We Keep Coding

Find element by CSS Selector behaviour difference

This is the xml snippet [div id=":6r" class="Ar Au" style="display: block;"]
To locate an element by css selector, i tried by giving the div element followed by the class name with dot(.), but it did not work
driver.find_element_by_css_selector("div.Ar.Au").send_keys("ABC")
However the below worked,
driver.find_element_by_css_selector(".Ar.Au div).send_keys("ABC")
Can someone please tell me why the 1st one did not work ?
Below is the error log using the 1st option:
selenium.common.exceptions.ElementNotInteractableException: Message:
element not interactable

div.Ar.Au mean locate the <div> which has these class name (ArandAu) as an attribute. So in this case its gonna locate <div id=":6r" class="Ar Au" style="display: block;">
.Ar.Au div this mean locate the child <div> tag under a tag which has class name (Arand Au). So this one is locating the child <div> under <div id=":6r" class="Ar Au" style="display: block;">
Note: dot(.) is used in CSS selector to represent a class name and select compound classes

How to find previous WebElements relative to a particular WebElement with selenium python

I have the below HTML snippet.
<div class="header">Planets</div>
<div class="event">Jupiter</div>
<div class="event">Mars</div>
<div class="header">Stars</div>
<div class="event">Acturus</div>
<div class="event">Pleaides</div>
Using driver.find_elements_by_class_name("event"), I am able to retrieve all the div tags with class "event".
I would want to navigate to the previous sibling and retrieve the div tag with class "header" for each WebElement.

Switch to by find_elements_by_xpath
driver.find_elements_by_xpath("//div[#class='event']/preceding-sibling::div[#class='header']")

Clicking button within span with no ID using Selenium

I'm trying to have click a button in the browser with Selenium and Python.
The button is within the following
<div id="generate">
<i class="fa fa-bolt"></i>
<span>Download Slides</span>
<div class="clear"></div>
</div>
Chrome's dev console tells me the button is within <span> but I have no idea how to reference the button for a .click().

Well, if you just want to click on an element without an id or name, I'd suggest three ways to do it:
use xpath:
driver.find_element_by_xpath('//*[#id="generate"]/span')
use CSS selector:
driver.find_element_by_css_selector('#generate > span')
Just try .find_element_by_tag_name() like:
driver.find_element_by_id('generate').find_elements_by_tag_name('span')[0]
Note that this way first try to get the generate <div> element by it's id, and then finds all the <span> elements under that <div>.
Finally, gets the first <span> element use [0].

Python scrapy, how to only get immediate children

so i have some html like this
<div class="content">
<div class="infobox">
<p> text </p>
<p> more text </p>
</div>
<p> text again </p>
<p> even more text </p>
</div>
And i am using this selector '.content p::text' i thought this would only get me the immediate children, so i wanted it to extract "text again" and "even more text" but it's also getting the text from the paragraphs inside the other div, how can i prevent this from happening, i only want text from the paragraphs that are the immediate children of the div with the class .content

Scrapy uses an extended set of CSS selectors and XPath selectors. In your case, you're using CSS selectors. The CSS relationship selector you want is > denoting a parent/child relationship, as in: .content > p::text. Scrapy's selectors are described in the section titled "Selectors" in its documentation.

to get the child: div>p ( text, more text )
In your case to get what you need: div+p
http://www.w3schools.com/cssref/css_selectors.asp
Worth reading

Python Selenium Webdriver - Grab div after specified one

I am trying to use Python Selenium Firefox Webdriver to grab the h2 content 'My Data Title' from this HTML
<div class="box">
<ul class="navigation">
<li class="live">
<span>
Section Details
</span>
</li>
</ul>
</div>
<div class="box">
<h2>
My Data Title
</h2>
</div>
<div class="box">
<ul class="navigation">
<li class="live">
<span>
Another Section
</span>
</li>
</ul>
</div>
<div class="box">
<h2>
Another Title
</h2>
</div>
Each div has a class of box so I can't easily identify the one I want. Is there a way to tell Selenium to grab the h2 in the box class that comes after the one that has the span called 'Section Details'?

If you want grab the h2 in the box class that comes after the one that has the span with text Section Details try below xpath using preceding :-
(//h2[preceding::span[normalize-space(text()) = 'Section Details']])[1]
or using following :
(//span[normalize-space(text()) = 'Section Details']/following::h2)[1]
and for Another Section just change the span text in xpath as:-
(//h2[preceding::span[normalize-space(text()) = 'Another Section']])[1]
or
(//span[normalize-space(text()) = 'Another Section']/following::h2)[1]

Here is an XPath to select the title following the text "Section Details":
//div[#class='box'][normalize-space(.)='Section Details']/following::h2

yeah, you need to do some complicated xpath searching:
referenceElementList = driver.find_elements_by_xpath("//span")
for eachElement in referenceElementList:
if eachElement.get_attribute("innerHTML") == 'Section Details':
elementYouWant = eachElement.find_element_by_xpath("../../../following-sibling::div/h2")
elementYouWant.get_attribute("innerHTML") should give you "My Data Title"
My code reads:
find all span elements regardless of where they are in HTML and store them in a list called referenceElementList;
iterate all span elements in referenceElementList one by one, looking for a span whose innerHTML attribute is 'Section Details'.
if there is a match, we have found the span, and we navigate backwards three levels to locate the enclosing div[#class='box'], and find this div element next sibling, which is the second div element,
Lastly, we locate the h2 element from its parent.
Can you please tell me if my code works? I might have gone wrong somewhere navigating backwards.
There is potential difficulty you may encounter, the innerHTML attribute may contain tab, new line and space characters, in that case, you need regex to do some filtering first.