[ 
https://issues.apache.org/jira/browse/THRIFT-3706?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=15874601#comment-15874601
 ] 

ASF GitHub Bot commented on THRIFT-3706:
----------------------------------------

Github user jeking3 commented on the issue:

    https://github.com/apache/thrift/pull/1191
  
    Discussing with folks who created crosstest what we should do.  My 
preference is to have a single "multiplexed" protocol test that uses the binary 
protocol.  That will be the most widely used combo, and layering multiplexed on 
top of any other protocol does not test any new functionality of multiplexed; 
but it might catch strange interactions that we could miss if we don't test all 
of them.  I'm not fond of the explosion of the test matrix, doubling all the 
tests potentially.  
    
    If you look in tests.json you will see protocols like "binary" and then 
"binary:accel", also "compact" and "compact:accelc".  This naming format 
basically says that if crosstest is testing "binary" it will match "binary" and 
"binary:accel", however the test wll be run once with a --protocol of "binary" 
and once with a --protocol of "accel".  I didn't want to add "binary:multi", 
"compact:multic", "json:multij", "header:multih" but that would be the logical 
extension of the pattern that already exists in crosstest.
    
    So I'm working on a copy of your branch right now to simplify down to 
"multiplexed" and then I want to see it run in the crosstest locally.  Then 
once I hear back from some folks we can move forward if they are okay with the 
simplification.


> There's no support for Multiplexed protocol on c_glib library
> -------------------------------------------------------------
>
>                 Key: THRIFT-3706
>                 URL: https://issues.apache.org/jira/browse/THRIFT-3706
>             Project: Thrift
>          Issue Type: Improvement
>          Components: C glib - Library
>    Affects Versions: 0.9.3
>            Reporter: Gonzalo Aguilar
>            Assignee: Gonzalo Aguilar
>
> There's no multiplexed protocol. 
> I will implement the same way it's done in Java an C++



--
This message was sent by Atlassian JIRA
(v6.3.15#6346)

Reply via email to