Selenium with Python: how to click preceding element

Question

I'm working with a calendar and the only element I can access (due to non-askii issues) is a button that says 'Today' at the top of it. Directly to the left is a left arrow that takes you back a month; similarly, directly to the right is a right arrow that takes you forward a month. The html looks like this:

<tr style="-moz-user-select: none;">
<td class="button nav" style="-moz-user-select: none;" colspan="1">
    «
</td>
<td class="button nav" style="-moz-user-select: none;" colspan="1">
    ‹
</td>
<td class="button nav" style="-moz-user-select: none;" colspan="3">
    Today
</td>
<td class="button nav" style="-moz-user-select: none;" colspan="1">
    ›
</td>
<td class="button nav" style="-moz-user-select: none;" colspan="1">
    »
</td>
</tr>

I can click the 'Today' button by using:

driver.find_element_by_xpath("//td[text()='Today']").click()

Also, I can click the right arrow using 'following':

driver.find_element_by_xpath("//td[text()='Today']/following::td").click()

However, I can not figure out how to click the left arrow (I believe with 'preceding'). This is what I'm trying and.. it says it is unable to find element with xpath == //...

driver.find_element_by_xpath("//td[text()='Today']/preceding::td").click()

Is there another command I'm supposed to be using in order to find the previous element/sibling?

If it matters, I'm using IE9 and Selenium 2.44.

Thanks.

Answer 1

Try the below xpaths:

1- For clicking on the "<" arrow :

//td[contains(text(),'Today')]/preceding-sibling::td[1]

Above xpath locates the first preceding "td" element to the "td" element that has innerHTML/text as "Today" .

2- For clicking on the "<<" arrow :

//td[contains(text(),'Today')]/preceding-sibling::td[2]

Above xpath locates the second preceding "td" element to the "td" element that has innerHTML/text as "Today" .

Answer 2

While it does not use 'preceding', nor look at a sibling, I figured out how to do it like so:

precedes = self.driver.find_elements_by_xpath("//td[text()='Today']/../td")
precedes[1].click()

It returns to the parent of the td containing 'today', and then finds the child td's and chooses the second one (index 1)