Smalltalk › Frameworks & Tools › Moose

PPCompositeTest

_‹ Previous Topic Next Topic _›

Classic

List

Threaded

3 messages Options

Tudor Girba

PPCompositeTest

Hi Lukas,

I noticed that you override two times the changes I pushed in the PPCompositeTest, and as expected, these changes break most of the external tests :).

In particular, there are two things that I found useful:
- the result instance variable holding the side effect of executing #parse:rule:. This is useful for reusing the grammar tests in the parser tests.
- the fail:rule: method for negative tests.

In the last comment, you mentioned that this was due to incompatibilities with Helvetia. Should we create a new class with these utilities?

Cheers,
Doru

--
www.tudorgirba.com

"When people care, great things can happen."

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Lukas Renggli

Re: PPCompositeTest

> I noticed that you override two times the changes I pushed in the PPCompositeTest, and as expected, these changes break most of the external tests :).

No, I didn't override the changes. I just didn't merge up to now them
because there was too much work involved. To my understanding the
changes are integrated now, no?

The problem was that the added 'result' instance-variable caused
variable shadowing in PPCompositeTest itself and many subclasses.
Additionally it broke some Helvetia tests that I refactored now.

> In particular, there are two things that I found useful:
> - the result instance variable holding the side effect of executing #parse:rule:. This is useful for reusing the grammar tests in the parser tests.

As far as I see this is in there.

> - the fail:rule: method for negative tests.

Ok sorry, I think I removed this because there were no senders.

Lukas

--
Lukas Renggli
www.lukas-renggli.ch

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev

Tudor Girba

Re: PPCompositeTest

Ah, my bad.

Indeed the result is there now, I did not look carefully in the diff view :).

I added #fail:rule: back.

Thanks a lot, and have a nice time in South America :)

Cheers,
Doru

On 30 Nov 2010, at 03:11, Lukas Renggli wrote:

>> I noticed that you override two times the changes I pushed in the PPCompositeTest, and as expected, these changes break most of the external tests :).
>
> No, I didn't override the changes. I just didn't merge up to now them
> because there was too much work involved. To my understanding the
> changes are integrated now, no?
>
> The problem was that the added 'result' instance-variable caused
> variable shadowing in PPCompositeTest itself and many subclasses.
> Additionally it broke some Helvetia tests that I refactored now.
>
>> In particular, there are two things that I found useful:
>> - the result instance variable holding the side effect of executing #parse:rule:. This is useful for reusing the grammar tests in the parser tests.
>
> As far as I see this is in there.
>
>> - the fail:rule: method for negative tests.
>
> Ok sorry, I think I removed this because there were no senders.
>
> Lukas
>
> --
> Lukas Renggli
> www.lukas-renggli.ch
>
> _______________________________________________
> Moose-dev mailing list
> [hidden email]
> https://www.iam.unibe.ch/mailman/listinfo/moose-dev

--
www.tudorgirba.com

"From an abstract enough point of view, any two things are similar."

_______________________________________________
Moose-dev mailing list
[hidden email]
https://www.iam.unibe.ch/mailman/listinfo/moose-dev