Using the Ajax helper for CakePHP (currently 1.2.3.8166) to provide an $ajax->autoComplete list of results, and giving a result list back as the rendered view, if you use the mouse (and even the mouse wheel) to scroll results, all is well. Using the arrow keys, on the other hand, has the nasty effect of awkwardly scrolling the view: if I press down, the select box and the whole page move to the bottom of the browser's view pane; pressing up has the opposite effect of moving it to the top.
Has anyone else noticed this behaviour, and thought of something? the resulting list is provided by, e.g., this code (this gets $people from the autoComplete() function in the controller):
<ul>
<?php foreach($people as $person): ?>
<li><?php echo $person['Person']['id']; ?></li>
<?php endforeach; ?>
</ul>
(Just an example, I actually show the id and name / surname / commercial name).
The CSS for the list is as follows:
div.auto_complete {
position: absolute;
width: 250px;
background-color: white;
border: 1px solid #888;
margin: 0px; padding: 0px;
}
div.auto_complete ul{
list-style: none;
margin: 0px;
}
I received the answer to this problem on the cake-php newsgroup (available on http://groups.google.com/group/cake-php ).
The poster pointed to this page with the solution, and I copy it here:
Open the controls.js file (should be in app/webroot/js)
Search for the markPrevious function and change it to:
markPrevious: function() {
if (this.index > 0) {
this.index--;
} else {
this.index = this.entryCount-1;
this.update.scrollTop = this.update.scrollHeight;
}
selection = this.getEntry(this.index);
selection_top = selection.offsetTop;
if (selection_top < this.update.scrollTop) {
this.update.scrollTop = this.update.scrollTop-
selection.offsetHeight;
}
},
Search the markNext function and change it to:
markNext: function() {
if(this.index < this.entryCount-1) {
this.index++;
} else {
this.index = 0;
this.update.scrollTop = 0;
}
selection = this.getEntry(this.index);
selection_bottom = selection.offsetTop+selection.offsetHeight;
if(selection_bottom > this.update.scrollTop+this.update.offsetHeight) {
this.update.scrollTop = this.update.scrollTop + selection.offsetHeight;
}
},
Search for the updateChoices function and change lines
this.stopIndicator();
this.index = 0;
to
this.stopIndicator();
this.update.scrollTop = 0;
this.index = 0;
Finally, try the behavior. If it doesn't work at first, try deleting the cache files in app/tmp/cache (or clear your favorite server-side cache), your browser cache, and try again. Clearing app/tmp/cache worked for me.
Related
I am using the Divi Builder in WordPress to make a list of patents in the accordion format. However, I would like to format the accordion title (of each accordion element) so that on the left is the patent#, and on the right is the title of the patent.
To do this, I would like to know how I can split the title into 2 separate sections; 1 where I can put the title number, and the other where I can type the name of the patent. Sort of like this:
Another sketch to show what I want it to look like: sketch of the goal
Again I am using Divi Builder to do this as I am a novice using WordPress. But I am assuming I would have to write some custom CSS to format the title in this way in the Divi Builder.
First is it possible to do what I am trying to do?
If so how can I do this (in the Divi Builder environment)?
Unfortunately, not all of our requirements can be met by means of ready-made tools for building a website.
In such cases, we are forced to write "fixes" in such individual cases.
I offer you a js script that does the following:
Searches for titles with text that meets the following requirements:
Starts with #
After a hash without spaces, numbers
Example: #111
Wraps the found in the span with the class specified in the configuration (at the very top of the file)
I tested here
(function() {
// Config
let classFirstHeading = 'my-heading';
let headerSelector = '.et_pb_toggle_title'; // You can replace to your selector of heading
// end config;
var DOMReady = function(callback) {
if (document.readyState === "interactive" || document.readyState === "complete") {
callback();
} else if (document.addEventListener) {
document.addEventListener("DOMContentLoaded", callback);
} else if (document.attachEvent) {
document.attachEvent("onreadystatechange", function() {
if (document.readyState != "loading") {
callback();
}
});
}
}
function updateHeading() {
let headings = document.querySelectorAll(headerSelector);
headings.forEach(el => {
let textHeading = el.innerText;
let regexp = /^\#\d+/;
if(textHeading && regexp.test(textHeading.trim())) {
el.innerHTML = textHeading.replace(regexp, (match) => {
return '<span class="'+classFirstHeading+'">'+match+'</span> ';
});
}
});
}
window.updateHeading = updateHeading; // Give access from outside
DOMReady(() => {
updateHeading();
/* Test, you can remove this section */
let styleEl = document.createElement('style'); styleEl.type = 'text/css';
let css = `
.${classFirstHeading} {
display: inline-block;
font-size: 23px;
font-weight: 700;
margin-right: 15px;
padding: 10px 15px 10px 0;
border-right: 2px solid #000;
}
`;
styleEl.appendChild(document.createTextNode(css));
document.querySelector('head').appendChild(styleEl);
/*************************************/
});
})();
<h5 class="et_pb_toggle_title">#111 What is your refund policy?</h5>
If you don't know where to insert such code, install the Custom CSS and JS in Header & Footer plugin and just copy my js code to your page where you use the accordion
I am using css variables in my angular7 application. Everything works fine on other browsers. But IE is not supporting css variables. Is there a way to make it work on IE. Can Autoprefixer do this?
color: var(--primary, #7F583F);
According to caniuse.com, of current browsers only IE, Edge (older versions) and Opera Mini do not support CSS variables. This polyfil appears to work on all three really well.
This is an attempt at a very basic CSS variables (custom properties) polyfil. In reality this is more of a partial polyfill as it will not cover variables inside of variables, DOM scoping or anything else "fancy". Just taking variables declared anywhere in the CSS and then re-parsing the CSS for var() statements and replacing them in browsers that don't natively support CSS variables.
I try to test this polyfil in IE 11 and looks like it is working with it.
/*!
* css-var-polyfill.js - v1.0.0
*
* Copyright (c) 2018 Aaron Barker <http://aaronbarker.net>
* Released under the MIT license
*
* Date: 2018-03-09
*/
let cssVarPoly = {
init: function() {
// first lets see if the browser supports CSS variables
// No version of IE supports window.CSS.supports, so if that isn't supported in the first place we know CSS variables is not supported
// Edge supports supports, so check for actual variable support
if (window.CSS && window.CSS.supports && window.CSS.supports('(--foo: red)')) {
// this browser does support variables, abort
console.log('your browser supports CSS variables, aborting and letting the native support handle things.');
return;
} else {
// edge barfs on console statements if the console is not open... lame!
console.log('no support for you! polyfill all (some of) the things!!');
document.querySelector('body').classList.add('cssvars-polyfilled');
}
cssVarPoly.ratifiedVars = {};
cssVarPoly.varsByBlock = {};
cssVarPoly.oldCSS = {};
// start things off
cssVarPoly.findCSS();
cssVarPoly.updateCSS();
},
// find all the css blocks, save off the content, and look for variables
findCSS: function() {
let styleBlocks = document.querySelectorAll('style:not(.inserted),link[rel="stylesheet"]');
// we need to track the order of the style/link elements when we save off the CSS, set a counter
let counter = 1;
// loop through all CSS blocks looking for CSS variables being set
[].forEach.call(styleBlocks, function(block) {
// console.log(block.nodeName);
let theCSS;
if (block.nodeName === 'STYLE') {
// console.log("style");
theCSS = block.innerHTML;
cssVarPoly.findSetters(theCSS, counter);
} else if (block.nodeName === 'LINK') {
// console.log("link");
cssVarPoly.getLink(block.getAttribute('href'), counter, function(counter, request) {
cssVarPoly.findSetters(request.responseText, counter);
cssVarPoly.oldCSS[counter] = request.responseText;
cssVarPoly.updateCSS();
});
theCSS = '';
}
// save off the CSS to parse through again later. the value may be empty for links that are waiting for their ajax return, but this will maintain the order
cssVarPoly.oldCSS[counter] = theCSS;
counter++;
});
},
// find all the "--variable: value" matches in a provided block of CSS and add them to the master list
findSetters: function(theCSS, counter) {
// console.log(theCSS);
cssVarPoly.varsByBlock[counter] = theCSS.match(/(--.+:.+;)/g) || [];
},
// run through all the CSS blocks to update the variables and then inject on the page
updateCSS: function() {
// first lets loop through all the variables to make sure later vars trump earlier vars
cssVarPoly.ratifySetters(cssVarPoly.varsByBlock);
// loop through the css blocks (styles and links)
for (let curCSSID in cssVarPoly.oldCSS) {
// console.log("curCSS:",oldCSS[curCSSID]);
let newCSS = cssVarPoly.replaceGetters(cssVarPoly.oldCSS[curCSSID], cssVarPoly.ratifiedVars);
// put it back into the page
// first check to see if this block exists already
if (document.querySelector('#inserted' + curCSSID)) {
// console.log("updating")
document.querySelector('#inserted' + curCSSID).innerHTML = newCSS;
} else {
// console.log("adding");
var style = document.createElement('style');
style.type = 'text/css';
style.innerHTML = newCSS;
style.classList.add('inserted');
style.id = 'inserted' + curCSSID;
document.getElementsByTagName('head')[0].appendChild(style);
}
};
},
// parse a provided block of CSS looking for a provided list of variables and replace the --var-name with the correct value
replaceGetters: function(curCSS, varList) {
// console.log(varList);
for (let theVar in varList) {
// console.log(theVar);
// match the variable with the actual variable name
let getterRegex = new RegExp('var\\(\\s*' + theVar + '\\s*\\)', 'g');
// console.log(getterRegex);
// console.log(curCSS);
curCSS = curCSS.replace(getterRegex, varList[theVar]);
// now check for any getters that are left that have fallbacks
let getterRegex2 = new RegExp('var\\(\\s*.+\\s*,\\s*(.+)\\)', 'g');
// console.log(getterRegex);
// console.log(curCSS);
let matches = curCSS.match(getterRegex2);
if (matches) {
// console.log("matches",matches);
matches.forEach(function(match) {
// console.log(match.match(/var\(.+,\s*(.+)\)/))
// find the fallback within the getter
curCSS = curCSS.replace(match, match.match(/var\(.+,\s*(.+)\)/)[1]);
});
}
// curCSS = curCSS.replace(getterRegex2,varList[theVar]);
};
// console.log(curCSS);
return curCSS;
},
// determine the css variable name value pair and track the latest
ratifySetters: function(varList) {
// console.log("varList:",varList);
// loop through each block in order, to maintain order specificity
for (let curBlock in varList) {
let curVars = varList[curBlock];
// console.log("curVars:",curVars);
// loop through each var in the block
curVars.forEach(function(theVar) {
// console.log(theVar);
// split on the name value pair separator
let matches = theVar.split(/:\s*/);
// console.log(matches);
// put it in an object based on the varName. Each time we do this it will override a previous use and so will always have the last set be the winner
// 0 = the name, 1 = the value, strip off the ; if it is there
cssVarPoly.ratifiedVars[matches[0]] = matches[1].replace(/;/, '');
});
};
// console.log(ratifiedVars);
},
// get the CSS file (same domain for now)
getLink: function(url, counter, success) {
var request = new XMLHttpRequest();
request.open('GET', url, true);
request.overrideMimeType('text/css;');
request.onload = function() {
if (request.status >= 200 && request.status < 400) {
// Success!
// console.log(request.responseText);
if (typeof success === 'function') {
success(counter, request);
}
} else {
// We reached our target server, but it returned an error
console.warn('an error was returned from:', url);
}
};
request.onerror = function() {
// There was a connection error of some sort
console.warn('we could not get anything from:', url);
};
request.send();
}
};
cssVarPoly.init();
:root {
--externalcolor: red;
--samename: orange;
--samename: #0f0;
--foo: green;
--FOO: #0f0;
--halfsuccess: orange;
--success: green;
--success2: #0f0;
}
html {
font-family: var(--fontsans);
}
.success {
color: green;
}
.fail {
color: red;
}
span {
display: inline-block;
margin: 5px;
}
.samename {
color: var(--samename);
}
.demo1 {
color: #f00;
color: var(--success);
}
.demo2 {
color: #f00;
color: var( --success2);
}
.demo3 {
color: #f00;
color: var(--halfsuccess);
color: var(--success);
}
.demo4 {
color: red;
border-color: #f00;
}
.inlineoverlink {
color: #f00;
}
p {
padding: var(--spacing-l);
}
.lower {
color: var(--foo);
}
.upper {
color: var(--FOO);
}
.externalcolor {
color: var(--externalcolor);
}
.fallback {
color: #f00;
color: var(--wrongname, green);
}
// for the top documentation
.supports {
color: green;
.no {
display:none;
}
}
.showforpolyfill {
display:none;
}
.cssvars-polyfilled {
.supports {
color: red;
.no {
display:inline;
}
}
.showforpolyfill {
display:inline;
}
.hideforpolyfill {
display:none;
}
}
.hide,
.hide-the-docs .documentation {
display:none;
}
/* declare some font-family stuff at bottom of file to reflect on stuff above it*/
:root {
--fontsans: arial;
}
<!-- Copy below for codepen update -->
<h1>CSS Variables Polyfill</h1>
<p>This is now managed (and available for PRs) at https://github.com/aaronbarker/css-variables-polyfill.</p>
<p>
This is an attempt at a very basic CSS variables (custom properties) polyfil. In reality this is more of a <em>partial</em> polyfill as it will not cover variables inside of variables, DOM scoping or anything else "fancy". Just taking variables declared anywhere in the CSS and
then re-parsing the CSS for var() statements and replacing them in browsers that don't natively support CSS variables.
</p>
<p>According to caniuse.com, of current browsers only IE, Edge and Opera Mini do not support CSS variables. This polyfil appears to work on all three really well. I don't see why this wouldn't work on older browsers as well, but I haven't been able to test it on them yet.</p>
<p>As far as we can tell your browser <span class="supports">does <span class="no">not</span> support</span> native CSS variables. <span class="showforpolyfill">That means if you see green tests results below, it is thanks to the polyfill :).</span> <span class="hideforpolyfill">All the green test results below are actually native CSS Variable support. Good job using a good browser :)</span></p>
<h3>Does this work on externally CSS files?</h3>
<p>Yes!</p>
<h3>Even ones loaded from another domain?</h3>
<p>To go across domain, CSS needs to be served up with <code>Access-Control-Allow-Origin:*</code> headers.</p>
</div>
Toggle documentation (for Opera Mini vs Codepen issue)
<style>
:root {
--newcolor: #0f0;
}
.inlineoverlink {
color: var(--success2);
}
</style>
<h2>Tests</h2>
<p>On mosts tests (unless otherwise noted) success will be green text. We start with a <code>color:red;</code> and then override it with a <code>color:var(--success);</code> (or similar) which is green.</p>
<ul>
<li><span class="samename">declare same variable over and over</span></li>
<li><span class="demo1">no whitespace on var() calls</span></li>
<li><span class="demo2">whitespace on var() calls</span></li>
<li><span class="demo3">Multiple variables in same call. orange means first var worked, green var worked</span></li>
<li><span class="inlineoverlink">orange if link won, green if style after link won</span></li>
<li><span class="lower">--foo: lowercase foo</span></li>
<li><span class="upper">--FOO: uppercase FOO</span></li>
<li><span class="fallback">uses fallback <code>--var(--wrongname, green)</code></span></li>
<li><span class="demo-import">css declared in an <code>#import</code></span> - not polyfilled yet. Identfied with a suggested fix, but will require a bit of a re-write (to use document.styleSheets), so haven't done it yet.</li>
</ul>
<h2>Tests on external, cross-domain file</h2>
<div class="documentation">
<p><strong>Edge</strong> appears to be working well on Edge 13. Edge 12 was having some problems.</p>
<p><strong>Opera mini</strong> seems to work well too. This demo fails because not all the page is displayed, but I think that is a codepen issue, not a polyfill issue. When the upper documentation is removed, all tests display well.</p>
<p><strong>IE 11</strong> seems to do fine.</p>
</div>
<ul>
<li><span class="demo4">Gets stuff from external .css file. Should start red and change to green on LINK load. border proves the CSS loaded, missing colors means script didn't get parsed and reinserted</span></li>
<li><span class="externalcolor">--externalcolor: should start red and change to green on LINK load</span></li>
<li><span class="externalfallback">uses fallback. should be green</span></li>
</ul>
<p>Another set of text under the test for Opera Mini testing.</p>
<!-- Copy above for codepen update -->
Testing result:
References:
(1) Codepen example link
(2) aaronbarker/css-variables-polyfill
I have two buttons are set position equal to "absolute", when the LastPass addon's bar dipslays, they displays wrong because LastPass had inserted an iframe to my webpage:
LastPass iFrame
<iframe id="lpiframe74158812" src="chrome-extension://hdokiejnpimakedhajhdlcegeplioahd/overlay.html?&add=1" scrolling="no"
style="height: 27px; width: 1263px; border: 0px;"></iframe>
The CSS:
.button-bar {
width: 175px;
float: left;
top: 113px;
text-align: right;
right: 20px;
position: absolute;
}
Image: http://i.stack.imgur.com/Rq5Z5.png
How can I avoid this case? Thanks so much!
I had this same issue and I found that last pass editing the DOM of my webpage! LastPass had added a div right after the body tag of my page.
<div id="lptopspacer48468746" style="height: 27px;"></div>
It looks like the div id is random, so I can't strip it out. I think this problem is with lastPass. I don't think there is a way to truly fix it.
Also annoyed by this <div id="lptopspacer[0-9]+" style="height:40px"></div> inserted in any page monitored by firefox lastPass plugin (after the site has shown a login form), I've come up with a jQuery solution.
Only adding some CSS rules don't seems to works as the div is obviously added after page load by a script. Changing style or trying to remove the div just after page load doesn't works either.
So this snippet run a delayed function to hide the div when found, or stop running after 5 attempts if no lastPass plugin is affecting the document.
<script>
var log = function(msg) {
if (console && console.log){
console.log(msg)
}
};
$(document).ready(function(){
var maxTry = 5, lptopHideTimeout;
var clearLptop = function(delay) {
var $lptop = $("div[id^='lptopspacer']");
if (lptopHideTimeout) {
window.clearTimeout(lptopHideTimeout);
}
if ($lptop.length && $lptop.is(':visible')) {
log("** Hiding lastPass lptopspacer...");
$lptop.css( "display","none" );
}
else {
maxTry -= 1;
if (maxTry > 0) {
log("## No lastPass lptopspacer div found yet. Retrying in " + (delay/1000) + ' second...');
lptopHideTimeout = window.setTimeout(function(){
clearLptop(delay);
},delay);
}
else {
log("## Giving up after too much attempts.");
}
}
};
clearLptop(500);
});
</script>
Better late than never if others also are having this problem. I had it to, that Lp spacer, in my case it was generated because I had Mcafee safekey installed on the computer. It showed up with a notice on every pageload on my website and caused an lp spacer that broke my site with a white bar on top.
Uninstalling Mcafee safekey solved it for me.
I need to set variable in LESS based on conditions.
#body-background-color: black;
#post-background-color: white;
#post-stacked-effect: true;
.content-padding() when ( #body-background-color = #post-background-color ) and not (#post-stacked-effect = true){
#content-padding: 0px;
}
.content-padding() when not (#body-background-color = #post-background-color), (#post-stacked-effect = true) {
#content-padding: 20px;
}
body {
.content-padding();
content: "#{post-stacked-effect}";
padding: #content-padding;
background-color: #body-background-color;
.post {
background-color: #post-background-color;
}
.some-math {
margin: -#content-padding -#content-padding #content-padding;
}
}
#content-padding is always 0px. Can you please help me fix that condition?
Its a bug in PHP Less Compiler. I have to do PHP workaround.
When I'm preparing variables for that compile i added function which calculates the padding in php and inject it into LESS.
Its only one row in PHP.
$lessvars["content-padding"] =
( $lessvars["body-background-color"] == $lessvars["post-background-color"] &&
!filter_var($lessvars["post-stacked-effect"], FILTER_VALIDATE_BOOLEAN) )
? "0px" : "20px";
You are creating a new scope with the 'when' guarded mixin which isn't visible outside.
So this isn't even supposed to work, and certainly isn't a bug.
You can instead create a function and call that from the 'when'.
#body-background-color: black;
#post-background-color: white;
#post-stacked-effect: true;
body {
.padding(#padding)
{
padding: #padding;
.some-math
{
margin: #padding + 2em;
}
}
& when ( #body-background-color = #post-background-color ) and not (#post-stacked-effect = true){
.padding(0px);
}
& when not (#body-background-color = #post-background-color), (#post-stacked-effect = true) {
.padding(20px);
}
}
I've removed the irrelevant css, and renamed it because stackoverflow formats hyphenated function names weirdly.
Is there a way to detect vertical scroll distance with a media query?
It seems that media queries are designed around detecting the medium (shocking right :P) so things like browser height are testable, but not specifically how far down the page is scrolled.
If is not possible, but you know a way in JS (not jQuery) feel free to post!
First off, the accepted answer doesn't work.
The correct name is
window.onscroll
and not
window.onScroll
https://developer.mozilla.org/en-US/docs/Web/API/window.onscroll
Second, this is horribly inefficient as the function is called way more than it needs to and can make the page laggy when scrolled. From John Resig:
http://ejohn.org/blog/learning-from-twitter/
Much better to use a timer that runs every 150 ms or so - something like:
var scrolled = false;
window.onscroll = function() {
scrolled = true;
}
setInterval(function(){
if (scrolled) {
scrolled = false;
// scrollFunction()
}
}, 150);
I don't believe it's possible with a CSS media query, but I do know that the scroll height can be found in JavaScript using window.pageYOffset. If you wanted to run this value through a function every time the users scrolled up or down on a page, you could do something like
window.onscroll = function() {
scrollFunctionHere(window.pageYOffset);
};
Or just:
window.onscroll = scrollFunctionHere;
If the function itself checked the value of window.pageYOffset.
For more advice on how to do use window.onscroll efficiently in JavaScript, refer to mynameistechno's answer.
Important note on efficiency: running a function every single time a scroll event is emitted can tear through CPU cycles if anything non-trivial is performed in the callback. Instead, it is good practice to only allow a callback to run so many times per second. This has been termed "debouncing".
Simple debounced scroll event handler code below. Notice how the text toggles between "HELLO" and "WORLD" every 250ms, rather than every single frame:
var outputTo = document.querySelector(".output");
var timeout_debounce;
window.addEventListener("scroll", debounce);
function debounce(event) {
if(timeout_debounce) {
return;
}
timeout_debounce = setTimeout(clearDebounce, 250);
// Pass the event to the actual callback.
actualCallback(event);
}
function clearDebounce() {
timeout_debounce = null;
}
function actualCallback(event) {
// Perform your logic here with no CPU hogging.
outputTo.innerText = outputTo.innerText === "HELLO"
? "WORLD"
: "HELLO";
}
p {
padding: 40vh;
margin: 20vh;
background: blue;
color: white;
}
<p class="output">Test!</p>
In Jquery you have the method .scrollTop()
http://api.jquery.com/scrolltop/
This example make a div scroll with the window scroll.
$(window).scroll(function(){
$("div").css("margin-top", $(window).scrollTop())
});
Here is one way to solution.f https://jsfiddle.net/oravckzx/1/
$(window).scroll(function(){
$('.post-sidebar').each(function(){
var ScrollTopVar = $(window).scrollTop();
var OffsetTopVar = $(this).offset().top;
var OuterHeightVar = $(this).outerHeight();
var PositionVar = OffsetTopVar-(OuterHeightVar*1.1);
if (ScrollTopVar >= PositionVar) {
$('.hide') .css('background','green').css('font-size','12px')
$('.post-sidebar') .css('background','orange').css('font-size','12px')
$('.hide') .css('background','green').css('font-size','12px')
$('.post-sidebar') .css('background','gray').css('font-size','12px')
document.getElementById("demo3").innerHTML = ScrollTopVar;
document.getElementById("demo2").innerHTML = PositionVar;
}else {
$('.hide') .css('background','yellow').css('background','yellow')
$('.hide') .css('background','yellow')
$('.hide') .css('background','yellow')
document.getElementById("demo").innerHTML = ScrollTopVar;
document.getElementById("demo12").innerHTML = ScrollTopVar;
document.getElementById("demo13").innerHTML = ScrollTopVar;
document.getElementById("demo14").innerHTML = ScrollTopVar;
document.getElementById("demo2").innerHTML = PositionVar;
}
});
});
.red {height:100px;
background:red;margin-bottom:20px;}
.hide {height:50px;background:blue;margin-bottom:20px;}
.post-sidebar {
height:50px;
background:yellow;
margin-bottom:20px;
box-sizing: border-box;
display: block;
font: normal 700 34px Lato, sans-serif;
padding-right: 20px;
width: 452px;
}
.p {
font: normal 700 14px Lato, sans-serif;
}
<script src="https://cdnjs.cloudflare.com/ajax/libs/jquery/3.3.1/jquery.min.js"></script>
<div class="red"><p id="demo"></div>
<div class="red"></div><p id="demo12">ok
<div class="red"><p id="demo13">ok</div>
<div class="post-sidebar"><p id="demo2"></p><p>lizard</p></div>
<div class="hide"><p id="demo2">If reaches to chosen class in html, saves value of ScrollTopVar (as distance from top) to separate variable as eventually PositionVar (which is the distance from top to the chosen class in html, - and that specific distance depends of device which is doing it), and then does certain action if ScrollTopVar value matches or exceeds PositionVar value. Else sets it back if needed, if not including Else it remains as it once met the value. <p id="demo3"></div>
<div class="red"><p id="demo14">ok</div>
<div class="red"><p id="demo"></div>
<div class="red"><p id="demo"></div>