Re: [xml-dev] Testing XML don't use xUnit

[James Fuller <james.fuller.2007@gmail.com>]

On Fri, Apr 12, 2013 at 10:16 AM, Ihe Onwuka <ihe.onwuka@gmail.com> wrote:
> On Fri, Apr 12, 2013 at 8:48 AM, Stephen D Green
> <stephengreenubl@gmail.com> wrote:
>> There is significant difference how Selenium API is used
>> compared to Selenium IDE (for Firefox), as far as XPath
>> locators of HTML attributes and elements is concerned.
>> Selenium API requires writing of code to run it which adds
>> time to development of the tests so developer is more likely
>> to be looking for repeated use with lower maintenance so
>> yes XPath expressions might be avoided. With the IDE,
>> on the other hand, the IDE itself provides XPath expressions
>> and tests are quickly produced and can be easily replaced
>> rather than maintained so XPath is very popular, in my
>> experience / awareness. With the API I'd think a developer
>> would be likely to take time to separate between XPaths
>> and more abstract test descriptions (usually by abstracting
>> away from the XPath perhaps). But like I say, there is still
>> a place for XPath in the IDE; and important place.
>>
>> Cheers
>>
>
> Ultimately we would like to translate CSS into markup. Does this mean
> that Selenium API could play a role in helping us achieve that goal.

Gerrit Imsieke (@gimsieke) latest work with CSSa presented at XML
Prague 2013, may provide some brain food for this.

https://github.com/gimsieke/CSSa

http://archive.xmlprague.cz/2013/presentations/Conveying_Layout_Information_with_CSSa/CSSa_xmlprague_gimsieke.html#/step-1

hth, Jim Fuller