Finding the position of an element

To find the position of element <b>col_d</b> in the following xml:

1 2 3 4 5 6 7

<a> <b>col_a</b> <b>col_b</b> <b>col_c</b> <b>col_d</b> <b>col_e</b> </a>

U can probably use the following xpath:

count(//b[normalize-space(text())="col_d"]/preceding-sibling::b)+1

But it doesn't work for me in nokogiri, so what i did was:

1 2	doc = Nokogiri::XML(above_xml_string) doc.xpath('//b[normalize-space(text())="col_d"]/preceding-sibling::b').count + 1

That's all !!

Finding Common Elements in Ruby Arrays

To find common elements in ruby arrays, use '|'. Eg.

%w{a b c d} | %w{c d e f}

# yields:

['c', 'd']


To find common elements in 2 dimensional array:

a1, a2, a3 = %w{a b c d}, %w{c d e f}, %w{d e f}
a = [a1, a2, a3]

a[1..-1].inject(a[0]) do |x, y|
x | y
end

# yields:

['d']


That's all !!

Getting the nearest xpath ancestor

Given the following xml fragment:

1 2 3 4 5

<a> <a> <b/> </a> <a>

To find the nearest ancestor <a/> w.r.t <b/>, i use:

//b/ancestor::a[1]

That's all !!

Renaming a screen window

To rename a screen window, just do:

'Ctrl+a' & 'Shift-a'.


Smart click_button

In previous post, i've mentioned abt webrat's smart click_link(), and do u know that click_button() is also quite smart, meaning:

1	<input type="image" alt="Update" />

Works as well if we do:

1	click_button("Update")

Nice ?!

Smart click_link()

WOW, just found out image within <a/> tag (see below):

1 2 3

<a class="image" href="..."> <img src="..." alt="Back"/> </a>

Actually work for the following pair of:

1 2	# Step Given I follow "Back"

&:

1 2 3 4

# Step definition Given /^I follow "([^\"]+)"$/ do |link| click_link(link) # plain vanilla webrat end

Cool !!

Workaround for Webrat's have_xpath()

I have the following xpath:

//td[normalize-space(text())='Peter']/parent::tr//label[normalize-space(text())="Present"]/@for

It is valid, yet webrat's have_xpath() kept complaining that it cannot find it. As a workaround, i did something like this in my cucumber:

1 2	hdoc = Nokogiri::HTML(response.body) (field_id = hdoc.xpath(the_above_rejected_xpath)).first.should_not be_nil

That's all.

Webrat's have_xpath() in diff modes

have_xpath() behaves slightly differently in :selenium & :webrat mode.

From webrat's doc, we have:

Webrat::Matchers#have_xpath(expected, options = {}, &block)

And

Webrat::Selenium::Matchers::have_xpath(xpath)

If u are a xpath fan like me, there is no way u can conveniently switch between the 2 modes without getting bitten.

2 Way to Fetch a Node's Attribute Value

There are at least 2 ways to fetch a node's attribute value using xpath:

# Method 1

//a/b/@c

# Method 2

//a/b/attribute::c


Fixing Gnome Tooltip

I've accidentally screwed up my gnome's config, as a result, all tooltips in my gnome desktop appear as meaningless black blocks. After some playing round & trial & error, i fixed the problem by replacing the following lines in ~/.gconf/desktop/gnome/interface/%gconf.xml:

1 2	tooltip_fg_color:#000000000000 tooltip_bg_color:#000000000000

With the more meaningful:

1 2	tooltip_fg_color:#000000000000 tooltip_bg_color:#ffffffffffff

That's all !!