aboutsummaryrefslogtreecommitdiffstats
diff options
context:
space:
mode:
-rw-r--r--phpBB/includes/acp/acp_forums.php16
-rw-r--r--phpBB/includes/utf/utf_tools.php43
2 files changed, 34 insertions, 25 deletions
diff --git a/phpBB/includes/acp/acp_forums.php b/phpBB/includes/acp/acp_forums.php
index 0bbaf96dec..d31404fa22 100644
--- a/phpBB/includes/acp/acp_forums.php
+++ b/phpBB/includes/acp/acp_forums.php
@@ -987,24 +987,14 @@ class acp_forums
}
/**
- * Replace Emojis and other 4bit UTF-8 chars not allowed by MySql to NCR.
+ * Replace Emojis and other 4bit UTF-8 chars not allowed by MySql to UCR / NCR.
* Using their Numeric Character Reference's Hexadecimal notation.
- * Doesn't interfere with Japanese or Cyrillic etc.
- *
- * @see https://www.w3.org/TR/xml11/
- * @see https://www.opentag.com/xfaq_charrep.htm
*/
- if (preg_match_all('/[\x{10000}-\x{10FFFF}]/u', $forum_data_ary['forum_name'], $matches))
- {
- foreach ($matches as $key => $emoji)
- {
- $forum_data_ary['forum_name'] = str_replace($emoji, utf8_encode_ncr($emoji), $forum_data_ary['forum_name']);
- }
- }
+ $forum_data_ary['forum_name'] = utf8_encode_ucr($forum_data_ary['forum_name']);
/**
* This should never happen again.
- * Leaving the fallback hre just in case there will be the need of it.
+ * Leaving the fallback here just in case there will be the need of it.
*/
if (preg_match_all('/[\x{10000}-\x{10FFFF}]/u', $forum_data_ary['forum_name'], $matches))
{
diff --git a/phpBB/includes/utf/utf_tools.php b/phpBB/includes/utf/utf_tools.php
index 89de454427..bb155aeae5 100644
--- a/phpBB/includes/utf/utf_tools.php
+++ b/phpBB/includes/utf/utf_tools.php
@@ -418,24 +418,43 @@ function utf8_recode($string, $encoding)
}
/**
-* Replace all UTF-8 chars that are not in ASCII with their NCR
-*
-* @param string $text UTF-8 string in NFC
-* @return string ASCII string using NCRs for non-ASCII chars
-*/
+ * Replace some special UTF-8 chars that are not in ASCII with their UCR.
+ * using their Numeric Character Reference's Hexadecimal notation.
+ *
+ * Doesn't interfere with Japanese or Cyrillic etc.
+ * Unicode character visualization will depend on the character support
+ * of your web browser and the fonts installed on your system.
+ *
+ * @see https://en.wikibooks.org/wiki/Unicode/Character_reference/1F000-1FFFF
+ *
+ * @param string $text UTF-8 string in NFC
+ * @return string ASCII string using NCR for non-ASCII chars
+ */
+function utf8_encode_ucr($text)
+{
+ return preg_replace_callback('/[\\xF0-\\xF4].../', 'utf8_encode_ncr_callback', $text);
+}
+
+/**
+ * Replace all UTF-8 chars that are not in ASCII with their NCR
+ * using their Numeric Character Reference's Hexadecimal notation.
+ *
+ * @param string $text UTF-8 string in NFC
+ * @return string ASCII string using NCRs for non-ASCII chars
+ */
function utf8_encode_ncr($text)
{
return preg_replace_callback('#[\\xC2-\\xF4][\\x80-\\xBF]{1,3}#', 'utf8_encode_ncr_callback', $text);
}
/**
-* Callback used in encode_ncr()
-*
-* Takes a UTF-8 char and replaces it with its NCR. Attention, $m is an array
-*
-* @param array $m 0-based numerically indexed array passed by preg_replace_callback()
-* @return string A HTML NCR if the character is valid, or the original string otherwise
-*/
+ * Callback used in utf8_encode_ncr() and utf8_encode_ucr()
+ *
+ * Takes a UTF-8 char and replaces it with its NCR. Attention, $m is an array
+ *
+ * @param array $m 0-based numerically indexed array passed by preg_replace_callback()
+ * @return string A HTML NCR if the character is valid, or the original string otherwise
+ */
function utf8_encode_ncr_callback($m)
{
return '&#' . utf8_ord($m[0]) . ';';