Thanks for sharing your experiences with Watir.
Any time you have to use sleep to keep a script synchronized you are likely to eventually run into the kinds of reliability problems you are seeing. I consider any situation where you need to use sleeps to reveal a bug in Watir.
The popup code has always been iffy. One of the problems we have had has been there has not been a clear way of distinguishing the stuff that is tested and reliable from other stuff that is more experimental. At one time, it was suggested that the popup code shouldn't be released because people might expect too much of it, but it was also pointed out that it useful and better than nothing.
In the future, we will be moving untested libraries of unknown reliability into a contrib directory. You will use "require 'watir/contrib/libname" to use these libraries.
In any case, i am working on creating a robust popup support library. Some of this code necessary is already checked in. Watir 1.5 will support modal web dialogs as well as have more robust and convenient support for modal windows dialogs as well.
I am always eager to get help. The biggest area of help that i can use is if people are to create more unit tests. For example, if someone created unit tests for popups with hard-coded sleeps, then i could remove the sleeps and use them to test my new popup code support. That would be a massive help.
What kinds of popups are you having trouble with? Do we have tests for them?
Bret
On 2/20/06,
[EMAIL PROTECTED] <[EMAIL PROTECTED]> wrote:
Good afternoon:
We created a testing suite using test environment A. Various strategic
sleeps were put in place to optimize the execution of WATIR scripts against
this test environment. For the most part these test suites execute with
some reliability.
On Friday I attempted to run the same set of scripts on my development
environment. Of course the scripts would break (primarily failed
assertions) because my laptop is not as powerful as the testing machines.
How do y'all handle the issue of pointing your tests to different
environments and not having the scripts execute with reliability?
Should we create scripts in the worst case environment and assume they will
function on the production ready hardware?
I believe the primary problem is the latency associated with Pop ups. Does
anyone use any tricks to make their scripts more reliable?
Despite everyone's hard work on WATIR, I am finding this to be an extreme
limitation and I am once again questioning my choice of WATIR as our
preferred testing solutions.
Will any of the new activities for the next release address this concern?
Am I isolated in having this concern?
Any thoughts or suggestions would be appreciated.
Thanks,
Carl
--
"NOTICE: The information contained in this electronic mail transmission is
intended by Convergys Corporation for the use of the named individual or
entity to which it is directed and may contain information that is
privileged or otherwise confidential. If you have received this electronic
mail transmission in error, please delete it from your system without
copying or forwarding it, and notify the sender of the error by reply email
or by telephone (collect), so that the sender's address records can be
corrected."
_______________________________________________
Wtr-general mailing list
[email protected]
http://rubyforge.org/mailman/listinfo/wtr-general
_______________________________________________ Wtr-general mailing list [email protected] http://rubyforge.org/mailman/listinfo/wtr-general
