | 60 +++++++++++++
 intl/uconv/tests/unit/test_bug381412.Big5.js  | 60 +++++++++++++
 intl/uconv/tests/unit/test_bug381412.cp936.js | 56 ++++++++++++
 .../uconv/tests/unit/test_bug381412.euc-kr.js | 60 +++++++++++++
 .../uconv/tests/unit/test_bug381412.euc_jp.js | 89 +++++++++++++++++++
 .../uconv/tests/unit/test_bug381412.gb2312.js | 56 ++++++++++++
 .../tests/unit/test_bug381412.hk.gb2312.js    | 73 +++++++++++++++
 intl/uconv/tests/unit/test_bug381412.johab.js | 60 +++++++++++++
 intl/uconv/tests/unit/test_bug381412.js       | 56 ++++++++++++
 9 files changed, 570 insertions(+)
 create mode 100644 intl/uconv/tests/unit/test_bug381412.Big5-HKSCS.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.Big5.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.cp936.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.euc-kr.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.euc_jp.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.gb2312.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.hk.gb2312.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.johab.js
 create mode 100644 intl/uconv/tests/unit/test_bug381412.js

diff --git a/intl/uconv/tests/unit/test_bug381412.Big5-HKSCS.js b/intl/uconv/tests/unit/test_bug381412.Big5-HKSCS.js
new file mode 100644
index 00000000000..14220d847e5
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.Big5-HKSCS.js
@@ -0,0 +1,60 @@
+const charset = "Big5-HKSCS";
+
+function dumpStrings(inString, outString) {
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+}
+
+function error(inString, outString, msg){
+    dumpStrings(inString, outString);
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.Big5.js b/intl/uconv/tests/unit/test_bug381412.Big5.js
new file mode 100644
index 00000000000..d42976d5b3d
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.Big5.js
@@ -0,0 +1,60 @@
+const charset = "Big5";
+
+function dumpStrings(inString, outString) {
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+}
+
+function error(inString, outString, msg){
+    dumpStrings(inString, outString);
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.cp936.js b/intl/uconv/tests/unit/test_bug381412.cp936.js
new file mode 100644
index 00000000000..903fed8ccfe
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.cp936.js
@@ -0,0 +1,56 @@
+const charset = "windows-936";
+
+function error(inString, outString, msg){
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.euc-kr.js b/intl/uconv/tests/unit/test_bug381412.euc-kr.js
new file mode 100644
index 00000000000..fa69f8772a1
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.euc-kr.js
@@ -0,0 +1,60 @@
+const charset = "EUC-KR";
+
+function dumpStrings(inString, outString) {
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+}
+
+function error(inString, outString, msg){
+    dumpStrings(inString, outString);
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.euc_jp.js b/intl/uconv/tests/unit/test_bug381412.euc_jp.js
new file mode 100644
index 00000000000..344a09628af
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.euc_jp.js
@@ -0,0 +1,89 @@
+const charset = "EUC-JP";
+const ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+var gConverter;
+
+function error(inString, outString, msg){
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+    do_throw("security risk: " + msg);
+}
+
+function test(inString) {
+    var outString = gConverter.ConvertToUnicode(inString) +
+	            gConverter.Finish();
+
+    switch (outString.length) {
+        case 0:
+        case 1:
+        case 2:
+	    error(inString, outString, "Unexpected error");
+	    break;
+        case 3:
+	    error(inString, outString, "3 byte sequence eaten");
+	    break;
+        case 4:
+	    if (outString.charCodeAt(0) < 0x80 &&
+		outString.charCodeAt(1) < 0x80 &&
+		outString.charCodeAt(2) < 0x80 &&
+		outString.charCodeAt(3) < 0x80) {
+		error(inString, outString,
+		      "3 byte sequence converted to 1 ASCII");
+	    }
+	    break;
+        case 5:
+	    if (outString != inString &&
+		outString.charCodeAt(0) < 0x80 &&
+		outString.charCodeAt(1) < 0x80 &&
+		outString.charCodeAt(2) < 0x80 &&
+		outString.charCodeAt(3) < 0x80 &&
+		outString.charCodeAt(4) < 0x80) {
+		error(inString, outString,
+		      "3 byte sequence converted to 2 ASCII");
+	    }
+	    break;
+        case 6:
+	    if (outString != inString &&
+		outString.charCodeAt(0) < 0x80 &&
+		outString.charCodeAt(1) < 0x80 &&
+		outString.charCodeAt(2) < 0x80 &&
+		outString.charCodeAt(3) < 0x80 &&
+		outString.charCodeAt(4) < 0x80 &&
+		outString.charCodeAt(5) < 0x80) {
+		error(inString, outString,
+		      "3 byte sequence converted to 3 ASCII");
+	    }
+	    break;
+    }
+}
+
+function run_test() {
+    gConverter = new ScriptableUnicodeConverter();
+    gConverter.charset = charset;
+
+    var byte1, byte2, byte3;
+    for (byte1 = 1; byte1 < 0x100; ++byte1) {
+	for (byte2 = 1; byte2 < 0x100; ++byte2) {
+	    if (byte1 == 0x8f) {
+		for (byte3 = 1; byte3 < 0x100; ++byte3) {
+		    test(String.fromCharCode(byte1, byte2, byte3) + "foo");
+		}
+	    } else {
+		test(String.fromCharCode(byte1, byte2) + " foo");
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.gb2312.js b/intl/uconv/tests/unit/test_bug381412.gb2312.js
new file mode 100644
index 00000000000..d342bf0b9f9
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.gb2312.js
@@ -0,0 +1,56 @@
+const charset = "GB2312";
+
+function error(inString, outString, msg){
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.hk.gb2312.js b/intl/uconv/tests/unit/test_bug381412.hk.gb2312.js
new file mode 100644
index 00000000000..27e58fe470f
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.hk.gb2312.js
@@ -0,0 +1,73 @@
+const charset = "hz-gb-2312";
+
+function dumpStrings(inString, outString) {
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+}
+
+function error(inString, outString, msg){
+    dumpStrings(inString, outString);
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    if (leadByte == 0x7e) {
+		if (trailByte == 0x7b ||
+		    trailByte == 0xa  ||
+		    trailByte == 0x7e) {
+		    // ignore escape sequences:
+		    //  ~{ (sets HZ-GB mode)
+		    //  ~\n ( ==> \n)
+		    //  ~~ ( ==> ~)
+		    continue;
+		}
+	    }
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) +
+		            converter.Finish();
+	    switch (outString.length) {
+                case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+                case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+                case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.johab.js b/intl/uconv/tests/unit/test_bug381412.johab.js
new file mode 100644
index 00000000000..5cf82ed26e7
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.johab.js
@@ -0,0 +1,60 @@
+const charset = "x-johab";
+
+function dumpStrings(inString, outString) {
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+}
+
+function error(inString, outString, msg){
+    dumpStrings(inString, outString);
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}
diff --git a/intl/uconv/tests/unit/test_bug381412.js b/intl/uconv/tests/unit/test_bug381412.js
new file mode 100644
index 00000000000..622b7bf7f5b
--- /dev/null
+++ b/intl/uconv/tests/unit/test_bug381412.js
@@ -0,0 +1,56 @@
+const charset = "Shift_JIS";
+
+function error(inString, outString, msg){
+    var dispIn = "";
+    var dispOut = "";
+    var i;
+    for (i = 0; i < inString.length; ++i) {
+	dispIn += " x" + inString.charCodeAt(i).toString(16);
+    }
+    if (outString.length == 0) {
+	dispOut = "<empty>";
+    } else {
+	for (i = 0; i < outString.length; ++i) {
+	    dispOut += " x" + outString.charCodeAt(i).toString(16);
+	}
+    }
+    dump("\"" + dispIn + "\" ==> \"" + dispOut + "\"\n");
+    do_throw("security risk: " + msg);
+}
+
+function run_test() {
+    var ScriptableUnicodeConverter =
+	Components.Constructor("@mozilla.org/intl/scriptableunicodeconverter",
+			       "nsIScriptableUnicodeConverter");
+
+    var converter = new ScriptableUnicodeConverter();
+    converter.charset = charset;
+
+    var leadByte, trailByte;
+    var inString;
+    for (leadByte = 1; leadByte < 0x100; ++leadByte) {
+	for (trailByte = 1; trailByte < 0x100; ++trailByte) {
+	    inString = String.fromCharCode(leadByte, trailByte, 65);
+	    var outString = converter.ConvertToUnicode(inString) + converter.Finish();
+	    switch (outString.length) {
+	        case 1:
+		    error(inString, outString, "2 byte sequence eaten");
+		    break;
+	        case 2:
+		    if (outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString, "2 byte sequence converted to 1 ASCII");
+		    }
+		    break;
+	        case 3:
+		    if (outString != inString &&
+			outString.charCodeAt(0) < 0x80 &&
+			outString.charCodeAt(1) < 0x80) {
+			error(inString, outString,
+			      "2 byte sequence converted to 2 ASCII");
+		    }
+		    break;
+	    }
+	}
+    }
+}