when S-CORE tests pass in S-CORE, but fail in S-RC

classic Classic list List threaded Threaded
2 messages Options
Reply | Threaded
Open this post in threaded view
|

when S-CORE tests pass in S-CORE, but fail in S-RC

Mike Williams-7
So, the S-RC build translates the S-CORE tests to RemoteControl
commands, and execute them that way.

This just bit me.  One of my checkins last night broke the S-RC build,
as follows:

    * I added a test that used "verifyBodyText"
    * test passed in S-CORE ("verifyBodyText" command is auto-generated
      from "getBodyText")
    * same test, translated, fails in S-RC, since "verifyBodyText" is
      not a valid command there (for reasons I don't understand)

Thanks to Jeff for fixing it up:
http://svn.openqa.org/fisheye/changelog/selenium/?cs=1284

How can we make this less brittle?  Seems to me that, if S-RC is
auto-translating the tests, it should be auto-generating the S-RC
command-set, too.

--
cheers, MikeW                            http://www.dogbiscuit.org/mdub/


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

Reply | Threaded
Open this post in threaded view
|

RE: when S-CORE tests pass in S-CORE, but fail in S-RC

Nelson Sproul
I think the long-term (or perhaps medium-term?) plan is to stop using
the selenium server test translator and adopt Shinya's translator
instead, which is written in a way that pays more attention to the
selenium core JavaScript and reflects that greater knowledge in the
decisions it makes, making it less vulnerable to this category of
failure.


-----Original Message-----
From: Mike Williams [mailto:[hidden email]]
Sent: Tuesday, August 15, 2006 2:58 AM
To: [hidden email]
Subject: [selenium-dev] when S-CORE tests pass in S-CORE, but fail in
S-RC

So, the S-RC build translates the S-CORE tests to RemoteControl
commands, and execute them that way.

This just bit me.  One of my checkins last night broke the S-RC build,
as follows:

    * I added a test that used "verifyBodyText"
    * test passed in S-CORE ("verifyBodyText" command is auto-generated
      from "getBodyText")
    * same test, translated, fails in S-RC, since "verifyBodyText" is
      not a valid command there (for reasons I don't understand)

Thanks to Jeff for fixing it up:
http://svn.openqa.org/fisheye/changelog/selenium/?cs=1284

How can we make this less brittle?  Seems to me that, if S-RC is
auto-translating the tests, it should be auto-generating the S-RC
command-set, too.

--
cheers, MikeW                            http://www.dogbiscuit.org/mdub/


---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]

_______________________________________________________________________
Notice:  This email message, together with any attachments, may contain
information  of  BEA Systems,  Inc.,  its subsidiaries  and  affiliated
entities,  that may be confidential,  proprietary,  copyrighted  and/or
legally privileged, and is intended solely for the use of the individual
or entity named in this message. If you are not the intended recipient,
and have received this message in error, please immediately return this
by email and then delete it.

---------------------------------------------------------------------
To unsubscribe, e-mail: [hidden email]
For additional commands, e-mail: [hidden email]