Re: [tap] Valids subtests (was: RFC Status?)

["Bruno P. Kinoshita" <[e-mail]>]

The source code is hosted at [1]. The web site is generated during the release process at [2]. The test app is deployed at [3] (though it is outdated and needs to be updated). 

I first tried writing a parser by taking a look at snake-yaml, but ended up depending a lot on regex's. So probably that's a part that could be reworked to better represent grammars such as the one you posted earlier. 

Please, feel free to submit issues, pull requests, comments/suggestions :)
Thanks!

[1] https://github.com/tupilabs/tap4j
[2] http://tap4j.org

[3] http://instanttap.appspot.com/



>________________________________
> From: Ovid <[e-mail]>
>To: Bruno P. Kinoshita <[e-mail]> 
>Cc: "[e-mail]" <[e-mail]> 
>Sent: Wednesday, September 17, 2014 5:40 PM
>Subject: Re: [tap] Valids subtests (was:  RFC Status?)
> 
>
>
>Bruno,
>
>
>Is there a git project for this anywhere? I haven't written any Java for a while and with my company, I don't have a lot of spare time, but I'd be happy to take a look.
> 
>Best,
>Ovid
>--
>IT consulting, training, international recruiting
>     
  http://www.allaroundtheworld.fr/.
>Buy my book! - http://bit.ly/beginning_perl
>Live and work overseas - http://www.overseas-exile.com/
>
>
>
>On Wednesday, 17 September 2014, 22:38, Bruno P. Kinoshita <[e-mail]> wrote:
> 
>
>>
>>
>>Haha, thanks Ovid!
>>
>>
>>Well, turns out tap4j TAP parser is likely broken for subtests then :)
>>
>>
>>I'll probably spend some time writing a new parser - which I'll try to make more compatible with Perl tools and with existing grammars. 
>>
>>
>>But some definitive documentation for subtests in TAP website would be grand too.
>>
>>
>>Cheers
>>Bruno
>>
>>
>>
>>>________________________________
>>> From: Ovid <[e-mail]>
>>>To: Bruno P. Kinoshita <[e-mail]> 
>>>Cc: "[e-mail]" <[e-mail]> 
>>>Sent: Wednesday, September 17, 2014 5:33 PM
>>>Subject: Re: [tap] Valids subtests (was:  RFC Status?)
>>> 
>>>
>>>
>>>Bruno,
>>>
>>>
>>>I realized that referring you to the grammar is the same as RTFM and that's not fair when there are questions about the FM :)
>>>
>>>
>>>    1..2
>>>    ok 1 - foo
>>>        not ok 1 - some text
>>>    ok 2 - bar passed
>>>        not ok 1 - another text
>>>
>>>
>>>The above is not valid because there must always be summary ok/not ok lines *after* the subtest and the subtest must have a leading XOR trailing plan (in other words, not both) with the same indentation level as the subtest and the summary line *must* have an ok/not ok status that matches the results of the subtest. The following would be valid:
>>>
>>>
>>>    1..2
>>>        not ok 1 - some text
>>>        1..1
>>>    not ok 1 - foo (because the subtest did not pass)
>>>        1..2
>>>        ok 1 - another text
>>>        not ok 2 - whee! # TODO (todo tests are always considered as passing)
>>>    ok 2 - bar passed (because the subtest did pass)
>>>
>>>
>>>
>>>
>>>
>>>
>>>
>>>Best,
>>>Ovid 
>>>
>>>--
>>>IT consulting, training, international recruiting
>>>       http://www.allaroundtheworld.fr/.
>>>Buy my book! - http://bit.ly/beginning_perl
>>>Live and work overseas - http://www.overseas-exile.com/
>>>
>>>
>>
>>
>
>