Skip to main content


Eclipse Community Forums
Forum Search:

Search      Help    Register    Login    Home
Home » Archived » Test and Performance Tools Platform (TPTP) » how do you use this stuff?
how do you use this stuff? [message #45884] Fri, 09 December 2005 23:50 Go to next message
Daniel Miles is currently offline Daniel MilesFriend
Messages: 11
Registered: July 2009
Junior Member
I've asked several questions in here before, but reading over some of
the other posts, it's becoming clear to me that I don't have a firm
grasp on what TPTP is. In casually browsing this newsgroup, haven't seen
a single answer that referred to documentation I hadn't already read (so
please don't just send me to the "getting started with tptp" document)
but I still don't understand. Perhaps if I laid out some of my
assumptions, somebody could tell me where I'm making my wrong-turn?

I'm trying to write a general case remote test runner that deploys some
test code on a remote machine, runs it and reports the results. The way
I envision it, each test is a stand alone executable. A TPTP client
deploys it across the data pipe to a TPTP agent and the agent launches
it, watches it and reports its results to the client.

For instance, if I was trying to verify that the correct kernel was
installed on the system under test, I might write a shell script like this:

#!/bin/bash
if [ "`uname -r`" == "2.6.9-11.ELsmp" ]; then
echo "pass"
else
echo "fail"
fi

I want to deploy that shell script, run it and view the results, ideally
in one of the cool common base event viewers that ship with TPTP.

To get this done, I believe I need to write an agent which would go with
the agent controller to live on the system under test and a client which
would plug into eclipse.

Is this the TPTP way, or have I missed something crucial?
Is there some existing functionality in TPTP that I could use to help me?
Re: how do you use this stuff? [message #46763 is a reply to message #45884] Thu, 15 December 2005 14:46 Go to previous messageGo to next message
Paul Slauenwhite is currently offline Paul SlauenwhiteFriend
Messages: 975
Registered: July 2009
Senior Member
Hi Daniel,
Firstly, the Common Base Event viewer (Log View in the Profiling and
Logging Perspective) is used for representing log data from the TPTP Log
model. The TPTP Log model is based on the Common Base Event v1.0.1
specification. Users typically populate the TPTP Log model by:

-Importing a log file (see File >> Import... >> Log File)
-Attaching and monitoring a Logging Agent that is logging Common Base Events
(see File >> New >> Example... >> Examples >> Logging >> JSR-047 Logging
Sample). Users may also write Common Base Event XML fragments directly to a
Logging Agent (see http://www.eclipse.org/tptp/ >> Documentation >>
EclipseWorld 2005 >> Using and Extending the Test and Performance Tools
Platform [Download: source]).

However, I believe you are looking to populate the TPTP Test model
with the execution results from your tests. Based on your problem
description, I would recommend using the existing TPTP JUnit support (see
Help >> Help Contents >> TPTP Tester Guide >> JUnit testing with TPTP) for
writing your tests. The TPTP JUnit test will launch the external script
(e.g. java.lang.Runtime#exec(*)) deployed to the target system when the test
is executed. Alternatively, you may want to write your own test runner (see
http://www.eclipse.org/tptp/ >> Documentation >> EclipseCon 2005 >> Building
a Custom Test Execution Environment).

Paul
"Daniel Miles" <daniel.t.miles@hp.com> wrote in message
news:dnd57d$ap0$1@news.eclipse.org...
> I've asked several questions in here before, but reading over some of
> the other posts, it's becoming clear to me that I don't have a firm
> grasp on what TPTP is. In casually browsing this newsgroup, haven't seen
> a single answer that referred to documentation I hadn't already read (so
> please don't just send me to the "getting started with tptp" document)
> but I still don't understand. Perhaps if I laid out some of my
> assumptions, somebody could tell me where I'm making my wrong-turn?
>
> I'm trying to write a general case remote test runner that deploys some
> test code on a remote machine, runs it and reports the results. The way
> I envision it, each test is a stand alone executable. A TPTP client
> deploys it across the data pipe to a TPTP agent and the agent launches
> it, watches it and reports its results to the client.
>
> For instance, if I was trying to verify that the correct kernel was
> installed on the system under test, I might write a shell script like
this:
>
> #!/bin/bash
> if [ "`uname -r`" == "2.6.9-11.ELsmp" ]; then
> echo "pass"
> else
> echo "fail"
> fi
>
> I want to deploy that shell script, run it and view the results, ideally
> in one of the cool common base event viewers that ship with TPTP.
>
> To get this done, I believe I need to write an agent which would go with
> the agent controller to live on the system under test and a client which
> would plug into eclipse.
>
> Is this the TPTP way, or have I missed something crucial?
> Is there some existing functionality in TPTP that I could use to help me?
Re: how do you use this stuff? [message #46789 is a reply to message #46763] Thu, 15 December 2005 14:49 Go to previous message
Paul Slauenwhite is currently offline Paul SlauenwhiteFriend
Messages: 975
Registered: July 2009
Senior Member
Hi Daniel,
See the
org.eclipse.hyades.logging.core.junit.utilities.TestingUtili ties#executeProc
ess(String[]) method for launching in the external script:

http://dev.eclipse.org/viewcvs/index.cgi/*checkout*/test-res ults/platform/org.eclipse.hyades.logging.core.tests/src/org/ eclipse/hyades/logging/core/junit/utilities/TestingUtilities .java?rev=HEAD&cvsroot=TPTP_Project&content-type=tex t/plain

Paul
"Paul Slauenwhite" <paules@ca.ibm.com> wrote in message
news:dnrvjs$25v$1@news.eclipse.org...
> Hi Daniel,
> Firstly, the Common Base Event viewer (Log View in the Profiling and
> Logging Perspective) is used for representing log data from the TPTP Log
> model. The TPTP Log model is based on the Common Base Event v1.0.1
> specification. Users typically populate the TPTP Log model by:
>
> -Importing a log file (see File >> Import... >> Log File)
> -Attaching and monitoring a Logging Agent that is logging Common Base
Events
> (see File >> New >> Example... >> Examples >> Logging >> JSR-047 Logging
> Sample). Users may also write Common Base Event XML fragments directly to
a
> Logging Agent (see http://www.eclipse.org/tptp/ >> Documentation >>
> EclipseWorld 2005 >> Using and Extending the Test and Performance Tools
> Platform [Download: source]).
>
> However, I believe you are looking to populate the TPTP Test model
> with the execution results from your tests. Based on your problem
> description, I would recommend using the existing TPTP JUnit support (see
> Help >> Help Contents >> TPTP Tester Guide >> JUnit testing with TPTP) for
> writing your tests. The TPTP JUnit test will launch the external script
> (e.g. java.lang.Runtime#exec(*)) deployed to the target system when the
test
> is executed. Alternatively, you may want to write your own test runner
(see
> http://www.eclipse.org/tptp/ >> Documentation >> EclipseCon 2005 >>
Building
> a Custom Test Execution Environment).
>
> Paul
> "Daniel Miles" <daniel.t.miles@hp.com> wrote in message
> news:dnd57d$ap0$1@news.eclipse.org...
> > I've asked several questions in here before, but reading over some of
> > the other posts, it's becoming clear to me that I don't have a firm
> > grasp on what TPTP is. In casually browsing this newsgroup, haven't seen
> > a single answer that referred to documentation I hadn't already read (so
> > please don't just send me to the "getting started with tptp" document)
> > but I still don't understand. Perhaps if I laid out some of my
> > assumptions, somebody could tell me where I'm making my wrong-turn?
> >
> > I'm trying to write a general case remote test runner that deploys some
> > test code on a remote machine, runs it and reports the results. The way
> > I envision it, each test is a stand alone executable. A TPTP client
> > deploys it across the data pipe to a TPTP agent and the agent launches
> > it, watches it and reports its results to the client.
> >
> > For instance, if I was trying to verify that the correct kernel was
> > installed on the system under test, I might write a shell script like
> this:
> >
> > #!/bin/bash
> > if [ "`uname -r`" == "2.6.9-11.ELsmp" ]; then
> > echo "pass"
> > else
> > echo "fail"
> > fi
> >
> > I want to deploy that shell script, run it and view the results, ideally
> > in one of the cool common base event viewers that ship with TPTP.
> >
> > To get this done, I believe I need to write an agent which would go with
> > the agent controller to live on the system under test and a client which
> > would plug into eclipse.
> >
> > Is this the TPTP way, or have I missed something crucial?
> > Is there some existing functionality in TPTP that I could use to help
me?
>
>
Previous Topic:New Technology Agent Controller - Now Available with TPTP 4.1 Release
Next Topic:Using command line parameters with IProces and new Agent Controller
Goto Forum:
  


Current Time: Fri Mar 29 09:59:46 GMT 2024

Powered by FUDForum. Page generated in 0.02348 seconds
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.2.
Copyright ©2001-2010 FUDforum Bulletin Board Software

Back to the top