[
https://issues.apache.org/jira/browse/MAPREDUCE-1623?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12860816#action_12860816
]
Chris Douglas commented on MAPREDUCE-1623:
------------------------------------------
bq. Agree with Amareshwari that not all of the package is public, and that we
probably need a public API to it, but it's not clear what it is yet. By marking
as public evolving we can evolve it to make things private that should be
private. The other way works too - mark it private unstable or evolving and
make public the parts that need to be made public in a later release. I can go
either way.
bq. Chris, I wanted to say that the whole package is not public, though some of
the classes in the package are public. For example, JobHistory.java has
framework code, which creates/deletes job history files, manages history files
and etc.
OK, that's fair. There may be a few tool authors who would be interested in a
stable API here, but I think Amareshwari is right, and whatever form that takes
belongs with the rest of the client APIs.
bq. I'm a bit confused about what happened here. In 0.20 there was only
o.a.h.filecache.DistributedCache, which was actually in Core. It's been
deprecated in trunk in favour of o.a.h.filecache.mapreduce.DistributedCache,
but this has been deprecated too - in favour of methods on Job. So, yes,
mapreduce.filecache package should be Private, and the deprecation message on
o.a.h.filecache.DistributedCache should direct users to Job.
I found this confusing, also. It looks like most of the javadoc for
{{mapreduce.filecache}} can be moved back to
{{o.a.h.filecache.DistributedCache}}, then the former package can be marked as
Private.
bq. Thanks for updating the patch with this. Is this the only change you made?
(It's hard to see with such a big patch.)
Yes, those should be the only changes.
> Apply audience and stability annotations to classes in mapred package
> ---------------------------------------------------------------------
>
> Key: MAPREDUCE-1623
> URL: https://issues.apache.org/jira/browse/MAPREDUCE-1623
> Project: Hadoop Map/Reduce
> Issue Type: Sub-task
> Components: documentation
> Reporter: Tom White
> Assignee: Tom White
> Attachments: M1623-1.patch, MAPREDUCE-1623.patch,
> MAPREDUCE-1623.patch, MAPREDUCE-1623.patch, MAPREDUCE-1623.patch
>
>
> There are lots of implementation classes in org.apache.hadoop.mapred which
> makes it difficult to see the user-level MapReduce API classes in the
> Javadoc. (See
> http://hadoop.apache.org/common/docs/r0.20.2/api/org/apache/hadoop/mapred/package-summary.html
> for example.) By marking these implementation classes with the
> InterfaceAudience.Private annotation we can exclude them from user Javadoc
> (using HADOOP-6658).
> Later work will move the implementation classes into o.a.h.mapreduce.server
> and related packages (see MAPREDUCE-561), but applying the annotations is a
> good first step.
--
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.