Correctly extracting email addresses

core/src/fpdftext/fpdf_text_int_unittest.cpp

+// Copyright 2015 PDFium Authors. All rights reserved.
+// Use of this source code is governed by a BSD-style license that can be
+// found in the LICENSE file.
+
+#include "testing/gtest/include/gtest/gtest.h"
+
+#include "core/src/fpdftext/text_int.h"
+
+// Class to help test functions in CPDF_LinkExtract class.
+class CPDF_TestLinkExtract : public CPDF_LinkExtract {
+ private:
+  // Add test cases as friends to access protected member functions.
+  // Access CheckMailLink.
+  FRIEND_TEST(fpdf_text_int, CheckMailLink);
+};
+
+TEST(fpdf_text_int, CheckMailLink) {
+  CPDF_TestLinkExtract extractor;
+  // Check cases that fail to extract valid mail link.
+  const wchar_t* invalid_strs[] = {
+      L"",
+      L"peter.pan"        // '@' is required.
+      L"abc@server",      // Host name needs at least one '.'.

[Lei Zhang, 2015/12/18 00:24:09]

As I mentioned previously, we need to investigate whether this is actually valid
or not. Please add a TODO.

[Wei Li, 2015/12/18 01:12:21]

Added in cpp file.

+      L"abc.@gmail.com",  // '.' can not immediately precede '@'.
+      L"abc@xyz&q.org",   // Host name should not contain '&'.
+      L"abc@.xyz.org",    // Host name should not start with '.'.
+      L"fan@g..com"       // Host name should not have consecutive '.'
+  };
+  for (int i = 0; i < FX_ArraySize(invalid_strs); ++i) {
+    CFX_WideString text_str(invalid_strs[i]);
+    EXPECT_EQ(FALSE, extractor.CheckMailLink(text_str));

[Lei Zhang, 2015/12/18 00:24:09]

It should be trivial to convert CheckMailLink() to return a bool. Can we do that
and use EXPECT_FALSE() here, and EXPECT_TRUE() below?

[Wei Li, 2015/12/18 01:12:21]

Done.

+  }
+
+  // Check cases that can extract valid mail link.
+  // An array of {input_string, expected_extracted_email_address}.
+  const wchar_t* valid_strs[][2] = {
+      {L"peter@abc.d", L"peter@abc.d"},
+      {L"red.teddy.b@abc.com", L"red.teddy.b@abc.com"},
+      {L"abc_@gmail.com", L"abc_@gmail.com"},  // '_' is ok before '@'.
+      {L"dummy-hi@gmail.com",
+       L"dummy-hi@gmail.com"},                  // '-' is ok in user name.
+      {L"a..df@gmail.com", L"df@gmail.com"},    // Stop at consecutive '.'.
+      {L".john@yahoo.com", L"john@yahoo.com"},  // Remove heading '.'.
+      {L"abc@xyz.org?/", L"abc@xyz.org"},       // Trim ending invalid chars.
+      {L"fan{abc@xyz.org", L"abc@xyz.org"},     // Trim beginning invalid chars.
+      {L"fan@g.com..", L"fan@g.com"},           // Trim the ending periods.
+      {L"CAP.cap@Gmail.Com", L"CAP.cap@Gmail.Com"},  // Keep the original case.
+  };
+  for (int i = 0; i < FX_ArraySize(valid_strs); ++i) {
+    CFX_WideString text_str(valid_strs[i][0]);
+    CFX_WideString expected_str(L"mailto:");
+    expected_str += valid_strs[i][1];
+    EXPECT_EQ(TRUE, extractor.CheckMailLink(text_str));
+    EXPECT_STREQ(text_str.c_str(), expected_str.c_str());
+  }
+}