I have to read a css file and remove its commentaries, so i've decided to use the preg_replace function from php:
$original = '
/*
foo
*/
.test, input[type="text"]{ color: blue; }
/* bar */
a:hover{
text-decoration: underline;
color: red;
background: #FFF;
}';
echo preg_replace('/\/\*.*\*\//s', '', $original);
the problem is that its losing the line .test, input[type="text"]{ color: blue; }
Change .* to .*?.
echo preg_replace('#/\*.*?\*/#s', '', $original);
This will only remove /* up to the closest */ rather than the furthest.
I would approach it as follows:
\/\*((?!\*\/).*?)\*\/
\/\* # Start-of-comment
((?!\*\/).*?) # Any character 0 or more times, lazy,
# without matching an end-of-comment
\*\/ # End-of-comment
Demo
Try this:
$buffer = preg_replace('/\*[^*]*\*+([^/][^*]*\*+)*/', '', $buffer);
It doesn't work for //Comments here, but it works for the rest of /* Comment here */, even for multiple lines.
for //Comments here, use this one:
$buffer = preg_replace('/\/\*.*?\*\//s', '', $buffer);
Related
I am showing a table in HTML on my web page where a check mark is shown properly (I am using ✔ for the bold check mark).
I am using classic-asp to display the HTML. The html buildup (tableOutput) is then posted to a PHP script ($output = $_POST["output"]) that uses mPDF.php (version 6.0) to print to PDF the same HTML table and somehow the check mark wont print correctly (%u2714 in printed on my PDF). All the rest of the table is printed correctly, only the check mark has a problem.
I tried adding the Symbola.ttf and Symbola_hint.ttf fonts in the mpdf\ttfonts folder but it didnt do the trick.
HTML (classic-asp):
tableOutput = tableOutput & "<TD class='center pass_x' style='font-family:symbola;'>✔</TD>"
PHP (create_report_pdf.php):
$output = $_POST["output"]; //This is the html buildup tableOutput discussed previously
$header = Irrelevant;
$footer= Irrelevant;
$mpdf = new mPDF( 'utf-8', 'A4-L', 0, '', 5, 5, 20, 20);
$mpdf->allow_charset_conversion = true;
$mpdf->WriteHTML($style,1);
$mpdf->SetHTMLHeader( $header );
$mpdf->SetHTMLFooter( $footer );
$mpdf->WriteHTML( $output );
$mpdf->Output($temp ,'F' );
config_fonts.php (I addded symbola.ttf and Symbola_hint.ttf in the mpdf\ttfonts folder):
$this->fontdata = array (
"symbola" => array (
'R' => "Symbola.ttf",
'I' => "Symbola_hint.ttf",
),
CSS (PHP $style variable points to create_report_pdf.css):
.report-table{
border: 1px solid black;
border-collapse: collapse;
font-size: 7pt;
width: 100%;
}
th,td{
font-size: 7pt;
border: 1px solid black !important;
vertical-align : middle;
}
.center{ text-align: center; vertical-align: middle; }
INPUT{
border-color:#ffffff !important;
}
.pf{ width: 45px; }
.fix-cell{ width: 90px; }
.dr-column, .dr-input{ width: 100px; }
.comments, .comments-input{ width: 130px; }
Thank you very much for your help
There are two possible solutions:
Replacing manually the %u2714 with the respective html entity ✔ by doing the following:
$output = str_replace('%u2714', '✔', $output);
Implement more generic approach that will handle all such special chars. The approach is based on the fact that %uXXXX is a non-standard notation scheme for URL-encoding Unicode characters. So you need to convert %uXXXX notation to HTML entity notation &#xXXXX; and then this can be decoded to actual UTF-8 by html_entity_decode.
$output = preg_replace('/%u([0-9A-F]+)/', '&#x$1;', $output);
$output = html_entity_decode($output, ENT_COMPAT, 'UTF-8');
I am trying to set the background of the html body dynamically. Basically if a file exists, use it, otherwise use the default. But it keeps using the default regardless of whether the file exists or not.
<style>
body
{
padding-top: 50px;
background-image: url("<?php
clearstatcache();
if(file_exists("/profile_img/".$profileData["ID"]."_bg.jpg"))
{
echo "/profile_img/".$profileData["ID"]."_bg.jpg?". rand(5, 15);
}
else
{
//echo "/profile_img/".$profileData["ID"]."_bg.jpg?". rand(5, 15);
echo "/profile_img/default.jpg?". rand(5, 15);
}
?>");
background-size: cover;
background-position: 50% 50%;
}
</style>
I have tried using the file (the commented line) and it works. I can not see why this doesn't work
Some issues:
Using / will be absolute, causing it to look in the root directory.
Always check vars are set before using.
All that's changing is the filename, so you can use a ternary, which will reduce alot of that code.
<?php
$background = isset($profileData["ID"]) && file_exists('./profile_img/'.$profileData["ID"].'_bg.jpg')
? $profileData["ID"].'_bg.jpg' : 'default.jpg';
?>
<style>
body {
padding-top: 50px;
background-image: url("/profile_img/<?= $background.'?_='.microtime(true) ?>");
background-size: cover;
background-position: 50% 50%;
}
</style>
If it's still not working:
Check the file actually exists.
I'm building emails from external variables of CSS and HTML and then combining them as inline styles. Before they are combined, however, I want to alter the CSS.
In the below, I want to remove the padding and margin from the body. How should I do this - preg_match or is there something "cleaner"?
$css = "body {
background: #fff;
color: #000;
padding: 0;
margin: 0;
font: 83%/1.4 verdana,sans-serif;
-webkit-text-size-adjust: 100%;
-ms-text-size-adjust: 100%;
}
a, a:link, a:visited, a:hover, a:active {
background: transparent;
text-decoration: none;
cursor: pointer;
}";
Also, it's well worth mentioning that I am getting the CSS (then inline-styling) like this with the below so maybe the DOMDocument can help me - that seems cleaner, but I don't know where to start.
use TijsVerkoyen\CssToInlineStyles\CssToInlineStyles;
$dom = new DOMDocument();
#$dom->loadHTML($row['htmlbody']); //from a database
$cssobj = $dom->getElementsByTagName('style');
$bodyobj = $dom->getElementsByTagName('body')->item(0);
$plainbody = new DOMDocument();
foreach ($bodyobj->childNodes as $child){
$plainbody->appendChild($plainbody->importNode($child, true));
}
$body = $plainbody->saveHTML();
$css = $cssobj->item(0)->nodeValue;
$cssToInlineStyles = new CssToInlineStyles();
$originalemail = $cssToInlineStyles->convert(
$body,
$css
);
If you know the exact text that you want to remove, you can use a str_replace() from php. Example: $str_replace('margin:0;','',$css)
or
$str = preg_replace('/margin:(.*?);/', '', $css)
How can I set style to var_dump() function and PHP errors style, like on the next image?
At the moment I have next view of var_dump() (with <pre>var_dump(...)</pre>, without it will be all in one line) and just plain text for errors.
I searched something for PHP colored errors, var_dump styles, but could not find anything.
I used OpenServer as localhost, and on previous version I had same styles for errors, but now just plain text. Is it real to customize?
Use this code. I've been using it for years. I can't even remember where it originally came from.
function var_dump_pretty($data, $label='', $return = false) {
$debug = debug_backtrace();
$callingFile = $debug[0]['file'];
$callingFileLine = $debug[0]['line'];
ob_start();
var_dump($data);
$c = ob_get_contents();
ob_end_clean();
$c = preg_replace("/\r\n|\r/", "\n", $c);
$c = str_replace("]=>\n", '] = ', $c);
$c = preg_replace('/= {2,}/', '= ', $c);
$c = preg_replace("/\[\"(.*?)\"\] = /i", "[$1] = ", $c);
$c = preg_replace('/ /', " ", $c);
$c = preg_replace("/\"\"(.*?)\"/i", "\"$1\"", $c);
$c = preg_replace("/(int|float)\(([0-9\.]+)\)/i", "$1() <span class=\"number\">$2</span>", $c);
// Syntax Highlighting of Strings. This seems cryptic, but it will also allow non-terminated strings to get parsed.
$c = preg_replace("/(\[[\w ]+\] = string\([0-9]+\) )\"(.*?)/sim", "$1<span class=\"string\">\"", $c);
$c = preg_replace("/(\"\n{1,})( {0,}\})/sim", "$1</span>$2", $c);
$c = preg_replace("/(\"\n{1,})( {0,}\[)/sim", "$1</span>$2", $c);
$c = preg_replace("/(string\([0-9]+\) )\"(.*?)\"\n/sim", "$1<span class=\"string\">\"$2\"</span>\n", $c);
$regex = array(
// Numberrs
'numbers' => array('/(^|] = )(array|float|int|string|resource|object\(.*\)|\&object\(.*\))\(([0-9\.]+)\)/i', '$1$2(<span class="number">$3</span>)'),
// Keywords
'null' => array('/(^|] = )(null)/i', '$1<span class="keyword">$2</span>'),
'bool' => array('/(bool)\((true|false)\)/i', '$1(<span class="keyword">$2</span>)'),
// Types
'types' => array('/(of type )\((.*)\)/i', '$1(<span class="type">$2</span>)'),
// Objects
'object' => array('/(object|\&object)\(([\w]+)\)/i', '$1(<span class="object">$2</span>)'),
// Function
'function' => array('/(^|] = )(array|string|int|float|bool|resource|object|\&object)\(/i', '$1<span class="function">$2</span>('),
);
foreach ($regex as $x) {
$c = preg_replace($x[0], $x[1], $c);
}
$style = '
/* outside div - it will float and match the screen */
.dumpr {
margin: 2px;
padding: 2px;
background-color: #fbfbfb;
float: left;
clear: both;
}
/* font size and family */
.dumpr pre {
color: #000000;
font-size: 9pt;
font-family: "Courier New",Courier,Monaco,monospace;
margin: 0px;
padding-top: 5px;
padding-bottom: 7px;
padding-left: 9px;
padding-right: 9px;
}
/* inside div */
.dumpr div {
background-color: #fcfcfc;
border: 1px solid #d9d9d9;
float: left;
clear: both;
}
/* syntax highlighting */
.dumpr span.string {color: #c40000;}
.dumpr span.number {color: #ff0000;}
.dumpr span.keyword {color: #007200;}
.dumpr span.function {color: #0000c4;}
.dumpr span.object {color: #ac00ac;}
.dumpr span.type {color: #0072c4;}
';
$style = preg_replace("/ {2,}/", "", $style);
$style = preg_replace("/\t|\r\n|\r|\n/", "", $style);
$style = preg_replace("/\/\*.*?\*\//i", '', $style);
$style = str_replace('}', '} ', $style);
$style = str_replace(' {', '{', $style);
$style = trim($style);
$c = trim($c);
$c = preg_replace("/\n<\/span>/", "</span>\n", $c);
if ($label == ''){
$line1 = '';
} else {
$line1 = "<strong>$label</strong> \n";
}
$out = "\n<!-- Dumpr Begin -->\n".
"<style type=\"text/css\">".$style."</style>\n".
"<div class=\"dumpr\">
<div><pre>$line1 $callingFile : $callingFileLine \n$c\n</pre></div></div><div style=\"clear:both;\"> </div>".
"\n<!-- Dumpr End -->\n";
if($return) {
return $out;
} else {
echo $out;
}
}
You get the colored output when you ínstall and enable Xdebug:
Xdebug replaces PHP's var_dump() function for displaying variables. Xdebug's version includes different colors for different types and places limits on the amount of array elements/object properties, maximum depth and string lengths. There are a few other functions dealing with variable display as well.
You can enable/disable this with the ini setting xdebug.overload_var_dump
By default Xdebug overloads var_dump() with its own improved version for displaying variables when the html_errors php.ini setting is set to 1. In case you do not want that, you can set this setting to 0, but check first if it's not smarter to turn off html_errors.
Check the documentation for further information.
Note that you do not want to have the Xdebug extension installed on a production server as it will significantly slow down code execution.
In xdebug 3 there is no more xdebug.overload_var_dump setting. When you upgrade to xdebug 3 and you use xdebug.mode=debug as said in xdebug upgrade guide, you will not get colored output anymore. To get the same result as it was in xdebug 2 you must set xdebug.mode to develop,debug
xdebug.mode=develop,debug
Of course if you only want colored output you can only use develop mode but then you lost step debugging. With both options you get colored output and also step debugging. You can specify as many modes as you want/need. They must be separated with ,. All available modes are explained here https://xdebug.org/docs/all_settings#mode
You can also use this extension for a colored debug: (this is very easy to install)
http://www.sitepoint.com/var_dump-introducing-symfony-vardumper/
Symfony VarDumper is a component designed to replace your var_dumps. It performs essentially the same functionality, but provides you with much, much more information in a much prettier format. It’s the var_dump you’ve always wanted.
Xdebug is what you are looking for.
Sample scripts for install on Ubuntu:
[Search for Xdebug]
$ apt-cache search xdebug
[Install Xdebug]
$ sudo apt-get install php-xdebug
[Restart Apache to make it work]
$ sudo /etc/init.d/apache2 restart
In latest version of xdebug (now is 3) you should just :
1 . open php.ini
2 . change (xdebug.mode=debug) to (xdebug.mode=develop,debug)
How would you go about Cutting strings short so it doesnt go to the next line in a div tag For example my message string contained the following:
We prefer questions that can be answered, not just discussed. Provide details. Write clearly and simply. If your question is about this website, ask it on meta instead.
And i want to preferably display it as
We prefer questions that can be answered, not just discussed. Provide ... Read More
Im thinking cutting the string short using PHP but then you have the following problem
!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! ... Read More
You can see that the above string can still be expanded a bit more. Any way to do this or Not, Thanks in Advance
How you can shorten a string is already answered, but your main question:
How would you go about Cutting strings short so it doesnt go to the next line in a div tag
this will not work in most cases.
for example:
iiiiiiiiii
wwwwwwwwww
You see the problem ?
this only works if your chars have the same width like:
iiiiiiiiii
wwwwwwwwww
Using PHP, you can use the wordwrap function to perform truncation:
function truncate_to_length($text, $length, $text_if_longer = "...")
{
if (strlen($text) > $length)
{
$text = wordwrap($text, $length);
$text = substr($text, 0, strpos($text, "\n"));
return $text . $text_if_longer;
}
return $text;
}
The following mostly works; the main issue is that the "...Read more" notice can cover up part of a letter. It also uses a bunch of presentational elements and requires JS. It makes me feel unclean.
The outermost element (of class .string) is used to set the overall size. .text holds the text to display. Between .string and .text in the descendent hierarchy is an element with a large width. This puts all the text in .text on one line. JS is used to show or hide the "...Read more" notice.
<style type="text/css">
.string {
height: 1em;
width: 25%;
position: relative; /* create containing block for children */
border: 1px solid black;
overflow: hidden;
background: white; /* or whatever, as long as it's not "transparent". */
}
.string .line {
width: 5000px; /* long enough for you? */
}
.string .notice {
position: absolute;
top: 0;
right: 0;
z-index: 1;
display: none;
background: inherit; /* so that the notice will completely cover up whatever's beneath */
}
</style>
<div class="string">
<div class="line"><span class="text">We prefer questions that can be answered,
not just discussed. Provide details. Write clearly and simply. If your
question is about this website, ask it on meta instead.</span></div>
</div>
<hr />
<div class="string">
<div class="line"><span class="text">Lorem ipsum dolor sit amet.</span></div>
</div>
<script>
function isOverflowed(elt) {
return elt.offsetWidth > elt.parentNode.parentNode.clientWidth;
}
function getNotice(textElt) {
try {
return (textElt.parentNode.parentNode.getElementsByClassName('notice'))[0];
} catch (e) {
return {style: {}};
}
}
function show(elt) {
elt.style.display = 'block';
}
function hide(elt) {
elt.style.display = 'none';
}
function showReadMoreNotices() {
var text=document.getElementsByClassName('text');
for (var i=0; i<text.length; ++i) {
if (isOverflowed(text[i])) {
show(getNotice(text[i]));
} else {
hide(getNotice(text[i]));
}
}
}
var strings = document.getElementsByClassName('string');
var notice = document.createElement('div');
notice.appendChild(document.createTextNode('\u2026Read more'));
notice.className = 'notice';
for (var i=0; i<strings.length; ++i) {
strings[i].appendChild(notice.cloneNode(true));
}
showReadMoreNotices();
/* use your favorite event registration method here. Not that the traditional
* model is my favorite, it's just simple.
*/
window.onresize = showReadMoreNotices;
</script>
Check the size, if the size is greater than your maximum, take the leftmost characters.
Leftmost # = Total size - size("... read more").
Then append the read more to the left most characters.
Jacob
Here is an example taken from http://snippetdb.com/php/truncate-string:
function Truncate ($str, $length=10, $trailing='...')
{
// take off chars for the trailing
$length-=strlen($trailing);
if (strlen($str) > $length)
{
// string exceeded length, truncate and add trailing dots
return substr($str,0,$length).$trailing;
}
else
{
// string was already short enough, return the string
$res = $str;
}
return $res;
}
Oooo i came up with something that works but not works completely... Id like to share
div.string{
height:15px;
overflow:hidden;
}
It solves the problem that it will hide the whole word that wont fit in to the end of line.., because of the overflow and the height is set to one line only. However the above still dosent do the Problem shown by Rufinus:
iiiiiiiiiiiiii
wwwwwwwwwwwwww
This is a function picked from smarty templates that shortens strings without cutting them.
function str_short($string, $length = 80, $etc = '...', $break_words = false, $middle = false) {
if ($length == 0)
return '';
if (strlen($string) > $length) {
$length -= min($length, strlen($etc));
if (!$break_words && !$middle) {
$string = preg_replace('/\s+?(\S+)?$/', '', substr($string, 0, $length+1));
}
if(!$middle) {
return substr($string, 0, $length) . $etc;
} else {
return substr($string, 0, $length/2) . $etc . substr($string, -$length/2);
}
} else {
return $string;
}
}
It all depends on the font. Of course, you can use a monospace font, but that isn't a solution. Anyway, why would you like to do this? Even if you manage to make it work on a certain font, it might be that an user doesn't have it. Then a different font will be used, and everything may break...
It would be basically impossible to do this in PHP with a non monospace font (even with a monospace font, it is difficult to tell exactly how long a string in going to be in a certain browser. The only way I can think of doing this in a browser is to put it in a and set the width of a the span to be a certain width then use overflow: hidden.
Depending on what browser you're using, this one line of CSS could do the trick:
div.string {
word-wrap: break-word;
}
If you know what true-type font the browser is using, you can use the imagettfbbox() in the GD library to calculate the width of your text. Then iterate and drop characters off of the end of the text until it fits into your div.
This may not be very efficient, I know that, but it does the job. You could fine-tune and optimize for your specific solution.
<?
$width = 280; // max. width in pixels that the text should have
$font = 'C:\Windows\Fonts\verdana.TTF';
$text = 'The quick brown fox jumps over the lazy dog';
function getWidth($text) {
list($x1, , $x2) = imagettfbbox(12, 0, $font, $text);
return $x2-$x1;
}
while (getWidth($text.'...') > $width) $text = substr($text, 0, -1);
?>
<style type='text/css'>
#my_div {
font-family: Verdana;
font-size: 12pt;
border: 1px solid black;
width: 280px;
padding: 0
}
</style>
<div id='my_div'>
<?=$text?>...
</div>