summaryrefslogtreecommitdiff
path: root/program/lib/Roundcube/rcube_text2html.php
diff options
context:
space:
mode:
authorAleksander Machniak <alec@alec.pl>2014-05-11 11:03:45 +0200
committerAleksander Machniak <alec@alec.pl>2014-05-11 11:03:45 +0200
commiteda92ed4c0d2735144df8fa2136584de69634bdb (patch)
treeab1f817904ed365e657380d784107ef4e14f18ce /program/lib/Roundcube/rcube_text2html.php
parent638ebf69c4bdc3717d8ae535ec3b1f4b753f5856 (diff)
Improved display of plain text messages and text to HTML conversion (#1488937)
Now instead of <pre> we use <div class="pre"> styled with monospace font. We replace whitespace characters with non-breaking spaces where needed. I.e. plain text is always unwrappable, until it uses format=flowed, in such a case only flowed paragraphs are wrappable. Also conversion of text to HTML in compose editor was modified in the same way.
Diffstat (limited to 'program/lib/Roundcube/rcube_text2html.php')
-rw-r--r--program/lib/Roundcube/rcube_text2html.php277
1 files changed, 277 insertions, 0 deletions
diff --git a/program/lib/Roundcube/rcube_text2html.php b/program/lib/Roundcube/rcube_text2html.php
new file mode 100644
index 000000000..5da771ac8
--- /dev/null
+++ b/program/lib/Roundcube/rcube_text2html.php
@@ -0,0 +1,277 @@
+<?php
+
+/**
+ +-----------------------------------------------------------------------+
+ | This file is part of the Roundcube Webmail client |
+ | Copyright (C) 2008-2014, The Roundcube Dev Team |
+ | |
+ | Licensed under the GNU General Public License version 3 or |
+ | any later version with exceptions for skins & plugins. |
+ | See the README file for a full license statement. |
+ | |
+ | PURPOSE: |
+ | Converts plain text to HTML |
+ +-----------------------------------------------------------------------+
+ | Author: Aleksander Machniak <alec@alec.pl> |
+ +-----------------------------------------------------------------------+
+ */
+
+/**
+ * Converts plain text to HTML
+ *
+ * @package Framework
+ * @subpackage Utils
+ */
+class rcube_text2html
+{
+ /**
+ * Contains the HTML content after conversion.
+ *
+ * @var string $html
+ */
+ protected $html;
+
+ /**
+ * Contains the plain text.
+ *
+ * @var string $text
+ */
+ protected $text;
+
+ /**
+ * Configuration
+ *
+ * @var array $config
+ */
+ protected $config = array(
+ // non-breaking space
+ 'space' => "\xC2\xA0",
+ // enables format=flowed parser
+ 'flowed' => false,
+ // enables wrapping for non-flowed text
+ 'wrap' => false,
+ // line-break tag
+ 'break' => "<br>\n",
+ // prefix and suffix (wrapper element)
+ 'begin' => '<div class="pre">',
+ 'end' => '</end>',
+ // enables links replacement
+ 'links' => true,
+ );
+
+
+ /**
+ * Constructor.
+ *
+ * If the plain text source string (or file) is supplied, the class
+ * will instantiate with that source propagated, all that has
+ * to be done it to call get_html().
+ *
+ * @param string $source Plain text
+ * @param boolean $from_file Indicates $source is a file to pull content from
+ * @param array $config Class configuration
+ */
+ function __construct($source = '', $from_file = false, $config = array())
+ {
+ if (!empty($source)) {
+ $this->set_text($source, $from_file);
+ }
+
+ if (!empty($config) && is_array($config)) {
+ $this->config = array_merge($this->config, $config);
+ }
+ }
+
+ /**
+ * Loads source text into memory, either from $source string or a file.
+ *
+ * @param string $source Plain text
+ * @param boolean $from_file Indicates $source is a file to pull content from
+ */
+ function set_text($source, $from_file = false)
+ {
+ if ($from_file && file_exists($source)) {
+ $this->text = file_get_contents($source);
+ }
+ else {
+ $this->text = $source;
+ }
+
+ $this->_converted = false;
+ }
+
+ /**
+ * Returns the HTML content.
+ *
+ * @return string HTML content
+ */
+ function get_html()
+ {
+ if (!$this->_converted) {
+ $this->_convert();
+ }
+
+ return $this->html;
+ }
+
+ /**
+ * Prints the HTML.
+ */
+ function print_html()
+ {
+ print $this->get_html();
+ }
+
+ /**
+ * Workhorse function that does actual conversion (calls _converter() method).
+ */
+ protected function _convert()
+ {
+ $text = stripslashes($this->text);
+
+ // Convert TXT to HTML
+ $this->html = $this->_converter($text);
+ $this->_converted = true;
+ }
+
+ /**
+ * Workhorse function that does actual conversion.
+ *
+ * @param string Plain text
+ */
+ protected function _converter($text)
+ {
+ // make links and email-addresses clickable
+ $attribs = array('link_attribs' => array('rel' => 'noreferrer', 'target' => '_blank'));
+ $replacer = new rcmail_string_replacer($attribs);
+
+ if ($this->config['flowed']) {
+ $flowed_char = 0x01;
+ $text = rcube_mime::unfold_flowed($text, chr($flowed_char));
+ }
+
+ // search for patterns like links and e-mail addresses and replace with tokens
+ if ($this->config['links']) {
+ $text = $replacer->replace($text);
+ }
+
+ // split body into single lines
+ $text = preg_split('/\r?\n/', $text);
+ $quote_level = 0;
+ $last = -1;
+
+ // find/mark quoted lines...
+ for ($n=0, $cnt=count($text); $n < $cnt; $n++) {
+ $flowed = false;
+ if ($this->config['flowed'] && ord($text[0]) == $flowed_char) {
+ $flowed = true;
+ $text[$n] = substr($text[$n], 1);
+ }
+
+ if ($text[$n][0] == '>' && preg_match('/^(>+ {0,1})+/', $text[$n], $regs)) {
+ $q = substr_count($regs[0], '>');
+ $text[$n] = substr($text[$n], strlen($regs[0]));
+ $text[$n] = $this->_convert_line($text[$n], $flowed || $this->config['wrap']);
+
+ if ($q > $quote_level) {
+ $text[$n] = $replacer->get_replacement($replacer->add(
+ str_repeat('<blockquote>', $q - $quote_level))) . $text[$n];
+ $last = $n;
+ }
+ else if ($q < $quote_level) {
+ $text[$n] = $replacer->get_replacement($replacer->add(
+ str_repeat('</blockquote>', $quote_level - $q))) . $text[$n];
+ $last = $n;
+ }
+ }
+ else {
+ $text[$n] = $this->_convert_line($text[$n], $flowed || $this->config['wrap']);
+ $q = 0;
+
+ if ($quote_level > 0) {
+ $text[$n] = $replacer->get_replacement($replacer->add(
+ str_repeat('</blockquote>', $quote_level))) . $text[$n];
+ }
+ }
+
+ $quote_level = $q;
+ }
+
+ if ($quote_level > 0) {
+ $text[$n] = $replacer->get_replacement($replacer->add(
+ str_repeat('</blockquote>', $quote_level))) . $text[$n];
+ }
+
+ $text = join("\n", $text);
+
+ // colorize signature (up to <sig_max_lines> lines)
+ $len = strlen($text);
+ $sig_max_lines = rcube::get_instance()->config->get('sig_max_lines', 15);
+
+ while (($sp = strrpos($text, "-- \n", $sp ? -$len+$sp-1 : 0)) !== false) {
+ if ($sp == 0 || $text[$sp-1] == "\n") {
+ // do not touch blocks with more that X lines
+ if (substr_count($text, "\n", $sp) < $sig_max_lines) {
+ $text = substr($text, 0, max(0, $sp))
+ .'<span class="sig">'.substr($text, $sp).'</span>';
+ }
+
+ break;
+ }
+ }
+
+ // insert url/mailto links and citation tags
+ $text = $replacer->resolve($text);
+
+ // replace \n before </blockquote>
+ $text = str_replace("\n</blockquote>", "</blockquote>", $text);
+
+ // replace line breaks
+ $text = str_replace("\n", $this->config['break'], $text);
+
+ return $this->config['begin'] . $text . $this->config['end'];
+ }
+
+ /**
+ * Converts spaces in line of text
+ */
+ protected function _convert_line($text, $is_flowed)
+ {
+ static $table;
+
+ if (empty($table)) {
+ $table = get_html_translation_table(HTML_SPECIALCHARS);
+ unset($table['?']);
+ }
+
+ // skip signature separator
+ if ($text == '-- ') {
+ return $text;
+ }
+
+ // replace HTML special characters
+ $text = strtr($text, $table);
+
+ $nbsp = $this->config['space'];
+
+ // replace some whitespace characters
+ $text = str_replace(array("\r", "\t"), array('', ' '), $text);
+
+ // replace spaces with non-breaking spaces
+ if ($is_flowed) {
+ $text = preg_replace_callback('/(^|[^ ])( +)/', function($matches) {
+ if (!strlen($matches[2])) {
+ return str_repeat($nbsp, strlen($matches[2]));
+ }
+ else {
+ return $matches[1] . ' ' . str_repeat($nbsp, strlen($matches[2])-1);
+ }
+ }, $text);
+ }
+ else {
+ $text = str_replace(' ', $nbsp, $text);
+ }
+
+ return $text;
+ }
+}