|
Administrator
|
Hey, all!
Beta5 is almost upon us. The last big merges have occurred, and at this time, the only code-related merge we're waiting on is some code from Evan Coury to streamline the bootstrap script slightly -- this code will have non-breaking effects. If you want to help between now and Thursday, here's what you can do: * Update your sites to master. Very likely, you'll have issues, so go into #zftalk.2 on Freenode IRC to ask what you need to do. We'll be trying to keep track of what changes you need to make so we can put this in a migration document in the changelog. * Test your site once you've updated. Let us know what is broken. Again, check with folks in IRC or the mailing list to get an idea of whether it's broken, or simply changed. Play around with new features -- we now have form annotation support, support for collections in forms, support for hydration of form fieldsets, a new i18n layer, new DB drivers and features, and more. * Help us document! If you come up with good use case samples, get them to somebody, or provide a patch with the use case for us to include. Pieces we know have changed and may affect your code: * The various plugin managers used in the MVC -- controller plugins, view helpers, and controllers themselves -- have slightly different configuration from before -- but the nice part is that they all follow the same pattern, and all exist at the top-level of the configuration. See the following as an example: https://github.com/ZF-Commons/ZfcUser/blob/master/config/module.config.php * Forms have changed slightly. They now differentiate options from attributes -- attributes should only ever translate 1:1 with HTML attributes, while options drive other behavior: labels, CSRF and CAPTCHA configuration, and more. You'll likely need to make some modifications to your forms. * As noted earlier: new i18n/L10n layer. The old code is gone, and the new code relies heavily on ext/intl. Make sure you have that installed, and see what you can and cannot do! For those in the US, enjoy the holiday Wednesday; for everyone else, thanks for your attention! -- Matthew Weier O'Phinney Project Lead | [hidden email] Zend Framework | http://framework.zend.com/ PGP key: http://framework.zend.com/zf-matthew-pgp-key.asc |
|
Hi!
Great, thank you all for hard work! I'm looking forward to new stuff in beta5 (didn't have much time to play with master recently). However, I've just tried to run tests on my Windows machine (using xampp with PHP 5.3.8) and I realized, that intl extension is not enabled there (and is even missing from the list of extension). I had to manually add a line with it. I think it would be tricky for new users.
Without it, the tests fail on Fatal Error immediately (so I suppose Locale stuff won't be working. The test should check for it and eventually skip the relevant tests: I:\xampp\zf2\tests>phpunit
Fatal error: Class 'NumberFormatter' not found in I:\xampp\zf2\tests\Zend\I18n\Filter\NumberFormatTest.php on line 64 Enabling it on localhost is not a problem. Worse is, that both PHP 5.3 webhosts I have access to, does not have intl extension enabled. It means that I don't have any place to run my ZF2 app except for the localhost or VPS.
There are many failing tests for me on Windows. Apart from those failing also in ZF1:
There are some other failures in Zend\Crypt, Zend\Config, Zend\I18n\Validator\FloatTest.php, Zend\Search\Lucene\Storage\File, ZendTest\Uri\UriTest, ZendTest\View\Helper\Navigation, ZendTest\View\Helper\RenderToPlaceholderTest, Zend\View\PhpRendererTest) - see full report https://gist.github.com/3046301.
I should have time to play with beta5 during the weekend, so I'll try to fix some of those failures. Regards,
Martin Hujer On Wed, Jul 4, 2012 at 5:18 AM, Matthew Weier O'Phinney <[hidden email]> wrote: Hey, all! |
|
These are the items on my list that I'd hope to make it into Beta 5:
Happy 4th to everyone in the US and Happy Wednesday to everyone else...time to have a beer or 6 ;) -Nick (@SocalNick) On Wed, Jul 4, 2012 at 2:25 AM, Martin Hujer <[hidden email]> wrote: Hi! |
|
I'd like to simplify what SocalNick asked for Di... I'd simply want the DiAbstractFactory working if there is a `di` config key (and that should happen once merged config is in place)... That pretty much makes up all I want to get back to work without the pain of defining a factory for everything (which, in this early development stage at work, is wasting a lot of my time and making things hard to modify/test). Conversion to SM already brought me to drop DoctrineModule and DoctrineORMModule from my project in favor of my own implementations, even if I'm maintaining them.
Also, I'd like to know if my patches to Di (https://github.com/zendframework/zf2/pull/1517) can be applied, since I am already using my OcraDiCompiler compiler in a staging environment and it is working like a charm, but is based on a patched ZF2 fork... Marco Pivetta http://twitter.com/Ocramius http://marco-pivetta.com On 4 July 2012 20:41, Nicholas Calugar <[hidden email]> wrote: These are the items on my list that I'd hope to make it into Beta 5: |
|
In reply to this post by SocalNick
On Wed, Jul 4, 2012 at 11:41 AM, Nicholas Calugar <[hidden email]> wrote:
These are the items on my list that I'd hope to make it into Beta 5: I've taken care of #3: https://github.com/zendframework/zf2/pull/1495#commits-pushed-3a53fb7
I started with the solution from Nick's hotfix/service-listener-lazy-load-service-manager branch, but refactored the ModuleManager to be more encapsulated within listeners so that the extra loadModule.finish event is no longer necessary.
Still looking into #4. --- Evan Coury, ZCE |
|
Am 05.07.2012 12:33, schrieb Evan Coury:
> On Wed, Jul 4, 2012 at 11:41 AM, Nicholas Calugar > <[hidden email]> wrote: >> * A hostname route that matches optional subdomains. I originally >> submitted this PR but DASPRiD said he wanted the feature built into >> the Hostname route. https://github.com/zendframework/zf2/pull/1386 >> [1] >> * I noticed that the ZendCacheStorageFactory doesn't work >> correctly for Memcached. You can't override the default localhost >> server when you instantiate via factory. There is a note in the >> adapter to allow changing of servers after instantiation, but here >> is another fix that I can PR if you like: I'm likely to touch this issue after B5. |
|
In reply to this post by EvanDotPro
On Thu, Jul 5, 2012 at 3:33 AM, Evan Coury <[hidden email]> wrote:
For anyone following this, Nicholas Calugar helped out with fixing #4, so DI fallback should be working well for beta5! If you're one of the people still using DI, you owe Nicholas a beer! ;)
|
|
Owing Nick a lot of beer! ( <3 Di )
Marco Pivetta http://twitter.com/Ocramius http://marco-pivetta.com On 6 July 2012 18:42, Evan Coury <[hidden email]> wrote:
|
|
In reply to this post by Ben Scholzen 'DASPRiD'
Ben,
Thanks for keeping this on your radar. Do you need me to file an issue to make it official? -Nick On Fri, Jul 6, 2012 at 7:29 AM, Ben Scholzen <[hidden email]> wrote: Am 05.07.2012 12:33, schrieb Evan Coury: |
| Powered by Nabble | Edit this page |
