db-derby-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Bryan Pendleton (JIRA)" <j...@apache.org>
Subject [jira] [Commented] (DERBY-6164) Improve Derby's Code Coverage - GSoC 2013 Project
Date Wed, 01 May 2013 00:44:16 GMT

    [ https://issues.apache.org/jira/browse/DERBY-6164?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13646227#comment-13646227

Bryan Pendleton commented on DERBY-6164:

Hi Ahsan,

I think your proposal looks quite clear and thorough, thanks for getting it submitted!

I think that your list of packages and classes to work on is a good list; it would be
good to improve the coverage of these classes.

I have two comments to make:

1) In some of the cases that you mentioned, the low coverage may 
be due to a flaw in the coverage measurement tool, rather than due to a lack
of tests. For example, it might be that we actually *do* have tests for
IndexStatisticsDaemonImpl, but for some reason those tests aren't being
counted in the coverage report.

If this is the case, then the task for these classes is not to write new tests,
but to figure out why our current coverage reports aren't correctly measuring
the coverage from the existing tests.

We can ask the development community for assistance with this if we
have trouble figuring it out together.

2) For some of the classes that you mention, we may need to enhance
the Derby test harness to perform better testing. For example, in order to
test the OSGI support in Derby we may need to include some of the
Open Source OSGI "mock" classes that are used for OSGI component
testing, such as PaxExam. In this case we should also see if we 
can fix some of the known OSGI bugs in Derby, for example  DERBY-6069

Similarly, in order to test some of the Security functionality, such as the
LDAP Authentication support, we may need to include some LDAP
mocking support in the Derby test harness.

Testing the security and authentication support would be good, as I suspect
these classes could use some additional testing.

So, overall, I think it is a good proposal, and you have understood the
idea of the project quite well.

> Improve Derby's Code Coverage - GSoC 2013 Project
> -------------------------------------------------
>                 Key: DERBY-6164
>                 URL: https://issues.apache.org/jira/browse/DERBY-6164
>             Project: Derby
>          Issue Type: Task
>          Components: Test
>            Reporter: Bryan Pendleton
>            Assignee: Bryan Pendleton
>            Priority: Minor
>              Labels: gsoc2013
> This year's edition of Google Summer of Code includes, once again, the Apache Software
> For the Derby project, I'm offering to mentor a student on the tasks described here.
> For this project, you are required to have some basic knowledge of JUnit and of how code
coverage works.
> Currently some packages of Derby have poor code coverage. Emma defines the minimum acceptable
code coverage as 70% and some engineers suggest a minimum of 80% is desired. 
> Students are invited to look at [1] to get familiar with Emma's code coverage reports
on Derby. 
> The Derby project conducted this effort last year, and substantial improvements were
> We are looking to continue this process this year, as there are still areas of the Derby
code base where code coverage could be substantially improved.
> The candidate students are also expected to come up with their own schedule for the tests
they plan on creating and to discuss this schedule publicly on the mailing list. 
> After a student applies, a small interview will be conducted via email as part of the
ASF's evaluation process. 
> [1] - http://dbtg.foundry.sun.com/derby/test/coverage/

This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira

View raw message