> On 2012-04-26 19:58:44, Henry Saputra wrote:
> > http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/core.io/io.js,
> >  line 436
> > <https://reviews.apache.org/r/4874/diff/5/?file=104573#file104573line436>
> >
> >     Why change from === to == ?

It only looks like I changed it, the original one isn't aligning correctly in 
this diff, it's further down.

=== is the exact same object and == is just as good, if not better for the 
spirit of what we're doing here.
I really think we're way overusing === in shindig, but I left the original 
check as-is.


> On 2012-04-26 19:58:44, Henry Saputra wrote:
> > http://svn.apache.org/repos/asf/shindig/trunk/features/src/test/javascript/features/core.io/iotest.js,
> >  line 179
> > <https://reviews.apache.org/r/4874/diff/5/?file=104574#file104574line179>
> >
> >     I thought this is a test to disable refresh? Why would we change it to 
> > GET?

Cause the shindig server will set no-cache headers (based on the request to not 
cache it) and tell our browser not to cache it.  There's no reason to redirect 
it through a post just because we want no refresh.


- Dan


-----------------------------------------------------------
This is an automatically generated e-mail. To reply, visit:
https://reviews.apache.org/r/4874/#review7272
-----------------------------------------------------------


On 2012-04-26 18:33:29, Dan Dumont wrote:
> 
> -----------------------------------------------------------
> This is an automatically generated e-mail. To reply, visit:
> https://reviews.apache.org/r/4874/
> -----------------------------------------------------------
> 
> (Updated 2012-04-26 18:33:29)
> 
> 
> Review request for shindig, Ryan Baxter, Stanton Sievers, and Brian Lillie.
> 
> 
> Summary
> -------
> 
> So I think this will clean up makeRequest from the client's perspective.
> 
> The implications of this are that if we use the refreshInterval param we 
> should not return the original response but instead return the response that 
> has the clobbered cache headers.  This is because we are processing the get 
> as-is and returned cache-control headers need to be valid for what we've 
> requested.
> 
> There is currently an issue in 
> org.apache.shindig.gadgets.http.AbstractHttpCache where when building the 
> response to cache, we clobber the cache-control headers and do not return the 
> modified response. If these changes listed here for makeRequest make sense, 
> we can continue and fix this other problem.
> 
> 
> Diffs
> -----
> 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/test/java/org/apache/shindig/gadgets/http/DefaultRequestPipelineTest.java
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/features/src/test/javascript/features/core.io/iotest.js
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/features/src/main/javascript/features/core.io/io.js
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/main/java/org/apache/shindig/gadgets/http/AbstractHttpCache.java
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/main/java/org/apache/shindig/gadgets/http/DefaultRequestPipeline.java
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/main/java/org/apache/shindig/gadgets/http/HttpCache.java
>  1329959 
>   
> http://svn.apache.org/repos/asf/shindig/trunk/java/gadgets/src/test/java/org/apache/shindig/gadgets/http/AbstractHttpCacheTest.java
>  1329959 
> 
> Diff: https://reviews.apache.org/r/4874/diff
> 
> 
> Testing
> -------
> 
> Updated tests.  All tests pass.
> 
> 
> Thanks,
> 
> Dan
> 
>

Reply via email to