You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@tapestry.apache.org by Taha Hafeez Siddiqi <ta...@gmail.com> on 2013/08/01 02:13:07 UTC
Re: Discussion: Future of tapestry-test & friends.
+1 for Spock and Geb.
I have been using Spock and Geb for some time now. Being both in Groovy saves you a lot of time.
One of the concerns that I have with Geb is unstable tests which seems to be related to WebDriver but, for some reason, are more visible with Geb.
BTW Spock extensions are really easy and fun to write!.
regarding Ulrich's point. I agree that Tapestry stack keeps changing and it is not easy to cope up with it. Actually, that is a great thing if you have time to learn. You get free examples in the clearest of codes from Howard!! On the flip side, if you don't have time, you lag behind and can't contribute much to the project. That is one reason my blog has dried out as I couldn't get time to experiment with Tapestry 5.4.
Thankfully, we have started migrating our libraries to Tapestry 5.4. I am now looking into the new source and it looks like an exciting next few weeks.
regards
Taha
On 31-Jul-2013, at 7:54 PM, Massimo Lusetti <ml...@gmail.com> wrote:
> On Wed, Jul 31, 2013 at 4:10 PM, Ulrich Stärk <ul...@spielviel.de> wrote:
>
> One reason I haven't contributed much in terms of code for quite some time
>> is the ever changing
>> technology stack Tapestry is built with. We have an increasingly complex
>> stack of bleeding-edge
>> tools and technologies that I simply lack the time of keeping up with.
>>
>> I have the feeling that this might be a turn-down for other potential
>> contributors as well. I won't
>> be against it but don't be surprised about continously declining
>> contributor activity.
>>
>>
>>
> One the other side I've always taken the "stack of bleeding-edge tools and
> technologies" used inside Tapestry as a reason to learn cool new stuff and
> ideas.
>
> BTW Spock and Geb doesn't seem so new in the market, well newer then TestNG
> and EasyMock yes.
>
> I can buy your point of view from a "enterprise business" point of view but
> I think we talking about a slow scrap with a deprecation cycle instead of
> simply throw away in the trash.
>
> --
> Massimo Lusetti
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
For additional commands, e-mail: dev-help@tapestry.apache.org
Re: Discussion: Future of tapestry-test & friends.
Posted by Barry Books <tr...@gmail.com>.
I have a few thoughts but they are conflicting.
1. Anything that makes testing easier would be welcome.
2. It would have to be beyond easy to create new test cases to make me want
to rewrite my old ones.
3. I'm sure (fill in you favorite language here) is great but I'm
mandated (It's more of a suggestion really) to use Java.
4. If I have to find Groovy programers they are going to know Grails.
5. I have enough trouble with Selenium running under Hudson on my build
machine. I don't really want to try and support something else also.
I can see this being used on new projects but I don't see much benefit to
existing ones.
On Thu, Aug 1, 2013 at 4:04 AM, Denis Delangle
<de...@gmail.com>wrote:
> Hello,
>
> As a Tapestry user, I use tapestry-test to test my applications and my
> widgets libraries to validate migration from tapestry versions. I did
> it from tapestry 5.1 to 5.4 quite easily. I would be happy to
> integrate new technologies for the tests I will write in the future
> but I see absolutely no good reason to rewrite existing tests. Writing
> regression tests is an investment on the long term, having to rewrite
> them because test technologies are no more supported would be a great
> loss.
> So please don't remove existing selenium support so we can keep our
> test harness when migrating tapestry to 5.5, 5.6 and more .
>
> I added to SeleniumTestCase a way to launch a forever waiting test so
> I can play with the screens and correct them. In parallel I can run
> testng tests from eclipse that will reuse the jetty instance. It make
> me save a lot of time while finding the correct selenium xpath
> expressions. You can find the code here.
> https://gist.github.com/ddelangle/6129694
>
> Do you know casperjs ? It provides a javascript DSL to run tests on
> headless webkit (phantomjs) and firefox, that makes it much faster
> that with selenium and can be run on simple servers without Xserver.
> http://casperjs.org/
>
>
> Regards,
>
> Denis
>
> 2013/8/1 Taha Hafeez Siddiqi <ta...@gmail.com>:
> > +1 for Spock and Geb.
> >
> > I have been using Spock and Geb for some time now. Being both in Groovy
> saves you a lot of time.
> >
> > One of the concerns that I have with Geb is unstable tests which seems
> to be related to WebDriver but, for some reason, are more visible with Geb.
> >
> > BTW Spock extensions are really easy and fun to write!.
> >
> >
> > regarding Ulrich's point. I agree that Tapestry stack keeps changing and
> it is not easy to cope up with it. Actually, that is a great thing if you
> have time to learn. You get free examples in the clearest of codes from
> Howard!! On the flip side, if you don't have time, you lag behind and can't
> contribute much to the project. That is one reason my blog has dried out as
> I couldn't get time to experiment with Tapestry 5.4.
> >
> > Thankfully, we have started migrating our libraries to Tapestry 5.4. I
> am now looking into the new source and it looks like an exciting next few
> weeks.
> >
> > regards
> > Taha
> >
> > On 31-Jul-2013, at 7:54 PM, Massimo Lusetti <ml...@gmail.com> wrote:
> >
> >> On Wed, Jul 31, 2013 at 4:10 PM, Ulrich Stärk <ul...@spielviel.de> wrote:
> >>
> >> One reason I haven't contributed much in terms of code for quite some
> time
> >>> is the ever changing
> >>> technology stack Tapestry is built with. We have an increasingly
> complex
> >>> stack of bleeding-edge
> >>> tools and technologies that I simply lack the time of keeping up with.
> >>>
> >>> I have the feeling that this might be a turn-down for other potential
> >>> contributors as well. I won't
> >>> be against it but don't be surprised about continously declining
> >>> contributor activity.
> >>>
> >>>
> >>>
> >> One the other side I've always taken the "stack of bleeding-edge tools
> and
> >> technologies" used inside Tapestry as a reason to learn cool new stuff
> and
> >> ideas.
> >>
> >> BTW Spock and Geb doesn't seem so new in the market, well newer then
> TestNG
> >> and EasyMock yes.
> >>
> >> I can buy your point of view from a "enterprise business" point of view
> but
> >> I think we talking about a slow scrap with a deprecation cycle instead
> of
> >> simply throw away in the trash.
> >>
> >> --
> >> Massimo Lusetti
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
> > For additional commands, e-mail: dev-help@tapestry.apache.org
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
> For additional commands, e-mail: dev-help@tapestry.apache.org
>
>
Re: Discussion: Future of tapestry-test & friends.
Posted by Denis Delangle <de...@gmail.com>.
Hello,
As a Tapestry user, I use tapestry-test to test my applications and my
widgets libraries to validate migration from tapestry versions. I did
it from tapestry 5.1 to 5.4 quite easily. I would be happy to
integrate new technologies for the tests I will write in the future
but I see absolutely no good reason to rewrite existing tests. Writing
regression tests is an investment on the long term, having to rewrite
them because test technologies are no more supported would be a great
loss.
So please don't remove existing selenium support so we can keep our
test harness when migrating tapestry to 5.5, 5.6 and more .
I added to SeleniumTestCase a way to launch a forever waiting test so
I can play with the screens and correct them. In parallel I can run
testng tests from eclipse that will reuse the jetty instance. It make
me save a lot of time while finding the correct selenium xpath
expressions. You can find the code here.
https://gist.github.com/ddelangle/6129694
Do you know casperjs ? It provides a javascript DSL to run tests on
headless webkit (phantomjs) and firefox, that makes it much faster
that with selenium and can be run on simple servers without Xserver.
http://casperjs.org/
Regards,
Denis
2013/8/1 Taha Hafeez Siddiqi <ta...@gmail.com>:
> +1 for Spock and Geb.
>
> I have been using Spock and Geb for some time now. Being both in Groovy saves you a lot of time.
>
> One of the concerns that I have with Geb is unstable tests which seems to be related to WebDriver but, for some reason, are more visible with Geb.
>
> BTW Spock extensions are really easy and fun to write!.
>
>
> regarding Ulrich's point. I agree that Tapestry stack keeps changing and it is not easy to cope up with it. Actually, that is a great thing if you have time to learn. You get free examples in the clearest of codes from Howard!! On the flip side, if you don't have time, you lag behind and can't contribute much to the project. That is one reason my blog has dried out as I couldn't get time to experiment with Tapestry 5.4.
>
> Thankfully, we have started migrating our libraries to Tapestry 5.4. I am now looking into the new source and it looks like an exciting next few weeks.
>
> regards
> Taha
>
> On 31-Jul-2013, at 7:54 PM, Massimo Lusetti <ml...@gmail.com> wrote:
>
>> On Wed, Jul 31, 2013 at 4:10 PM, Ulrich Stärk <ul...@spielviel.de> wrote:
>>
>> One reason I haven't contributed much in terms of code for quite some time
>>> is the ever changing
>>> technology stack Tapestry is built with. We have an increasingly complex
>>> stack of bleeding-edge
>>> tools and technologies that I simply lack the time of keeping up with.
>>>
>>> I have the feeling that this might be a turn-down for other potential
>>> contributors as well. I won't
>>> be against it but don't be surprised about continously declining
>>> contributor activity.
>>>
>>>
>>>
>> One the other side I've always taken the "stack of bleeding-edge tools and
>> technologies" used inside Tapestry as a reason to learn cool new stuff and
>> ideas.
>>
>> BTW Spock and Geb doesn't seem so new in the market, well newer then TestNG
>> and EasyMock yes.
>>
>> I can buy your point of view from a "enterprise business" point of view but
>> I think we talking about a slow scrap with a deprecation cycle instead of
>> simply throw away in the trash.
>>
>> --
>> Massimo Lusetti
>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
> For additional commands, e-mail: dev-help@tapestry.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@tapestry.apache.org
For additional commands, e-mail: dev-help@tapestry.apache.org