Add test coverage that our definition of BMP agrees with MySQL

Summary:
Ref T1191. Test that MySQL's rules match those of `phutil_is_utf8_with_only_bmp_characters()`:

  - Build a string with //every// character that we consider to be a BMP character.
  - Write it into MySQL.
  - Read it back out.
  - Make sure MySQL didn't truncate it.

Test Plan: Ran unit test. This test runs pretty quickly (50ms), the string with every character isn't all that enormous.

Reviewers: btrahan, arice

Reviewed By: arice

CC: chad, arice, aran

Maniphest Tasks: T1191

Differential Revision: https://secure.phabricator.com/D8314

resources/sql/autopatches/20140223.bigutf8scratch.sql

@@ -0,0 +1,2 @@
+ALTER TABLE {$NAMESPACE}_harbormaster.harbormaster_scratchtable
+  ADD bigData LONGTEXT COLLATE utf8_bin;

src/applications/harbormaster/storage/HarbormasterScratchTable.php

@@ -9,5 +9,6 @@
 final class HarbormasterScratchTable extends HarbormasterDAO {
 
   protected $data;
+  protected $bigData;
 
 }

src/infrastructure/__tests__/PhabricatorInfrastructureTestCase.php

@@ -3,6 +3,12 @@
 final class PhabricatorInfrastructureTestCase
   extends PhabricatorTestCase {
 
+  protected function getPhabricatorTestCaseConfiguration() {
+    return array(
+      self::PHABRICATOR_TESTCONFIG_BUILD_STORAGE_FIXTURES => true,
+    );
+  }
+
   /**
    * This is more of an acceptance test case instead of a unittest. It verifies
    * that all symbols can be loaded correctly. It can catch problems like
@@ -22,6 +28,54 @@ final class PhabricatorInfrastructureTestCase
       'In test cases, all applications should default to installed.');
   }
 
+  public function testMySQLAgreesWithUsAboutBMP() {
+    // Build a string with every BMP character in it, then insert it into MySQL
+    // and read it back. We expect to get the same string out that we put in,
+    // demonstrating that strings which pass our BMP checks are also valid in
+    // MySQL and no silent data truncation will occur.
+
+    $buf = '';
+
+    for ($ii = 0x01; $ii <= 0x7F; $ii++) {
+      $buf .= chr($ii);
+    }
+
+    for ($ii = 0xC2; $ii <= 0xDF; $ii++) {
+      for ($jj = 0x80; $jj <= 0xBF; $jj++) {
+        $buf .= chr($ii).chr($jj);
+      }
+    }
+
+    // NOTE: This is \xE0\xA0\xZZ.
+    for ($ii = 0xE0; $ii <= 0xE0; $ii++) {
+      for ($jj = 0xA0; $jj <= 0xBF; $jj++) {
+        for ($kk = 0x80; $kk <= 0xBF; $kk++) {
+          $buf .= chr($ii).chr($jj).chr($kk);
+        }
+      }
+    }
+
+    // NOTE: This is \xE1\xZZ\xZZ through \xEF\xZZ\xZZ.
+    for ($ii = 0xE1; $ii <= 0xEF; $ii++) {
+      for ($jj = 0x80; $jj <= 0xBF; $jj++) {
+        for ($kk = 0x80; $kk <= 0xBF; $kk++) {
+          $buf .= chr($ii).chr($jj).chr($kk);
+        }
+      }
+    }
+
+    $this->assertEqual(194431, strlen($buf));
+    $this->assertEqual(true, phutil_is_utf8_with_only_bmp_characters($buf));
+
+    $write = id(new HarbormasterScratchTable())
+      ->setData('all.utf8.bmp')
+      ->setBigData($buf)
+      ->save();
+
+    $read = id(new HarbormasterScratchTable())->load($write->getID());
+
+    $this->assertEqual($buf, $read->getBigData());
+  }
 
 }