php字符串处理（邮件处理）
今日做单据邮件提醒功能，发现系统发出的邮件总是显示不全。调用发送邮件方法前记录下来的日志是完全的。
经查询原来是邮件格式对邮件单行有长度限制：
ps1：邮件字符限制
本标准对行字符数量有2种限制。每行字符数必须（MUST）不超过998个，应该（SHOULD）不超过78个， 除了CRLF。
ps2：邮件正文换行符规定
邮件正文一般由US-ASCII字符组成的多行文字。只有2个限制，如下：
-CR，LF必须同时出现为CRLF，他们不可以分开出现在正文里。
-正文的每行字符数必须（MUST）不超过998个，应该（SHOULD）不超过78个，除了CRLF。
（原文链接）：http://hi.baidu.com/zhufangtian/item/ba54cb5f09780413abf6d74b
所以在发送邮件时必须对邮件字符串进行换行处理
开始尝试用php
后来测试codeigniter框架邮件发送插件，发现它支持这种单行超长邮件的发送，于是查看源码。分析其邮件正文处理方法，
最终处理步骤:
1.将字符串全部转化为小写
2.将回车换行符替换为<br>。因为本项目邮件为html格式的
3.去除多余空格
4.替换a标签内容，防止标签内容超长被截断
5.wordwrap方法处理字符串自动换行。（）
在实际项目中发现换行符必须为\r\n(在windows情况下实际测试，网上很多都说为\n\r，个人感觉是说错了)，所以必须要规定wordwrap方法的替换参数，注意在替换格式后添加空格是为了有些邮件客户端（outlook）可以正常解析换行的属性标签，有些则不能（foxmail）
如截断后标签换行 tr和span中间有换行符，无空格。
<tr span='3'>6.替换回a标签内容
ps1：改造后邮件处理方法
//邮件文本替换
function __mailreplace($content, $type) {
//替换回车换行符（注意替换字符串后添加 空格，以便在wordwrap方法中可以将字符串换行）
if (strtolower ( $type ) == "html") {
$content = ltrim ( $content );
$content = str_replace ( array ('\n', '\r\n' ), "<br/> ", $content );
$content = str_replace ( array (chr ( 13 ), chr ( 13 ) . chr ( 10 ) ), "<br/> ", $content );
//替换空格
$content = str_replace ( " ", " ", $content );
//处理表格过长问题，将每行换行
$content = str_replace ( "</tr>", "</tr> ", $content );
} else {
$content = str_replace ( array ("<br>", "<br/>" ), "\r\n", $content );
$content = str_replace ( " ", " ", $content );
}
//替换多余空格
$content = preg_replace ( "| +|", " ", $content );
//替换a标签（防止链接过长被截断）
$unwrap = array ();
if (preg_match_all ( '|(<a.*>.*</a>)|U', $content, $matches )) {
for($i = 0; $i < count ( $matches ['0'] ); $i ++) {
$unwrap [] = $matches ['1'] [$i];
$content = str_replace ( $matches ['1'] [$i], "{{a-link" . $i . "}}", $content );
}
}
//过长字符串自动换行（添加空格，避免html标签内属性换行时无法正常解析）
$content = wordwrap ( $content, 75, "\r\n ", FALSE );
//替换回a标签
if (count ( $unwrap ) > 0) {
foreach ( $unwrap as $key => $val ) {
$content = str_replace ( "{{a-link" . $key . "}}", $val, $content );
}
}
return $content;
}ps3：codeigniter源码：
public function word_wrap($str, $charlim = '') {
// Se the character limit
if ($charlim == '') {
$charlim = ($this->wrapchars == "") ? "76" : $this->wrapchars;
}
// Reduce multiple spaces 去除多余空格
$str = preg_replace ( "| +|", " ", $str );
// Standardize newlines 替换回车 换行符
if (strpos ( $str, "\r" ) !== FALSE) {
$str = str_replace ( array ("\r\n", "\r" ), "\n", $str );
}
// If the current word is surrounded by {unwrap} tags we'll
// strip the entire chunk and replace it with a marker. 批量替换特定标签（改为批量替换a链接）
$unwrap = array ();
if (preg_match_all ( "|(\{unwrap\}.+?\{/unwrap\})|s", $str, $matches )) {
for($i = 0; $i < count ( $matches ['0'] ); $i ++) {
$unwrap [] = $matches ['1'] [$i];
$str = str_replace ( $matches ['1'] [$i], "{{unwrapped" . $i . "}}", $str );
}
}
// Use PHP's native public function to do the initial wordwrap.
// We set the cut flag to FALSE so that any individual words that are
// too long get left alone. In the next step we'll deal with them.
// 截断字符串
$str = wordwrap ( $str, $charlim, "\n", FALSE );
// Split the string into individual lines of text and cycle through them
$output = "";
foreach ( explode ( "\n", $str ) as $line ) {
// Is the line within the allowed character count?
// If so we'll join it to the output and continue
if (strlen ( $line ) <= $charlim) {
$output .= $line . $this->newline;
continue;
}
$temp = '';
while ( (strlen ( $line )) > $charlim ) {
// If the over-length word is a URL we won't wrap it
if (preg_match ( "!\[url.+\]|://|wwww.!", $line )) {
break;
}
// Trim the word down
$temp .= substr ( $line, 0, $charlim - 1 );
$line = substr ( $line, $charlim - 1 );
}
// If $temp contains data it means we had to split up an over-length
// word into smaller chunks so we'll add it back to our current line
if ($temp != '') {
$output .= $temp . $this->newline . $line;
} else {
$output .= $line;
}
$output .= $this->newline;
}
// Put our markers back
// 将替换字符串替换回来
if (count ( $unwrap ) > 0) {
foreach ( $unwrap as $key => $val ) {
$output = str_replace ( "{{unwrapped" . $key . "}}", $val, $output );
}
}
return $output;
}
