Recent changes to feature-requests

Java 11 Compatibility check: EMMA

2018-08-16T10:54:57.608000Z

Hi,
We are reaching out to you to check Java 11 compatibility of the library EMMA.

Could you help us by answering the following questions:

Library Name: <library name.="" e.g:="" activation=""></library>
Latest version: <latest 1.1.1="" eg:="" version.=""></latest>
Latest version Is the library compatible with Java 11 Compatible? (Y/N)
Is the library supported with Java 11? (Y/N)
(If “N” in compatibility or support) What is the versions that would be compatible and Supported?
Date of support availability?

Appreciate your response by 19.08.2018.

Thanks,
Krzysztof Dragan,
PTC Inc. Contractor

EMMA news

2011-12-29T15:27:45Z

Hi, I would like to know if you guys are still working on EMMA. If you are working in some new released, bugfix or something. And i'm not talking aboute the plug-in.

Thank you.

MIDlet support

2011-02-19T10:32:56Z

Hi,

I would really like to use emma on a J2ME MIDP 2.0 MIDlet.

Don't know how you would implement it. Maybe via some hooks in the WTK emulator?

Thanks!

In-place instrumentation should add emma.jar to classpath

2011-01-26T12:12:51Z

I love EMMA's in-place .jar file instrumentation ability - EMMA is the only tool that has this feature as far as I can tell, and it allows us to run an instrumentation step on all our built jars and then execute our enormous python-based test framework - without making changes to hundreds of python and ant files to turn coverage on for each testcase individually (e.g. no need to change classpath to include the instrumented jar directory each time we invoke java.exe).

The only problem is that emma.jar needs to be explicitly added to the classpath whenever we invoke java - and that's *really* hard when there are a large number of places it's invoked (from python, from hundreds of separate ant scripts, from eclipse, etc); putting emma.jar into the jvm lib/ext directory helps but is clunky and doesn't always work (e.g. ClassDefNotFound if executing java from ant under linux if forked=false!).

So I wonder if it would be possible to have an instrumentation option for hardcoding the location of the emma.jar during the instrumentation process so that once a .jar has been instrumented it can run with NO additional jvm/classpath arguments at all.

This would be a really awesome feature :o)

Filter methods from the coverage report

2010-02-09T05:20:13Z

Hi all,
I think that would be interesting to have the ability to filter methods from the coverage report. There are methods that aren't important (for a lack of a better adjective) enough to be tested. For example, if I override a toString() method for debug purposes, or dummy get/set methods. The problem is that the coverage report is polluted from non-tested code that from a developer point of view shouldn't be tested (well, just simply not interested in test it).
Thanks

Block coverage makes it impossible to cover unit tests

2009-08-09T02:26:56Z

When running coverage tests, I'm in the habit of running them on my unit tests in addition to the code under test. This has allowed me to catch code in unit tests that I thought was behaving a certain way but turned out not to.

But EMMA won't let me do that. Consider code like this:

wordsManager = new WordsManager(mDataAccessor, new PriorityWordsFilter(3, 2, true));
assertIsTrue(wordsManager.getNotificationMode() == WordsManager.NOTIFICATION_MODE_NO_CHANGE);
assertNotNull(wordsManager.getPriorityWords());
assertIsTrue(wordsManager.getPriorityWords().length == 0);
wordsManager.updatePriorityWordsFilter(new PriorityWordsFilter(3, 2, true));
assertIsTrue(wordsManager.getNotificationMode() == WordsManager.NOTIFICATION_MODE_NO_CHANGE);
assertNotNull(wordsManager.getPriorityWords());
assertIsTrue(wordsManager.getPriorityWords().length == 0);

Each of the lines with an equality test is marked yellow, not covered, by EMMA. But of *course* the expressions are only going to evaluated to true or to false--that's the whole point of passing them into the asserts. That's how JUnit works!

Annotation support

2009-08-07T18:00:38Z

Using spring annotation to declare a method to expect a specific type of exception.

These methods do not get identified by the coverage tool and show they are 0% covered.

Need an annotation to force the coverage tool to do the following:

Omit method from being displayed and calculated in the coverage results

Force the coverage tool to identify the method has been covered at %100

Add more infos in XML report

2009-04-07T07:58:20Z

Hi,

i am using emma with hudson and it seems the xml report produced does not contains all the same infos than the HTML report.
The result is then quite ugly ...

see https://hudson.dev.java.net/issues/show_bug.cgi?id=3402

Cheers

request CoverageData.diff() method

2008-08-18T23:43:21Z

I request a method in CoverageData that returns the difference between two runs instead of a merge between two runs. I'll start this RFE by explaining why, then discuss how, then go into a more extended discussion.

WHY

I've been doing research on programmer productivity, and noted that there are situations when developers have a really hard time finding where the code that implements something happens. For example, in one of the tasks that I observed, developers needed to find where to add code that executed only when a specific pane got resized. They flailed completely.

I modified EMMA and EclEmma so that the developers could do two runs of their code and then "diff" them, so they could see what was different between the two runs. If they did one run where they {created and resized a pane, then quit} and then one run where they {created a pane and then hovered on the corner long enough to get the cursor to change its shape, then quit}, then the diffed coverage report had one and only one method in it -- which was the method that they needed to change.

I implemented this tool by modifying EclEmma and EMMA. I'd like to release this tool to the world, but feel uncomfortable about releasing a modified version of EMMA. I'd much rather my mod were rolled into EMMA.

HOW

I made a few very minor changes in CoverageData.java in emma-2.0.5312 to add a diff() routine that is a very very slight modification of CoverateData.merge(). I've attached my version of the file; stuff I changed is marked with "@@@".

I wasn't completely sure what m_stamp was, but the code I attached does what I need, and shouldn't break any existing code.

EXTENDED DISCUSSION

It would be really nice to be able to colour lines of codes based on whether something was in the first run but not the second, in the second but not the first, in both, or in neither. I didn't see an easy way to do that; if you have any ideas, I'd be quite open to hearing them.

Support for comment to exclude lines

2008-07-30T15:15:18Z

Please add a special comment to mark lines in the source as "these will never be executed".