list[0] mangles XPath results

[sjehuda]

Consider using method str.join(iterable).

descriptions = node.xpath('.//p/descendant-or-self::text()')
description = '' if len(descriptions) == 0 else '<br>'.join(descriptions)

Here is where the problem currently relies.

descriptions = node.xpath('.//p/descendant-or-self::text()')
description = '' if len(descriptions) == 0 else descriptions[0]

Our example target is this <p>We are currently looking at a node with a<b> child node </b>within it.</p>, and we would use .//p/descendant-or-self::text() to catch anything between <p> and </p>;

Once we would print list, we would see
[u'We are currently looking at a node with a ', u'child node', u' within it.', u'.'];

This means that list[0] contains We are currently looking at a node with a, list[1] contains child node, and list[2] contains within it.;

list	content
list[0]	We are currently looking at a node with a
list[1]	child node
list[2]	within it.

By obtaining only list[0], we would corrupt the desired XPath result by outputing only the content between the first <p> and the first <b>.

This also significantly hardens the use of XPath String Functions such as substring-before, substring-after and substring, and may require us to make an excessive use of for loops, considering the functions that XPath already provides.