Xpath builder in Python

Question

I'm building relatively complicated xpath expressions in Python, in order to pass them to selenium. However, its pretty easy to make a mistake, so I'm looking for a library that allows me to build the expressions without messing about with strings. For example, instead of writing

locator='//ul[@class="comment-contents"][contains(., "West")]/li[contains(., "reply")]

I could write something like:

import xpathbuilder as xpb
locator = xpb.root("ul")
             .filter(attr="class",value="comment-contents")
             .filter(xpb.contains(".", "West")
             .subclause("li")
             .filter(xpb.contains (".", "reply"))

which is maybe not as readable, but is less error-prone. Does anything like this exist?

Answer 1

though this is not exactly what you want.. you can use css selector

...
import lxml.cssselect
csssel = 'div[class="main"]'
selobj = lxml.cssselect.CSSSelector(csssel)
elements = selobj(documenttree)

generated XPath expression is in selobj.path

>>> selobj.path
u"descendant-or-self::div[@class = 'main']"