From 410c2ecbfdf7ca7795c70a2d576c1605973b117b Mon Sep 17 00:00:00 2001
From: epriestley <git@epriestley.com>
Date: Thu, 2 Oct 2014 11:47:25 -0700
Subject: [PATCH] Remove UTF8 BMP unit test and replace it with a general UTF8
 test

Summary:
Ref T1191. After utf8mb4 conversion, these tests no longer pass because MySQL allows emoji and gclefs and such.

We could keep these tests running by keeping a `ut8f_bin` table around somewhere, but we have no other use cases for it and it does not seem worth the added complexity. All these BMP-only codepaths are on the way out.

Update the `%s` / `%B` test to make sure it's rejecting invalid byte sequences, which are still not permitted.

Test Plan: Tests now pass.

Reviewers: btrahan

Reviewed By: btrahan

Subscribers: epriestley

Maniphest Tasks: T1191

Differential Revision: https://secure.phabricator.com/D10621
---
 .../PhabricatorInfrastructureTestCase.php     | 57 ++-----------------
 1 file changed, 4 insertions(+), 53 deletions(-)

diff --git a/src/__tests__/PhabricatorInfrastructureTestCase.php b/src/__tests__/PhabricatorInfrastructureTestCase.php
index d90e004af4..88c7b07418 100644
--- a/src/__tests__/PhabricatorInfrastructureTestCase.php
+++ b/src/__tests__/PhabricatorInfrastructureTestCase.php
@@ -50,69 +50,20 @@ final class PhabricatorInfrastructureTestCase extends PhabricatorTestCase {
       '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));
-    $this->assertTrue(phutil_is_utf8_with_only_bmp_characters($buf));
-
-    $write = id(new HarbormasterScratchTable())
-      ->setData('all.utf8.bmp')
-      ->setBigData($buf)
-      ->save();
-
-    $read = id(new HarbormasterScratchTable())->load($write->getID());
-
-    $this->assertEqual($buf, $read->getBigData());
-  }
-
-  public function testRejectMySQLBMPQueries() {
+  public function testRejectMySQLNonUTF8Queries() {
     $table = new HarbormasterScratchTable();
     $conn_r = $table->establishConnection('w');
 
     $snowman = "\xE2\x98\x83";
-    $gclef = "\xF0\x9D\x84\x9E";
+    $invalid = "\xE6\x9D";
 
     qsprintf($conn_r, 'SELECT %B', $snowman);
     qsprintf($conn_r, 'SELECT %s', $snowman);
-    qsprintf($conn_r, 'SELECT %B', $gclef);
+    qsprintf($conn_r, 'SELECT %B', $invalid);
 
     $caught = null;
     try {
-      qsprintf($conn_r, 'SELECT %s', $gclef);
+      qsprintf($conn_r, 'SELECT %s', $invalid);
     } catch (AphrontCharacterSetQueryException $ex) {
       $caught = $ex;
     }