[ 
https://issues.apache.org/jira/browse/HADOOP-11293?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Steve Loughran updated HADOOP-11293:
------------------------------------
    Attachment: HADOOP-11293-branch-2-005.patch

This is the patch merged with Branch-2; some minor conflict in 
{{TestLocalFileSystem}}

Now what?

If it weren't for {{Shell}} doing too much in its static initialisers I'd be -1 
on the basis if "needless change across the codebase whose purpose is to make 
patches hard"

And, if the patch didn't maintain 100% backwards compatibility with the Shell.* 
constants, again, It'd be a -1, here on "breaks things"

But shell is messy; this code does factor it out, ultimately it will be good if 
broadly adopted, which is what this patch does across the hadoop code.

So, while I'm ~=0 on the need for it, there's enough slightly positive to say 
+1, if we can do the merge right.

Which is where we are now. 

I propose that 
# hadoop-common patch goes in to trunk
# there are separate JIRAs created for HDFS, YARN and MAPREDUCE, each with 
their own patches.
That way: each project gets to see what is happening, and isn't surprised by a 
big change across the codebase.

There's one more remaining issue: branch-2. We do need consistency across the 
branches to aid in applying may other patches. 

So: branch-2-for-2.7 vs 2.8? 



> Factor OSType out from Shell
> ----------------------------
>
>                 Key: HADOOP-11293
>                 URL: https://issues.apache.org/jira/browse/HADOOP-11293
>             Project: Hadoop Common
>          Issue Type: Improvement
>          Components: fs, util
>    Affects Versions: 2.7.0
>            Reporter: Yongjun Zhang
>            Assignee: Yongjun Zhang
>         Attachments: HADOOP-11293-branch-2-005.patch, HADOOP-11293.001.patch, 
> HADOOP-11293.002.patch, HADOOP-11293.003.patch, HADOOP-11293.004.patch, 
> HADOOP-11293.005.patch, HADOOP-11293.005.patch, HADOOP-11293.005.patch, 
> HADOOP-11293.005.patch
>
>
> Currently the code that detects the OS type is located in Shell.java. Code 
> that need to check OS type refers to Shell, even if no other stuff of Shell 
> is needed. 
> I am proposing to refactor OSType out to  its own class, so to make the 
> OSType easier to access and the dependency cleaner.
>  



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)

Reply via email to