From 9a52bd030189280f48a35d3b9e52f9d77071cb5e Mon Sep 17 00:00:00 2001 From: Igor Wiedler Date: Mon, 3 Jan 2011 22:21:54 +0100 Subject: [task/phpunit-xml] Use phpunit.xml for test suite PHPBB3-9967 --- tests/utf/all_tests.php | 43 ------------------------------------ tests/utf/utf8_clean_string_test.php | 3 +-- tests/utf/utf8_wordwrap_test.php | 3 +-- 3 files changed, 2 insertions(+), 47 deletions(-) delete mode 100644 tests/utf/all_tests.php (limited to 'tests/utf') diff --git a/tests/utf/all_tests.php b/tests/utf/all_tests.php deleted file mode 100644 index 0d5d44d695..0000000000 --- a/tests/utf/all_tests.php +++ /dev/null @@ -1,43 +0,0 @@ -addTestSuite('phpbb_utf_utf8_wordwrap_test'); - $suite->addTestSuite('phpbb_utf_utf8_clean_string_test'); - - return $suite; - } -} - -if (PHPUnit_MAIN_METHOD == 'phpbb_utf_all_tests::main') -{ - phpbb_utf_all_tests::main(); -} - diff --git a/tests/utf/utf8_clean_string_test.php b/tests/utf/utf8_clean_string_test.php index 870ad76fc4..148297ad4b 100644 --- a/tests/utf/utf8_clean_string_test.php +++ b/tests/utf/utf8_clean_string_test.php @@ -7,8 +7,7 @@ * */ -require_once 'test_framework/framework.php'; -require_once '../phpBB/includes/utf/utf_tools.php'; +require_once __DIR__ . '/../../phpBB/includes/utf/utf_tools.php'; class phpbb_utf_utf8_clean_string_test extends phpbb_test_case { diff --git a/tests/utf/utf8_wordwrap_test.php b/tests/utf/utf8_wordwrap_test.php index ef1165a897..fbc947b92a 100644 --- a/tests/utf/utf8_wordwrap_test.php +++ b/tests/utf/utf8_wordwrap_test.php @@ -7,8 +7,7 @@ * */ -require_once 'test_framework/framework.php'; -require_once '../phpBB/includes/utf/utf_tools.php'; +require_once __DIR__ . '/../../phpBB/includes/utf/utf_tools.php'; class phpbb_utf_utf8_wordwrap_test extends phpbb_test_case { -- cgit v1.2.1 From e00c5544d260741c6639d9a005ea933888ee8317 Mon Sep 17 00:00:00 2001 From: Igor Wiedler Date: Mon, 10 Jan 2011 23:20:06 +0100 Subject: [ticket/9990] Integrate utf normalizer tests into test suite PHPBB3-9990 --- tests/utf/data/.gitkeep | 0 tests/utf/normalizer_test.php | 318 ++++++++++++++++++++++++++++++++++++++++++ 2 files changed, 318 insertions(+) create mode 100644 tests/utf/data/.gitkeep create mode 100644 tests/utf/normalizer_test.php (limited to 'tests/utf') diff --git a/tests/utf/data/.gitkeep b/tests/utf/data/.gitkeep new file mode 100644 index 0000000000..e69de29bb2 diff --git a/tests/utf/normalizer_test.php b/tests/utf/normalizer_test.php new file mode 100644 index 0000000000..9a9011c0fe --- /dev/null +++ b/tests/utf/normalizer_test.php @@ -0,0 +1,318 @@ + array( + 'c2' => array('c1', 'c2', 'c3'), + 'c4' => array('c4', 'c5') + ), + + /** + * NFD + * c3 == NFD(c1) == NFD(c2) == NFD(c3) + * c5 == NFD(c4) == NFD(c5) + */ + 'NFD' => array( + 'c3' => array('c1', 'c2', 'c3'), + 'c5' => array('c4', 'c5') + ), + + /** + * NFKC + * c4 == NFKC(c1) == NFKC(c2) == NFKC(c3) == NFKC(c4) == NFKC(c5) + */ + 'NFKC' => array( + 'c4' => array('c1', 'c2', 'c3', 'c4', 'c5') + ), + + /** + * NFKD + * c5 == NFKD(c1) == NFKD(c2) == NFKD(c3) == NFKD(c4) == NFKD(c5) + */ + 'NFKD' => array( + 'c5' => array('c1', 'c2', 'c3', 'c4', 'c5') + ) + ); + + $tested_chars = array(); + + $fp = fopen(__DIR__.'/data/NormalizationTest.txt', 'rb'); + while (!feof($fp)) + { + $line = fgets($fp); + + if ($line[0] == '@') + { + continue; + } + + if (!strpos(' 0123456789ABCDEF', $line[0])) + { + continue; + } + + list($c1, $c2, $c3, $c4, $c5) = explode(';', $line); + + if (!strpos($c1, ' ')) + { + /** + * We are currently testing a single character, we add it to the list of + * characters we have processed so that we can exclude it when testing + * for invariants + */ + $tested_chars[$c1] = 1; + } + + foreach ($test_suite as $form => $serie) + { + foreach ($serie as $expected => $tests) + { + $hex_expected = ${$expected}; + $utf_expected = $this->hexseq_to_utf($hex_expected); + + foreach ($tests as $test) + { + $utf_result = $utf_expected; + call_user_func(array('utf_normalizer', $form), &$utf_result); + + $hex_result = $this->utf_to_hexseq($utf_result); + $this->assertEquals($utf_expected, $utf_result, "$expected == $form($test) ($hex_expected != $hex_result)"); + } + } + } + } + fclose($fp); + + return $tested_chars; + } + + /** + * @depends test_normalizer + */ + public function test_invariants(array $tested_chars) + { + $fp = fopen(__DIR__.'/data/UnicodeData.txt', 'rb'); + + while (!feof($fp)) + { + $line = fgets($fp, 1024); + + if (!$pos = strpos($line, ';')) + { + continue; + } + + $hex_tested = $hex_expected = substr($line, 0, $pos); + + if (isset($tested_chars[$hex_tested])) + { + continue; + } + + $utf_expected = $this->hex_to_utf($hex_expected); + + if ($utf_expected >= UTF8_SURROGATE_FIRST + && $utf_expected <= UTF8_SURROGATE_LAST) + { + /** + * Surrogates are illegal on their own, we expect the normalizer + * to return a replacement char + */ + $utf_expected = UTF8_REPLACEMENT; + $hex_expected = $this->utf_to_hexseq($utf_expected); + } + + foreach (array('nfc', 'nfkc', 'nfd', 'nfkd') as $form) + { + $utf_result = $utf_expected; + call_user_func(array('utf_normalizer', $form), &$utf_result); + $hex_result = $this->utf_to_hexseq($utf_result); + + $this->assertEquals($utf_expected, $utf_result, "$hex_expected == $form($hex_tested) ($hex_expected != $hex_result)"); + } + } + fclose($fp); + } + + /** + * Convert a UTF string to a sequence of codepoints in hexadecimal + * + * @param string $utf UTF string + * @return integer Unicode codepoints in hex + */ + protected function utf_to_hexseq($str) + { + $pos = 0; + $len = strlen($str); + $ret = array(); + + while ($pos < $len) + { + $c = $str[$pos]; + switch ($c & "\xF0") + { + case "\xC0": + case "\xD0": + $utf_char = substr($str, $pos, 2); + $pos += 2; + break; + + case "\xE0": + $utf_char = substr($str, $pos, 3); + $pos += 3; + break; + + case "\xF0": + $utf_char = substr($str, $pos, 4); + $pos += 4; + break; + + default: + $utf_char = $c; + ++$pos; + } + + $hex = dechex($this->utf_to_cp($utf_char)); + + if (!isset($hex[3])) + { + $hex = substr('000' . $hex, -4); + } + + $ret[] = $hex; + } + + return strtr(implode(' ', $ret), 'abcdef', 'ABCDEF'); + } + + /** + * Convert a UTF-8 char to its codepoint + * + * @param string $utf_char UTF-8 char + * @return integer Unicode codepoint + */ + protected function utf_to_cp($utf_char) + { + switch (strlen($utf_char)) + { + case 1: + return ord($utf_char); + + case 2: + return ((ord($utf_char[0]) & 0x1F) << 6) | (ord($utf_char[1]) & 0x3F); + + case 3: + return ((ord($utf_char[0]) & 0x0F) << 12) | ((ord($utf_char[1]) & 0x3F) << 6) | (ord($utf_char[2]) & 0x3F); + + case 4: + return ((ord($utf_char[0]) & 0x07) << 18) | ((ord($utf_char[1]) & 0x3F) << 12) | ((ord($utf_char[2]) & 0x3F) << 6) | (ord($utf_char[3]) & 0x3F); + + default: + throw new RuntimeException('UTF-8 chars can only be 1-4 bytes long'); + } + } + + /** + * Return a UTF string formed from a sequence of codepoints in hexadecimal + * + * @param string $seq Sequence of codepoints, separated with a space + * @return string UTF-8 string + */ + protected function hexseq_to_utf($seq) + { + return implode('', array_map(array($this, 'hex_to_utf'), explode(' ', $seq))); + } + + /** + * Convert a codepoint in hexadecimal to a UTF-8 char + * + * @param string $hex Codepoint, in hexadecimal + * @return string UTF-8 char + */ + protected function hex_to_utf($hex) + { + return $this->cp_to_utf(hexdec($hex)); + } + + /** + * Convert a codepoint to a UTF-8 char + * + * @param integer $cp Unicode codepoint + * @return string UTF-8 string + */ + protected function cp_to_utf($cp) + { + if ($cp > 0xFFFF) + { + return chr(0xF0 | ($cp >> 18)) . chr(0x80 | (($cp >> 12) & 0x3F)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F)); + } + else if ($cp > 0x7FF) + { + return chr(0xE0 | ($cp >> 12)) . chr(0x80 | (($cp >> 6) & 0x3F)) . chr(0x80 | ($cp & 0x3F)); + } + else if ($cp > 0x7F) + { + return chr(0xC0 | ($cp >> 6)) . chr(0x80 | ($cp & 0x3F)); + } + else + { + return chr($cp); + } + } + + // chunked download helper + static protected function download($url, $to) + { + $target = $to . '/' . basename($url); + + if (file_exists($target)) + { + return; + } + + if (!$fpr = fopen($url, 'rb')) + { + throw new RuntimeException("Failed to download $url"); + } + + if (!$fpw = fopen($target, 'wb')) + { + throw new RuntimeException("Failed to open $target for writing"); + } + + $chunk = 32768; + + while (!feof($fpr)) + { + fwrite($fpw, fread($fpr, $chunk)); + } + fclose($fpr); + fclose($fpw); + } +} -- cgit v1.2.1 From 14891cdf4ecf0e3ee164962be6685b8563f37a15 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Erik=20Fr=C3=A8rejean?= Date: Mon, 31 Jan 2011 12:58:18 +0100 Subject: [ticket/10011] Tests don't work on PHP < 5.3 Due to the usage of `__DIR__` for the file includes the tests can't be ran on systems with PHP < 5.3. Replace all occurances of `__DIR__` with `dirname(__FILE__)`. PHPBB3-10011 --- tests/utf/normalizer_test.php | 10 +++++----- tests/utf/utf8_clean_string_test.php | 2 +- tests/utf/utf8_wordwrap_test.php | 2 +- 3 files changed, 7 insertions(+), 7 deletions(-) (limited to 'tests/utf') diff --git a/tests/utf/normalizer_test.php b/tests/utf/normalizer_test.php index 9a9011c0fe..38b4ec1b6b 100644 --- a/tests/utf/normalizer_test.php +++ b/tests/utf/normalizer_test.php @@ -7,7 +7,7 @@ * */ -require_once __DIR__ . '/../../phpBB/includes/utf/utf_normalizer.php'; +require_once dirname(__FILE__) . '/../../phpBB/includes/utf/utf_normalizer.php'; /** * @group slow @@ -16,8 +16,8 @@ class phpbb_utf_normalizer_test extends phpbb_test_case { static public function setUpBeforeClass() { - self::download('http://www.unicode.org/Public/UNIDATA/NormalizationTest.txt', __DIR__.'/data'); - self::download('http://www.unicode.org/Public/UNIDATA/UnicodeData.txt', __DIR__.'/data'); + self::download('http://www.unicode.org/Public/UNIDATA/NormalizationTest.txt', dirname(__FILE__).'/data'); + self::download('http://www.unicode.org/Public/UNIDATA/UnicodeData.txt', dirname(__FILE__).'/data'); } public function test_normalizer() @@ -62,7 +62,7 @@ class phpbb_utf_normalizer_test extends phpbb_test_case $tested_chars = array(); - $fp = fopen(__DIR__.'/data/NormalizationTest.txt', 'rb'); + $fp = fopen(dirname(__FILE__).'/data/NormalizationTest.txt', 'rb'); while (!feof($fp)) { $line = fgets($fp); @@ -117,7 +117,7 @@ class phpbb_utf_normalizer_test extends phpbb_test_case */ public function test_invariants(array $tested_chars) { - $fp = fopen(__DIR__.'/data/UnicodeData.txt', 'rb'); + $fp = fopen(dirname(__FILE__).'/data/UnicodeData.txt', 'rb'); while (!feof($fp)) { diff --git a/tests/utf/utf8_clean_string_test.php b/tests/utf/utf8_clean_string_test.php index 148297ad4b..e5a771eafa 100644 --- a/tests/utf/utf8_clean_string_test.php +++ b/tests/utf/utf8_clean_string_test.php @@ -7,7 +7,7 @@ * */ -require_once __DIR__ . '/../../phpBB/includes/utf/utf_tools.php'; +require_once dirname(__FILE__) . '/../../phpBB/includes/utf/utf_tools.php'; class phpbb_utf_utf8_clean_string_test extends phpbb_test_case { diff --git a/tests/utf/utf8_wordwrap_test.php b/tests/utf/utf8_wordwrap_test.php index fbc947b92a..03fa9dc38c 100644 --- a/tests/utf/utf8_wordwrap_test.php +++ b/tests/utf/utf8_wordwrap_test.php @@ -7,7 +7,7 @@ * */ -require_once __DIR__ . '/../../phpBB/includes/utf/utf_tools.php'; +require_once dirname(__FILE__) . '/../../phpBB/includes/utf/utf_tools.php'; class phpbb_utf_utf8_wordwrap_test extends phpbb_test_case { -- cgit v1.2.1 From 28801d01985dcee7c94ad5c54ca5f9ee34592026 Mon Sep 17 00:00:00 2001 From: Nils Adermann Date: Fri, 10 Jun 2011 02:16:19 +0200 Subject: [ticket/10206] Failed unicode.org downloads no longer terminate tests Instead a warning is triggered and an explanation echo'd. PHPBB3-10206 --- tests/utf/normalizer_test.php | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-) (limited to 'tests/utf') diff --git a/tests/utf/normalizer_test.php b/tests/utf/normalizer_test.php index 38b4ec1b6b..f78dba8004 100644 --- a/tests/utf/normalizer_test.php +++ b/tests/utf/normalizer_test.php @@ -298,12 +298,14 @@ class phpbb_utf_normalizer_test extends phpbb_test_case if (!$fpr = fopen($url, 'rb')) { - throw new RuntimeException("Failed to download $url"); + echo "Failed to download $url\n"; + return; } if (!$fpw = fopen($target, 'wb')) { - throw new RuntimeException("Failed to open $target for writing"); + echo "Failed to open $target for writing\n"; + return; } $chunk = 32768; -- cgit v1.2.1