Tool suggestion: JTReg test module dependencies verifier.

Alexandre (Shura) Iline alexandre.iline at oracle.com
Thu Mar 29 18:05:53 UTC 2018


Jon, do you think I should start JEP for this tool?

I think now is the right time.

Shura

> On Mar 28, 2018, at 12:05 PM, Jonathan Gibbons <jonathan.gibbons at oracle.com> wrote:
> 
> 
> 
> On 03/28/2018 11:34 AM, Alexandre (Shura) Iline wrote:
>> 
>> 
>>> On Mar 28, 2018, at 11:22 AM, Jonathan Gibbons <jonathan.gibbons at oracle.com <mailto:jonathan.gibbons at oracle.com>> wrote:
>>> 
>>> 
>>> 
>>> On 3/28/18 10:57 AM, Alexandre (Shura) Iline wrote:
>>>>> There's a 3rd (weaker?) category of "implementation desires", such as whether or not or how much this is integrated into jtreg.
>>>> Correct, these are the non-functional requirements, of which at least one is missing right now:
>>>> 2. The tool must use jdeps tool or jdeps API to perform test selection and test execution.
>>> It's quibbling, but I don't think these are requirements. If jdeps or the jdeps API did not exist, you could still implement what you are proposing.  jdeps and the jdeps API are just technologies which you may elect to use to make your task simpler.
>> 
>> Well, it must be a requirement to treat modules in a way compatible with JDK. By requiring what I am requiring, I am guarantying that and also, in addition, since I am strongly opposed to re-implement what is already implemented in jdeps, this is my proposal for the tool: it must use jdeps.
>> 
>> Same, word for word, goes for JTReg, FWIW.
>> 
>> Shura
>>> 
>>> -- Jon
>> 
> 
> I'm not against using jdeps, or jtreg. But there is a difference between a Requirement and an implementation choice. But this is just an email thread, and not any more formal document such as a CSR or JEP, so I'll not push on this any more.
> 
> -- Jon



More information about the jtreg-dev mailing list