I'm new to XPath so am sure my understanding is the issue here and could
use some help....

I've got some fairly messy (X)HTML which has TD elements sprinkled in
solely for the purposes of formatting (i.e. <td height="2"></td>) and
was thinking that I could use XPath to select only the TD nodes which
actually display a value. I've verified the basic path to get all
elements in the table (./html/body/table/tr), but when tried to extend
this with various permutations such as
../html/body/table/tr/td[string-length(text()) 0] without success.

Thanks in advance,
Phil

In article <78udnRhMcdenqlvZnZ2dnUVZ_vednZ2d@comcast.com>,
Foxpointe <foxpointe@comcast.netwrote:

>./html/body/table/tr/td[string-length(text()) 0] without success.

That should work, but you could just use

./html/body/table/tr/td[text()]

since empty ones won't have a text node at all.

-- Richard

Richard Tobin wrote:

That should work, but you could just use
>
./html/body/table/tr/td[text()]
>
since empty ones won't have a text node at all.
>
-- Richard

It turns out it did work (user error ;-) but thanks for the tip as I
didn't realize I didn't need to explicitly check the length.

Thanks,
Phil