Hi All,

With the new Engine APIs in place, which is essentially uid-based such as

click "myui.link"

instead of the Selenium way

click(locator_of_link)

I wonder if it is a good idea to split the current core into two parts:

1) Selenium-wrapper: Generate runtime locator based on UI definition and
call Selenium APIs. It needs to work with Selenium RC + Selenium server.

2) Tellurium-Core: UI module parse, not runtime locator, but pass uid
directly to Tellurium Engine. May consider merging Selenium RC
into it as a layer. Perhaps a new Tellurium server.

The benefits of doing it is

1) Separate the wrapper from the Tellurium core so that Tellurium core does
not need to be twisted to be compatible with Selenium APIs. Simplify the
design.
2) Separate builds for users to choose which one they like to use. In this
way, Tellurium can really go on its own way without the burden of being
compatible with Selenium.
3) Make it easy to port Tellurium Core to other languages such as Ruby.
4) Perhaps, we can also create our own Engine without Selenium Core code,
easier to maintain, i.e., split the current Engine project as well.

With that, Selenium-wrapper would be a pretty stable artifact so that users
can keep using it together with Selenium server, but Tellurium core and
Engine keep evolving rapidly.

What do you think?

Thanks,

Jian

-- 
You received this message because you are subscribed to the Google Groups 
"tellurium-users" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/tellurium-users?hl=en.

Reply via email to