Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [cdt-dev] new metric plug-in based on Codan checkers

Hi Everyone

metriculator is now available on github!

Feel free to fork the repo on: https://github.com/ideadapt/metriculator
or to create new "open issues" on github of ideadapt (or just in your own fork ;) )

The CI-Server builds every night, if there are changes, at 02:00 am  (UTC+01:00) 

Test coverage reports are available on http://ideadapt.net/share/metriculator/coverage/coverage.html after every successful build.

If you plan to do "pull requests" you can watch if your potential contribution would build successfully at http://travis-ci.org/#!/ideadapt/metriculator/pull_requests

If you have any further questions please do not hesitate to ask.

Best regards,

Ueli Kunz (http://www.ideadapt.net/)
Jules Weder

2012/5/16 Ueli Kunz <kunz@xxxxxxxxxxxx>
Just want to announce that metriculator is now available at the marketplace (http://marketplace.eclipse.org/content/metriculator). And we successfully ran metriculator against the chromium source code base which has about 9 mio. psloc and 3 mio lsloc.


2012/4/17 Ueli Kunz <kunz@xxxxxxxxxxxx>
Hey,

Okey. We already tested metriculator against the llvm project (1 mio.
sloc) and the doom project (1/2 mio sloc). Do you now how many sloc
your project has? We used the following command to get a number:
find . -print | egrep '\.cpp$|\.h$' | xargs cat | sed '/^\s*$/d' | wc -l

Maybe I try running it against the source code of the google chromium
project, which seems to be fairly big as well, according to
https://www.ohloh.net/p/chrome/analyses/latest

2012/4/17 Marc-Andre Laperle <malaperle@xxxxxxxxxxxxx>:
> Hi,
>
> Unfortunately, I can't share the code but you can probably get similar
> results with open source projects of reasonable size. I have tested the new
> version (0.0.1.201204081558) and it seems to run much faster and without
> memory leaks. But it seems I can't see the results if I run it on a folder,
> I can only see the results when selecting files.
>
> Marc-Andre
>
>
> On 12/04/2012 7:01 AM, Ueli Kunz wrote:
>>
>> Hey Marc-Andre
>>
>> Thanks for testing and your feedback. We have further improved
>> metriculator, solved some bugs and again increased the performance. Is
>> there a way to get the source code you ran metriculator against? We
>> would like to see if it now works and if not be able to further
>> investigate the issue. Having the "problematic" source code could
>> simplify that task.
>>
>> 2012/3/28 Marc-Andre Laperle<malaperle@xxxxxxxxxxxxx>:
>>>
>>> Thanks, with this change, I can install it. I ran it on a single file and
>>> it
>>> worked. Then I tried running it on only a subfolder (around 500
>>> source/header files) and quickly ran out of memory (using -Xmx1100m, it
>>> should be plenty). I attached a screenshot from Eclipse MAT.
>>>
>>>
>>> On 20/03/2012 8:33 AM, Ueli Kunz wrote:
>>>>
>>>> @Marc-Andre
>>>> The HSR is willing to maintain the plug-in in the future. Peopple at
>>>> HSR had already contributed other plug-ins and classes to cdt.
>>>
>>>
>>> That's good to hear. Unfortunately, there is no (active) committer from
>>> HSR
>>> at the moment so this is more difficult. It would take a committer who
>>> has
>>> sufficient interest in the feature to review and apply the patches.
>>> Perhaps
>>> Metriculator would be a good fit for the Eclipse Marketplace.
>>>
>>> Marc-Andre
>>>
>>>
>>> On 20/03/2012 8:33 AM, Ueli Kunz wrote:
>>>>
>>>> Hi,
>>>>
>>>> Thank you for your feedbacks. We have beein working on metriculator
>>>> the last weeks and respected your feedbacks in our latest
>>>> implementations. Metriculator now ignores C source code files, since
>>>> our aim is to analyse C++ and to produce accurate metric values. The
>>>> tag cloud component is now a separate feature. So the installation of
>>>> just metriculator is simpler now, since no more zest dependencies
>>>> exist.
>>>> Further more we have heavely increased the performance, analysing 1
>>>> mio. sloc is not a problem anymore.
>>>>
>>>> The latest nightly build is available for installation at
>>>> http://sinv-56013.edu.hsr.ch/metriculator/updatesite-nightly/
>>>>
>>>> @kessel
>>>> I can not reproduce any NPE. Could you please provide further details
>>>> about the exception? Maybe the NPE is fixed in the nightly build.
>>>>
>>>> @Marc-Andre
>>>> The HSR is willing to maintain the plug-in in the future. Peopple at
>>>> HSR had already contributed other plug-ins and classes to cdt.
>>>>
>>>> ueli
>>>>
>>>> 2012/3/2 kesselhaus<kesselhaus@xxxxxxx>:
>>>>>
>>>>> Hi,
>>>>>
>>>>> Same here, needed a fresh new install of Eclipse/CDT (Indigo SR2) since
>>>>> I
>>>>> have another plugin requiringGEF/Zest 1.x ...
>>>>> I get NPEs when scrolling through the metricular list to fast.
>>>>>
>>>>> Another frustration comes from the C++ only functionality again .. for
>>>>> C,
>>>>> this plugins output makes no sense.
>>>>> C headers get strange McCabe values>0 ... some even get values of>350!!
>>>>> Since I do no C++ programming, I can't say anything about the
>>>>> functionality
>>>>> on C++ projects.
>>>>>
>>>>> - kessel
>>>>>
>>>>>
>>>>> Am 29.02.2012 23:04, schrieb Marc-Andre Laperle:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> I tried to install it but couldn't because I couldn't install
>>>>>> zest.cloudio. I think the zest.cloudio integration should be an
>>>>>> optional
>>>>>> feature. I have no opinion right now on whether or not it should be
>>>>>> committed to CDT but we have to make sure someone will maintain it.
>>>>>>
>>>>>> Marc-Andre
>>>>>>
>>>>>> On 29/02/2012 3:49 PM, Ueli Kunz wrote:
>>>>>>>
>>>>>>> Greetings,
>>>>>>>
>>>>>>> We have worked on a CDT plug-in (metriculator) that uses Codan
>>>>>>> checkers to create static software metrics. As far as we know there
>>>>>>> is
>>>>>>> no metric plug-in available for CDT. Metriculator aims to fill that
>>>>>>> gap.
>>>>>>>
>>>>>>> Metriculator was born as subject of our semester thesis at the HSR,
>>>>>>> where Prof. Peter Sommerlad (initiator of CUTE) supervised our work.
>>>>>>>
>>>>>>> For more details please visit our project wiki at
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> http://sinv-56013.edu.hsr.ch/redmine/projects/metricular/wiki/Documentation
>>>>>>> We created a bug report at bugzilla
>>>>>>> https://bugs.eclipse.org/bugs/show_bug.cgi?id=372722
>>>>>>>
>>>>>>> We would really appreciate your feedback. Do you think its reasonable
>>>>>>> to commit it to CDT at some point?
>>>>>>>
>>>>>>> Regards
>>>>>>> Ueli Kunz, Jules Weder
>>>>>>> _______________________________________________
>>>>>>> cdt-dev mailing list
>>>>>>> cdt-dev@xxxxxxxxxxx
>>>>>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>>>>
>>>>>>
>>>>>> _______________________________________________
>>>>>> cdt-dev mailing list
>>>>>> cdt-dev@xxxxxxxxxxx
>>>>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>>>>
>>>>> _______________________________________________
>>>>> cdt-dev mailing list
>>>>> cdt-dev@xxxxxxxxxxx
>>>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>>
>>>>
>>>>
>>>
>>> _______________________________________________
>>> cdt-dev mailing list
>>> cdt-dev@xxxxxxxxxxx
>>> https://dev.eclipse.org/mailman/listinfo/cdt-dev
>>>
>>
>>
>
> _______________________________________________
> cdt-dev mailing list
> cdt-dev@xxxxxxxxxxx
> https://dev.eclipse.org/mailman/listinfo/cdt-dev



--
http://www.ideadapt.net - where ideas become code



--
http://www.ideadapt.net - where ideas become code

_______________________________________________
cdt-dev mailing list
cdt-dev@xxxxxxxxxxx
https://dev.eclipse.org/mailman/listinfo/cdt-dev



Back to the top