Accessing the composed DOM for a polymer element - polymer

I'm trying to access the browser-rendered DOM for a polymer element, without caring in the slightest which bits from from the "light" or "shadow" DOM as described in http://www.polymer-project.org/platform/shadow-dom.html, but I cannot find any API documentation on access functions that would let me do this.
I can see the .shadowRoots list that is on polymer elements, and I can see the regular .children property, but neither of these is particularly useful to get the content that is actually being shown on the page.
What is the correct way to get the DOM fragment that is currently visible in the browser? (or if someone knows where this is documented on the polymer site, I'd love to know that too. Google couldn't find it for me)

The general idea is to only consider your virtual DOM (and even better, only consider local DOM [divide and conquer ftw]).
Therefore, today, there is no (easy) way to examine the composed DOM.
There will be such a way eventually for certain specialized uses, but I encourage you to try to adopt a mental model that allows you to avoid asking this question.

Here is something I wrote for the same purpose:
function getComposedDOMChildren (root) {
if (root.shadowRoot) {
root = root.shadowRoot;
}
var children = [];
for (var i = 0; i < root.children.length; i++) {
if (root.children[i].tagName === 'CONTENT') {
children.push.apply(children, root.children[i].getDistributedNodes());
} else if (root.children[i].tagName === 'SHADOW') {
var shadowRoot = root;
while (!shadowRoot.host) {
shadowRoot = shadowRoot.parentNode;
}
children.push.apply(children, getComposedDOMChildren(shadowRoot.olderShadowRoot));
} else {
children.push(root.children[i]);
}
}
return children;
}
It returns an array with the children of the root as per the composed DOM.
It works by going into the shadow root and replacing all <content> elements with what is actually rendered in them and recursing with the earlier shadow root when <shadow> elements are found.
Of course, to get the entire DOM tree, one has to iterate through the returned array and invoke getComposedDOMChildren on each element.
Something like:
function traverse (root) {
var children = getComposedDOMChildren(root);
for (var i = 0; i < children.length; i++) {
traverse(children[i]);
}
}
Please correct me if this isn't right.

Related

How to improve ui Sentry.io breadcrumbs?

I was wondering if there is a good practice on how to write your HTML code in order to get better Sentry.io breadcrumbs inside of the issues.
It's not possible to identify the elements that the user has interacted and I think using CSS class or IDs for it is not the ideal - although we can customize the breadcrumbs, looks like it's not a good practice to get the text inside the tag as per some issues found on Sentry Github repository.
I was thinking about aria-label, does anyone has any advices on it?
Right now is very hard to understand the user steps when reading the breadcrumbs.
This can be solved using the beforeBreadcrumb hook / filtering events.
Simply add
beforeBreadcrumb(breadcrumb, hint) {
if (breadcrumb.category === 'ui.click') {
const { target } = hint.event;
if (target.ariaLabel) {
breadcrumb.message = target.ariaLabel;
}
}
return breadcrumb;
}
... to your Sentry.init() configuration.
Sentry.init({
dsn:...
Resulting in something like this:
Sentry.init({
dsn: '....',
beforeBreadcrumb(breadcrumb, hint) {
if (breadcrumb.category === 'ui.click') {
const { target } = hint.event;
if (target.ariaLabel) {
breadcrumb.message = target.ariaLabel;
}
}
return breadcrumb;
}
});
More about this here: sentry.io filtering events documentation

How to access html element text in chrome devtools console

I have a page with a whole bunch of blackquote tags.
In dev console I am typing document.getElementsByTagName("blockquote") that giving me an array.
But if I do document.getElementsByTagName("blockquote").innerText document.getElementsByTagName("blockquote").innerHTML document.getElementsByTagName("blockquote").textContent
document.getElementsByTagName("blockquote").outerText document.getElementsByTagName("blockquote").outerHTML
All return undefined
However if I inspect elements of the array document.getElementsByTagName("blockquote") I can see all above properties in place.
How to access at least one of them (innerText, outerHTLM, innerText, outerHTML, textContent) ?
Or if you want to access any specific element you can use index in array
for (var i=0; i <document.getElementsByTagName("blockquote").length; i++ ){
var singleElement = document.getElementsByTagName("blockquote")[i];
console.log(singleElement.innerHTML);
}
You need to iterate the array in order to access those properties. Something like this will work for them:
var elements = document.getElementsByTagName("blockquote");
for (var prop in elements)
{
if(elements.hasOwnProperty(prop)) {
console.log(elements[prop].innerHTML);
}
}
You can also try the following commands:
var elements = document.getElementsByTagName("blockquote");.This will return the list of elements.To access the text of your required element at i index
elements[i].value.

HTML parsing individual tables/not all data being parsed?

I'm rather new when it comes to Windows Phone 8 development and I've been toying around with a few things as part of the application I'm developing.
Right now I'm trying to parse information from a website such as the RuneScape 07 High Scores - http://services.runescape.com/m=hiscore_oldschool/hiscorepersonal.ws?user1=zezima
I'm using HTML Agility Pack and I'm able to parse some data (down to Woodcutting), but anything passed that doesn't appear? (Is that down to the size of my ListBox?)
Ideally, I'd like to be able to parse the table information individually rather than in one block like so:
public MainPage()
{
InitializeComponent();
HtmlWeb.LoadAsync("http://services.runescape.com/m=hiscore_oldschool/hiscorepersonal.ws?user1=zezima", DownLoadCompleted);
}
void DownLoadCompleted(object sender, HtmlDocumentLoadCompleted e)
{
if(e.Error == null)
{
HtmlDocument doc = e.Document;
if (doc != null)
{
var result = doc.DocumentNode.SelectNodes("//div[#id='contentHiscores']");
foreach (var htmlNode in result)
{
lBox.Items.Add(htmlNode.InnerText);
}
}
}
But if I try and access an individual table such as this one using
var result = doc.DocumentNode.SelectNodes("//div[#id='contentHiscores']/table/tbody/tr[5]/td[2]");
I get a NullReferenceException.
Is this possible or am I doing something exceptionally wrong?
You probably relied on a developper tools such as FireBug or Chrome, etc... to determine the XPATH for the nodes you're after.
You can' really do this as the XPATH given by such tools correspond to the in memory HTML DOM while the Html Agility Pack only knows about the raw HTML sent back by the server.
What you need to do is look at what's sent back (or just do a view source). You'll see there is no TBODY element for example. So you want to find anything discriminant, and use XPATH axes for example.
Here is a code that seems to work:
// get all TD nodes with ALIGN attribute set to left
foreach (var node in doc.DocumentNode.SelectNodes("//div[#id='contentHiscores']//td[#align='left']"))
{
var item = lBox.Items.Add(node.InnerText.Trim());
// use an 'XPATH axe': get all sibling TD nodes with ALIGN attribute set to 'right'
foreach (var sibling in node.SelectNodes("following-sibling::td[#align='right']"))
{
item.SubItems.Add(sibling.InnerText.Trim());
}
}

What is the best way to remove all tabindex attribute to html elements?

What is the best way to remove all tabindex attributes from html elements? GWt seems to put this attribute even it is not set anywhere in the code. It sets tabindex to -1.
I have the code below as working but it is tedious because I have to search every element according to its tag name and that slows the page loading. Any other suggestions? I'd prefer the solution not use javascript, as I am new to it.
NodeList<Element> input = this.getElement().getElementsByTagName("input");
if(input.getLength()>0)
{
for(int i=0; i<=input.getLength(); i++)
{
input.getItem(i).removeAttribute("tabIndex");
}
}
NodeList<Element> div = this.getElement().getElementsByTagName("div");
if(div.getLength()>0)
{
for(int i=0; i<=div.getLength(); i++)
{
div.getItem(i).removeAttribute("tabIndex");
}
}
I'm not entirely sure what you are asking then. You want to remove the tab index attribute. You either:
set the tabindex attribute to -1 manually in the HTML.
use the code you already have.
or use the simplified JQuery version in the other thread.
Perhaps I have misunderstood what you are trying to achieve?
EDIT
Okay perhaps this:
$(document).ready(function(){
$('input').removeAttr("tabindex");
});
This should remove it rather than set it to -1... hopefully. Sorry if I've misunderstood again!
JQuery removeAttr Link
Use querySelectorAll function which Returns a list of the elements within the document (using depth-first pre-order traversal of the document's nodes) that match the specified group of selectors.
function removeTagAttibute( attributeName ){
var allTags = '*';
var specificTags = ['ARTICLE', 'INPUT'];
var allelems = document.querySelectorAll( specificTags );
for(i = 0, j = 0; i < allelems.length; i++) {
allelems[i].removeAttribute( attributeName );
}
}
removeTagAttibute( 'tabindex' );
I finally figured it out.
I tried Javascirpt/jquery but they couldn't remove tabindexes because the page was not fully rendered yet - my panels are placed programmatically after window.load. What I did is make use of the RootPanel.class of gwt (which was being used already, but I didn't know).
The task: to get rid of all tabindex with -1 value, add type="tex/javascript" for all script tags, type="text/css" for style tags and out an alt to all img tags. These are all for the sake of html validation.
I am not sure this is the best way, it sure does add up to slow loading, but client is insisting that I do it. So here it is:
RootPanel mainPanel = RootPanel.get(Test_ROOT_PANEL_ID);
Widget widget = (Widget) getEntryView();
mainPanel.add((widget));
// Enable the view disable the loading view. There should always be
// the loading panel to disable.
Element mainPanelelement = DOM.getElementById(Test_ROOT_PANEL_ID);
Element loadingMessage = DOM.getElementById(LOADING_MESSAGE);
Element parent = loadingMessage.getParentElement();
if(parent!=null)
{
//i had to use prev sibling because it is the only way that I know of to access the body //tag that contains the scripts that are being generated by GWT ex.bigdecimal.js
Element body = parent.getPreviousSibling().getParentElement();
if(body!=null)
{
NodeList<Element> elms = body.getElementsByTagName("*");
if(elms.getLength()>0)
{
Element element=null;
for(int i=0; i<=elms.getLength(); i++)
{
if(elms.getItem(i)!=null)
{
element = elms.getItem(i);
if(element.getTagName().compareToIgnoreCase("script")==0)
element.setAttribute("type", "text/javascript");
else if(element.getTagName().compareToIgnoreCase("style")==0)
element.setAttribute("type", "text/css");
else if(element.getTagName().compareToIgnoreCase("img")==0)
{
if(element.getAttribute("alt")=="")
element.setAttribute("alt", element.getAttribute("title")!=" " ? element.getTitle() : " " );
}
else
{
if(element.getTabIndex()<=0)
element.removeAttribute("tabindex");
}
}
}
}
}
}
DOM.setStyleAttribute((com.google.gwt.user.client.Element) loadingMessage, "display", "none");
DOM.setStyleAttribute((com.google.gwt.user.client.Element) mainPanelelement, "display", "inline");
// Change cursor back to default.
RootPanel.getBodyElement().getStyle().setProperty("cursor", "default");
}

Remove attributes using HtmlAgilityPack

I'm trying to create a code snippet to remove all style attributes regardless of tag using HtmlAgilityPack.
Here's my code:
var elements = htmlDoc.DocumentNode.SelectNodes("//*");
if (elements!=null)
{
foreach (var element in elements)
{
element.Attributes.Remove("style");
}
}
However, I'm not getting it to stick? If I look at the element object immediately after Remove("style"). I can see that the style attribute has been removed, but it still appears in the DocumentNode object. :/
I'm feeling a bit stupid, but it seems off to me? Anyone done this using HtmlAgilityPack? Thanks!
Update
I changed my code to the following, and it works properly:
public static void RemoveStyleAttributes(this HtmlDocument html)
{
var elementsWithStyleAttribute = html.DocumentNode.SelectNodes("//#style");
if (elementsWithStyleAttribute!=null)
{
foreach (var element in elementsWithStyleAttribute)
{
element.Attributes["style"].Remove();
}
}
}
Your code snippet seems to be correct - it removes the attributes. The thing is, DocumentNode .InnerHtml(I assume you monitored this property) is a complex property, maybe it get updated after some unknown circumstances and you actually shouldn't use this property to get the document as a string. Instead of it HtmlDocument.Save method for this:
string result = null;
using (StringWriter writer = new StringWriter())
{
htmlDoc.Save(writer);
result = writer.ToString();
}
now result variable holds the string representation of your document.
One more thing: your code may be improved by changing your expression to "//*[#style]" which gets you only elements with style attribute.
Here is a very simple solution
VB.net
element.Attributes.Remove(element.Attributes("style"))
c#
element.Attributes.Remove(element.Attributes["style"])