makary/phpbb
1
0
Fork 0
mirror of https://github.com/phpbb/phpbb.git synced 2025-06-28 14:18:52 +00:00
Code Issues Projects Releases Packages Wiki Activity

[feature/postgresql-fulltext-search] remove mbstring support

Browse source 
Remove the usage of mbstring regex engine when PCRE does not support UTF8
since its a requirement for phpbb 3.1

PHPBB3-9730
This commit is contained in:
Dhruv Goel 2012-06-12 04:18:39 +05:30
parent 8981399ae1
commit 84054afc9d
2 changed files with 8 additions and 83 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

89
phpBB/includes/search/fulltext_postgres.php
View file

@ -29,7 +29,6 @@ class phpbb_search_fulltext_postgres extends phpbb_search_base
var $tsearch_query;
var $common_words = array();
var $pcre_properties = false;
var $mbstring_regex = false;
var $tsearch_usable = false;
public function __construct(&$error)
@ -44,11 +43,6 @@ class phpbb_search_fulltext_postgres extends phpbb_search_base
$this->pcre_properties = true;
}
if (function_exists('mb_ereg'))
{
$this->mbstring_regex = true;
}
if ($db->sql_layer == 'postgres')
{
$pgsql_version = explode('.', substr($db->sql_server_info(), 10));
@ -115,41 +109,10 @@ class phpbb_search_fulltext_postgres extends phpbb_search_base
$split_keywords = preg_replace("#[\"\n\r\t]+#", ' ', trim(htmlspecialchars_decode($keywords)));
// Split words
if ($this->pcre_properties)
{
$split_keywords = preg_replace('#([^\p{L}\p{N}\'*"()])#u', '$1$1', str_replace('\'\'', '\' \'', trim($split_keywords)));
}
else if ($this->mbstring_regex)
{
$split_keywords = mb_ereg_replace('([^\w\'*"()])', '\\1\\1', str_replace('\'\'', '\' \'', trim($split_keywords)));
}
else
{
$split_keywords = preg_replace('#([^\w\'*"()])#u', '$1$1', str_replace('\'\'', '\' \'', trim($split_keywords)));
}
if ($this->pcre_properties)
{
$matches = array();
preg_match_all('#(?:[^\p{L}\p{N}*"()]|^)([+\-|]?(?:[\p{L}\p{N}*"()]+\'?)*[\p{L}\p{N}*"()])(?:[^\p{L}\p{N}*"()]|$)#u', $split_keywords, $matches);
$this->split_words = $matches[1];
}
else if ($this->mbstring_regex)
{
mb_regex_encoding('UTF-8');
mb_ereg_search_init($split_keywords, '(?:[^\w*"()]|^)([+\-|]?(?:[\w*"()]+\'?)*[\w*"()])(?:[^\w*"()]|$)');
while (($word = mb_ereg_search_regs()))
{
$this->split_words[] = $word[1];
}
}
else
{
$matches = array();
preg_match_all('#(?:[^\w*"()]|^)([+\-|]?(?:[\w*"()]+\'?)*[\w*"()])(?:[^\w*"()]|$)#u', $split_keywords, $matches);
$this->split_words = $matches[1];
}
$split_keywords = preg_replace('#([^\p{L}\p{N}\'*"()])#u', '$1$1', str_replace('\'\'', '\' \'', trim($split_keywords)));
$matches = array();
preg_match_all('#(?:[^\p{L}\p{N}*"()]|^)([+\-|]?(?:[\p{L}\p{N}*"()]+\'?)*[\p{L}\p{N}*"()])(?:[^\p{L}\p{N}*"()]|$)#u', $split_keywords, $matches);
$this->split_words = $matches[1];
// to allow phrase search, we need to concatenate quoted words
$tmp_split_words = array();
@ -260,42 +223,10 @@ class phpbb_search_fulltext_postgres extends phpbb_search_base
global $config;
// Split words
if ($this->pcre_properties)
{
$text = preg_replace('#([^\p{L}\p{N}\'*])#u', '$1$1', str_replace('\'\'', '\' \'', trim($text)));
}
else if ($this->mbstring_regex)
{
$text = mb_ereg_replace('([^\w\'*])', '\\1\\1', str_replace('\'\'', '\' \'', trim($text)));
}
else
{
$text = preg_replace('#([^\w\'*])#u', '$1$1', str_replace('\'\'', '\' \'', trim($text)));
}
if ($this->pcre_properties)
{
$matches = array();
preg_match_all('#(?:[^\p{L}\p{N}*]|^)([+\-|]?(?:[\p{L}\p{N}*]+\'?)*[\p{L}\p{N}*])(?:[^\p{L}\p{N}*]|$)#u', $text, $matches);
$text = $matches[1];
}
else if ($this->mbstring_regex)
{
mb_regex_encoding('UTF-8');
mb_ereg_search_init($text, '(?:[^\w*]|^)([+\-|]?(?:[\w*]+\'?)*[\w*])(?:[^\w*]|$)');
$text = array();
while (($word = mb_ereg_search_regs()))
{
$text[] = $word[1];
}
}
else
{
$matches = array();
preg_match_all('#(?:[^\w*]|^)([+\-|]?(?:[\w*]+\'?)*[\w*])(?:[^\w*]|$)#u', $text, $matches);
$text = $matches[1];
}
$text = preg_replace('#([^\p{L}\p{N}\'*])#u', '$1$1', str_replace('\'\'', '\' \'', trim($text)));
$matches = array();
preg_match_all('#(?:[^\p{L}\p{N}*]|^)([+\-|]?(?:[\p{L}\p{N}*]+\'?)*[\p{L}\p{N}*])(?:[^\p{L}\p{N}*]|$)#u', $text, $matches);
$text = $matches[1];
// remove too short or too long words
$text = array_values($text);
@ -858,10 +789,6 @@ class phpbb_search_fulltext_postgres extends phpbb_search_base
<dt><label>' . $user->lang['FULLTEXT_POSTGRES_PCRE'] . '</label><br /><span>' . $user->lang['FULLTEXT_POSTGRES_PCRE_EXPLAIN'] . '</span></dt>
<dd>' . (($this->pcre_properties) ? $user->lang['YES'] : $user->lang['NO']) . ' (PHP ' . PHP_VERSION . ')</dd>
</dl>
<dl>
<dt><label>' . $user->lang['FULLTEXT_POSTGRES_MBSTRING'] . '</label><br /><span>' . $user->lang['FULLTEXT_POSTGRES_MBSTRING_EXPLAIN'] . '</span></dt>
<dd>' . (($this->mbstring_regex) ? $user->lang['YES'] : $user->lang['NO']). '</dd>
</dl>
<dl>
<dt><label>' . $user->lang['FULLTEXT_POSTGRES_TS_NAME'] . '</label><br /><span>' . $user->lang['FULLTEXT_POSTGRES_TS_NAME_EXPLAIN'] . '</span></dt>
<dd><select name="config[fulltext_postgres_ts_name]">';

2
phpBB/language/en/acp/search.php
View file

@ -64,12 +64,10 @@ $lang = array_merge($lang, array(
'FULLTEXT_POSTGRES_INCOMPATIBLE_VERSION' => 'The PostgreSQL fulltext backend can only be used with PostgreSQL.',
'FULLTEXT_POSTGRES_TS_NOT_USABLE' => 'The PostgreSQL fulltext backend can only be used with PostgreSQL 8.3 and above.',
'FULLTEXT_POSTGRES_TOTAL_POSTS' => 'Total number of indexed posts',
'FULLTEXT_POSTGRES_MBSTRING' => 'Support for non-latin UTF-8 characters using mbstring:',
'FULLTEXT_POSTGRES_PCRE' => 'Support for non-latin UTF-8 characters using PCRE:',
'FULLTEXT_POSTGRES_TS_NAME' => 'Tsearch2 Configuration Profile:',
'FULLTEXT_POSTGRES_MIN_WORD_LEN' => 'Minimum word length for keywords',
'FULLTEXT_POSTGRES_MAX_WORD_LEN' => 'Maximum word length for keywords',
'FULLTEXT_POSTGRES_MBSTRING_EXPLAIN' => 'If PCRE does not have unicode character properties, the search backend will try to use mbstring’s regular expression engine.',
'FULLTEXT_POSTGRES_PCRE_EXPLAIN' => 'This search backend requires PCRE unicode character properties, only available in PHP 4.4, 5.1 and above, if you want to search for non-latin characters.',
'FULLTEXT_POSTGRES_TS_NAME_EXPLAIN' => 'The Tsearch2 configuration profile used to determine the parser and dictionary.',
'FULLTEXT_POSTGRES_MIN_WORD_LEN_EXPLAIN' => 'Words with at least this many characters will be included in the query to the database.',