Trying to understand a CSS chain - forms

can anybody explain the following code snippet (used in CSS):
[type="checkbox"]:checked
I've tried to find this on various sites, I understand the pseudo class on the end but the square brackets have really got me stumped.
Thanks for reading.

this [type="checkbox"] is an attribute selector
[attr=value]
Represents an element with an attribute name of attr and whose value is exactly "value".
:checked is:
a pseudo class selector represents any radio (<input type="radio">),
checkbox (<input type="checkbox">) or option (<option> in a <select>)
element that is checked or toggled to an on state. The user can change
this state by clicking on the element, or selecting a different value,
in which case the :checked pseudo-class no longer applies to this
element, but will to the relevant one.
Which means you have a checkbox element checked

[type="checkbox"] is an attribute selector.
This specific selector will match any element that has the attribute type and that attribute's value is checkbox. Most would identify this as a selector for and input but is not specific enough to be limited to that element type. Other elements that accept the type attribute are <button> <command>, <embed>, <object>, <script>, <source>, <style> and <menu>.
You'll often see input pre-pended to a selector like the one above, i.e. input[type="checkbox"], when targeting specific types of input.

The square brackets target an attribute such as type for an input elemet.
In your case you're selecting a checked checkbox.

Related

Accessibility ARIA issues with react-select

We are using react-select component in one of our projects and trying to resolve accessibility issues. Right now, there is no way to provide the required aria attributes to the input element which has the aria-autocomplete="list" attribute.
https://www.digitala11y.com/aria-autocomplete-properties/
Above is the reference link to what we are trying to achieve. Below is the short story from the above link.
If an element has aria-autocomplete set to list or both, authors MUST ensure both of the following conditions are met:
The element has a value specified for aria-controls that refers to the element that contains the collection of suggested values.
Either the element or a containing element with role combobox has a value for aria-haspopup that matches the role of the element that contains the collection of suggested values.
I opened a PR to add the ability to add ARIA properties to the input element. https://github.com/JedWatson/react-select/pull/4132
But I wanted to check if I'm missing something if anyone has already worked on this. Please let me know if you need any other details.

How can I locate an element by header or any particular property value in Protractor?

I have an element that have no Id defined. Class name do not allow me to identify the element uniquely.
That's why I would like to locate it using role or header property:
<p-dialog header="Deleting" showeffect="fade" role="dialog"
class="ng-tns"> </p-dialog>
How should I do that in Protractor / Selenium Webdriver?
I can do that by means of xpath, but as I understand it's rather suggested to avoid it at all.
Do I need to define my own locator using addLocator or is there simpler solution?
Protractor recommend to use css selector as first option, xpath as the secondary option when css selector can resolve your problem in some cases.
The primary reason of such recommendation considered followings:
xpath is not fast than css selector
xpath is case sensitive in some browser
for lower IE, like 6 not support xpath natively
xpath generally is longer than css selector to find same element
But css selector has two shortage which xpath is able to do:
can't find element by text
can't find parent element (only can find downward)
For your case, you can try following locator:
// css selector
p-dialog[header="Deleting"]
p-dialog[role="dialog"]
p-dialog[header="Deleting"][role="dialog"]

Find CURRENTLY selected <option> with XPath

What's the correct XPath syntax to check if an option element is currently selected, or just to get the selected option element from a select element, on an open page with which the user, and JavaScript, may have interacted? Is this even possible with XPath, or does it lack the ability to look at DOM properties?
I can't find any documentation on this, and have (speculatively) tried:
//option[#selected=true]
//option[#selected="selected"]
//option[#selected]
but none of these work; they simply don't match any elements.
(In case it matters, I've tried this both using the $x function in the Chrome developer console, and using the find_elements_by_xpath method in Selenium for Python.)
Short answer: it's not possible.
Longer answer: XPath can look at HTML attributes, but it can't look at DOM properties. Selecting an <option> element in a <select> changes the selected property of the <option> to true, and also changes the value property of its parent <select> element, but it doesn't affect the attributes of either, so it is invisible to XPath.
To find <option> elements that have the selected attribute set, which is often how a page author might determine which option is initially selected, you can use //option[#selected]. But this does not find the currently selected <option>; changes that the user makes to the selection are invisible to XPath. There's no guarantee it will even find the initially selected option, since it's possible that the page author didn't put the selected attribute on any elements and either let the browser select the first option by default or had some JavaScript select the initial option via the selected property.
The multiple other answers here claiming that a selector like //option[#selected] can detect selection changes made by the user after the page loads are simply completely wrong.
Of course, if you're able to use CSS selectors instead of XPath selectors, then option:checked will do the job.
The problem could be the " (double quotes).
//select/option[#selected='selected'] - Will match the selected option, i am using this successfully.
//select/option[#selected='selected' and #value='specific value'] - Will only match the selected option if it has a 'specific value', i'm also using this.
If you are still having trouble, it could be an entirely different problem, perhaps there is no option node. I hope this helps.
I think we can use a knowledge from #Mark's answer and account that. Let's just find a node which HAS desired attribute:
tree.xpath('//select/option[#selected]/text()')[0].strip()
I tried "//option[#selected=''] and it has worked for me.
it is able to highlight the selected option within Page objects model.
I would try //option[#selected='true']
i.e. driver.findElements(By.xpath("//option[#selected='true']")).getText();

Is it allowed to use <label> tag without labeled control?

I need to show in a page a list of, let's say, person's properties that should be rendered more or less as follow:
name: Name
date: 1/1/2000
other: Other
Reading the doc they say:
The LABEL element may be used to attach information to controls.
So, is it the right tag to encompass the names of the properties like name, date...
even if there's not an <input> to associate with?
Nope, as per Quentin’s answer.
However, in HTML5, <dl> can be used for generic associations where <label> isn’t appropriate.
No.
It says that it can associate information with controls.
It does not say that it can associate information with anything else.
See also the newer (but draft) specification:
Some elements, not all of them form-associated, are categorized as
labelable elements. These are elements that can be associated with a
label element.
button input (if the type attribute is not in the hidden state) keygen
meter output progress select textarea
No, it is not correct to use the label element like that.
The important thing here is the meaning of may.
The LABEL element may be used to attach information to controls.
RFC 2119 (which the HTML4 spec follows) defines may:
May: This word, or the adjective "OPTIONAL", mean that an item is truly optional
So here, may does not mean the label element can be used for other purposes; it just means that it is not required to use a label element with controls.
As far as alternatives go, it depends what you want to achieve. If you are just trying to follow the spec closely, then I suggest using p and a strong element for each pair. If you want the data to be more meaningful to computers, you could consider using one of the Microformat specifications.
I partially agree with the answers so far but want to add that if you want to use labels for some reason, then I would display the property values in read-only controls, and then your labels will be valid.
I've done this using appropriate styling to differentiate the read-only controls from the functioning controls to avoid confusing your users. This was on a sequence of pages which built up the information gathered from the user in a wizard.
I have this link to W3C - the "Editor's Draft" as opposed to the link above which is the "Working Draft", which states that you can associate it with any element - as long as it's 'labelable' - even though this is a subsection of the Form section. It states earlier that a labelable element does not have to be within a form.
http://dev.w3.org/html5/spec/single-page.html#the-label-element

jQuery: Select all 'select' elements with certain val()

Does anyone know of an easy way, using jQuery, to select all <select> elements whose val() attribute yields a certain value?
I'm trying to do some validation logic and would like to just select all those elements with a single selector, then apply a warning class to each of their parents. This I know how to do once I select all the elements, but I didn't see a selector that handles this case.
Am I going to have to select all of the <select> elements into a selector, then iterate through them and check each of their values? I was hoping there would be a simpler way.
Thanks.
Why doesn't select[value=x] work? Well firstly because <select> doesn't actually have a value attribute. There is not a single value of a select box: there may be no selected options (there shouldn't normally be, but there can be in at least IE), and, in a <select multiple>, there can be any number of selected options.
Even input[value=x] doesn't work, even though <input> does have a value attribute. Well, it does work, it just doesn't do what you think. It fetches the value of the value="..." attribute in the HTML, not the current value you have entered into the form. The value="..." attribute actually corresponds to the defaultValue property and not value.
Similarly, option[value=x][selected] doesn't work because it is checking the <option selected> attribute from the HTML source (selected attribute -> defaultSelected property) and not the current selectedness of the option (selected property not attribute) - which might have changed since the page was loaded.
Except in IE, which gets the value, selected etc form attributes wrong.
Except (again): Tesserex's example may seem to work, and the reason for that is that that it's using a non-standard jQuery-specific selector, :has. This causes the native querySelectorAll methods of modern browsers to fail, and consequently jQuery falls back to its own (native JavaScript, slow) selector engine instead. This selector engine has a bug where it confuses properties for attributes, allowing [value=x] to do what you expected, and not fail like it should! (Update: this is probably no longer the case in newer jQuery versions.)
Summary: form field state checking and selectors don't mix. Apart from these issues, you also have to worry about escaping issues - for example, what if the value you want to test against contains quotes or square brackets?
So instead, yes, you should check it manually. For example using a filter:
$('select').filter(function() {
return $(this).val()==='the target value';
}).parent().addClass('warning');
(There is a value property in HTML5 and supported by modern browsers, that when you read it gives you the value of the first selected <option>. jQuery's val() is safe to use here because it provides the same method of getting the first selected option even on browsers that don't support this.)
The existing answers don't work on select tags, but I found something that does. Ask for a select that has a selected option.
$("select:has(option[value=blah]:selected)")
You can use :
$("select[value=X]");
where X is the value against which you want to check the select's value.
Attribute selectors Is what you're looking for I believe.
Something like $+('element[attribute="value"]')
See also:
*= anywhere
^= starts with
$= ends with
~= contains word
etc.
You can create a change event that puts the value in a custom attribute on the select element whenever the value changes. You can then use a simple selector to find all of the select elements that have that value. For example:
$("select").on("change", function (e) {
var $select = $(e.currentTarget);
$select.attr("select-value", $select.val());
});
And then you can do this:
var $matches = $("select[select-value='" + searchVal + "']");
$matches will have all of your matching selects.
This is a lot easier than having to iterate through elements. Remember to set select-value to the initial value when rendering the page so you don't need to trigger a change event for each select so the select-value is set.