@recaptime-dev's working patches + fork for Phorge, a community fork of Phabricator. (Upstream dev and stable branches are at upstream/main and upstream/stable respectively.)
hq.recaptime.dev/wiki/Phorge
phorge
phabricator
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
+4
-53
+4
-53
src/__tests__/PhabricatorInfrastructureTestCase.php
+4
-53
src/__tests__/PhabricatorInfrastructureTestCase.php
···
50
50
'In test cases, all applications should default to installed.');
51
51
}
52
52
53
-
public function testMySQLAgreesWithUsAboutBMP() {
54
-
// Build a string with every BMP character in it, then insert it into MySQL
55
-
// and read it back. We expect to get the same string out that we put in,
56
-
// demonstrating that strings which pass our BMP checks are also valid in
57
-
// MySQL and no silent data truncation will occur.
58
-
59
-
$buf = '';
60
-
61
-
for ($ii = 0x01; $ii <= 0x7F; $ii++) {
62
-
$buf .= chr($ii);
63
-
}
64
-
65
-
for ($ii = 0xC2; $ii <= 0xDF; $ii++) {
66
-
for ($jj = 0x80; $jj <= 0xBF; $jj++) {
67
-
$buf .= chr($ii).chr($jj);
68
-
}
69
-
}
70
-
71
-
// NOTE: This is \xE0\xA0\xZZ.
72
-
for ($ii = 0xE0; $ii <= 0xE0; $ii++) {
73
-
for ($jj = 0xA0; $jj <= 0xBF; $jj++) {
74
-
for ($kk = 0x80; $kk <= 0xBF; $kk++) {
75
-
$buf .= chr($ii).chr($jj).chr($kk);
76
-
}
77
-
}
78
-
}
79
-
80
-
// NOTE: This is \xE1\xZZ\xZZ through \xEF\xZZ\xZZ.
81
-
for ($ii = 0xE1; $ii <= 0xEF; $ii++) {
82
-
for ($jj = 0x80; $jj <= 0xBF; $jj++) {
83
-
for ($kk = 0x80; $kk <= 0xBF; $kk++) {
84
-
$buf .= chr($ii).chr($jj).chr($kk);
85
-
}
86
-
}
87
-
}
88
-
89
-
$this->assertEqual(194431, strlen($buf));
90
-
$this->assertTrue(phutil_is_utf8_with_only_bmp_characters($buf));
91
-
92
-
$write = id(new HarbormasterScratchTable())
93
-
->setData('all.utf8.bmp')
94
-
->setBigData($buf)
95
-
->save();
96
-
97
-
$read = id(new HarbormasterScratchTable())->load($write->getID());
98
-
99
-
$this->assertEqual($buf, $read->getBigData());
100
-
}
101
-
102
-
public function testRejectMySQLBMPQueries() {
53
+
public function testRejectMySQLNonUTF8Queries() {
103
54
$table = new HarbormasterScratchTable();
104
55
$conn_r = $table->establishConnection('w');
105
56
106
57
$snowman = "\xE2\x98\x83";
107
-
$gclef = "\xF0\x9D\x84\x9E";
58
+
$invalid = "\xE6\x9D";
108
59
109
60
qsprintf($conn_r, 'SELECT %B', $snowman);
110
61
qsprintf($conn_r, 'SELECT %s', $snowman);
111
-
qsprintf($conn_r, 'SELECT %B', $gclef);
62
+
qsprintf($conn_r, 'SELECT %B', $invalid);
112
63
113
64
$caught = null;
114
65
try {
115
-
qsprintf($conn_r, 'SELECT %s', $gclef);
66
+
qsprintf($conn_r, 'SELECT %s', $invalid);
116
67
} catch (AphrontCharacterSetQueryException $ex) {
117
68
$caught = $ex;
118
69
}