[CANCEL VOTE] PSR-7: HTTP Message Interface

[Beau Simensen]

Hello all,

Since we put PSR-7 up for a vote, a number of issues have arisen that we feel require attention. In most cases these are clarifications that, had they been made during REVIEW, could have been merged without dropping the spec back to DRAFT.

Sadly, since PSR-7 is now up for a vote, we cannot make clarifications to the spec. We cannot even make clarifications after the spec is accepted, either, except by way of annotations and errata in the meta document.

We've weighed the risk of leaving the spec as-is against canceling the vote and making the required changes directly to the spec itself. This has been an ongoing discussion since the middle of last week.

I had a meeting with Mathew and Paul this morning in which we decided that it would be in the best interest of everyone for us to cancel the vote and make the changes directly.

Specifically:

 * clarify the reasons for why the with* methods must return a new instance even if the values do not change in the meta document

 * clarify withScheme in the spec as to :, :// issues

 * clarify withPath/getPath in the spec as to encoding issues

 * add information on how to handle base path issues in the meta document such that getPath() can always safely return at least "/"

There is no set procedure for what happens when a PSR vote is canceled. As best as we can interpret it, the PSR is still actually in REVIEW. As such, consider PSR-7 still in review but the acceptance vote CANCELED.

Furthermore, there was no guidance in the bylaws to state how long a PSR that has had its vote canceled has to wait before going up for VOTE again. As such, we have decided to leave PSR-7 in REVIEW for at least two weeks, starting today, before we will put PSR-7 up for VOTE once again.

With just over 24 hours from PSR-7 being accepted (based on the votes gathered so far) it was a tough call to make but we believe it is the right one.

[Jeremy Lindblom]

April Fools?

[Beau Simensen]

On Wednesday, April 1, 2015 at 2:55:03 PM UTC-5, Jeremy Lindblom wrote:

April Fools?

Sadly, no. This is not for April Fools' day.

[Jeremy Lindblom]

Where is the thread about: "clarify the reasons for why the with* methods must return a new instance even if the values do not change"?

--
Jeremy Lindblom (@jeremeamia)
Software Engineer at Amazon Web Services

Co-author of the AWS SDK for PHP (@awsforphp)

President of the Seattle PHP User Group (@seaphp)

Organizer of the Pacific Northwest PHP conference (pnwphp.com)

--
You received this message because you are subscribed to the Google Groups "PHP Framework Interoperability Group" group.
To unsubscribe from this group and stop receiving emails from it, send an email to php-fig+u...@googlegroups.com.
To post to this group, send email to php...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/php-fig/79fc2d78-7c09-45c7-8852-82e34a59a913%40googlegroups.com.

For more options, visit https://groups.google.com/d/optout.

[Korvin Szanto]

I don't think you can officially cancel anything on april fools day...

To view this discussion on the web visit https://groups.google.com/d/msgid/php-fig/CALDVupKuxsezf_NY2zJaAgnGZUYUM1f1g1FUM5SLLt9aGzW4Kw%40mail.gmail.com.

[Matthew Weier O'Phinney]

On Wed, Apr 1, 2015 at 3:02 PM, Jeremy Lindblom <jerem...@gmail.com> wrote:
> Where is the thread about: "clarify the reasons for why the with* methods
> must return a new instance even if the values do not change"?

I've opened the following pull requests with the various clarifications:

- Scheme delimiter clarification:
https://github.com/php-fig/fig-standards/pull/480
- Encoding clarifications: https://github.com/php-fig/fig-standards/pull/481
- New instance clarifications: https://github.com/php-fig/fig-standards/pull/482

Beau is working on the one regarding base path issues.

> On Wed, Apr 1, 2015 at 1:00 PM, Beau Simensen <sime...@gmail.com> wrote:
>>
>> On Wednesday, April 1, 2015 at 2:55:03 PM UTC-5, Jeremy Lindblom wrote:
>>>
>>> April Fools?
>>
>>
>> Sadly, no. This is not for April Fools' day.
>>
>> --
>> You received this message because you are subscribed to the Google Groups
>> "PHP Framework Interoperability Group" group.
>> To unsubscribe from this group and stop receiving emails from it, send an
>> email to php-fig+u...@googlegroups.com.
>> To post to this group, send email to php...@googlegroups.com.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/php-fig/79fc2d78-7c09-45c7-8852-82e34a59a913%40googlegroups.com.
>>
>> For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups
> "PHP Framework Interoperability Group" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to php-fig+u...@googlegroups.com.
> To post to this group, send email to php...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msgid/php-fig/CALDVupKuxsezf_NY2zJaAgnGZUYUM1f1g1FUM5SLLt9aGzW4Kw%40mail.gmail.com.

>
> For more options, visit https://groups.google.com/d/optout.

--
Matthew Weier O'Phinney
mweiero...@gmail.com
https://mwop.net/

[Beau Simensen]

We are actively working on making the changes Matthew was adding as errata to being properly represented in the spec/meta docs directly.

 * the with* reasoning: https://github.com/php-fig/fig-standards/pull/482

 * the withScheme* clarification: https://github.com/php-fig/fig-standards/pull/480

The other two may require more work and we need more feedback on them. However, the start of the work for the encoding clarification is here:

 * the encoding clarification: https://github.com/php-fig/fig-standards/pull/481

On Wednesday, April 1, 2015 at 3:02:41 PM UTC-5, Jeremy Lindblom wrote:

Where is the thread about: "clarify the reasons for why the with* methods must return a new instance even if the values do not change"?

Thanks for asking. These didn't exist even 30 minutes ago when I finally clicked send on the first email to this thread but I will add them as they come up.

The clarification on the with* / new instance question was to address the fact that the spec currently dictated that a new instance must be returned. Several implementations were already doing the optimization which made them no longer compliant with the spec. This was a problem.

Since we are in review, we can revisit this and change it so that the spec no longer requires returning a new instance. In the meantime, we need to ensure that the metadoc backs up the reasons why it is written the way it is.

[Alexander Makarov]

Makes sense. Please consider merging (or not) my minor adjustments:

https://github.com/php-fig/fig-standards/pull/468

https://github.com/php-fig/fig-standards/pull/469