This adds red and green bars to the html report. It shouldn't make the report any wider than it already is.
I still can't attach files to bug reports, so I am continuing to post patches to my web site for you to download. Please let me know as the patches are applied to the tree, so that I can remove them from my site.
Ryan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks. I have downloaded both this and the patch for excluding empty constructors. Sorry about being slow here: it's a challenge to keep this project going along with my day job.
Incidentally, the reason the empty constructors contains the aload_0/invokespecial/return insructions sequence is that they are never truly empty: every constructor in Java chains to the super constructor (except for java.lang.Object). Thus, I think this exclusion should be triggered in a different way.
I hope the stream of bugs I consider reasonably high priority has dried up now and we can start new feature development in 2.1 branch. Because the new exclusion rules (empty constructors etc) will change the coverage stats, I think this should go into 2.1, not 2.0.
Incidentally, I sent an invite to join EMMA dev team to your SF alias. The offer is still open if you are interested.
I hope to find some time to publish instructions on the build and dev process for the project.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I understand how hard it is to keep an open source project going and work a day job. No worries about how fast you get to my patches.
Personally, I would like to be able to exclude methods based on some metadata that the developer includes in their source file, but that can't be done easily until java 1.5. But, for private constructors, using the logic I posted actually makes sense. If you think about what must be done in a method, if you just have those three byte-code instructions in a private constructor, it must be an empty no-arg constructor. Even simply passing a different argument to the parent's constructor will cause more byte-code instructions to load the variable into a register.
I haven't received the invitation, but I would like to join the project. I will dig through the SF site to try to figure out how to do so. If I can't find it though, can you invite rbb@rkbloom.net?
Ryan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I haven't been able to find the invitation to join the project, and SF doesn't seem to have a page I can go to to accept the invitation. Can you send another one?
Ryan
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I am still figuring out what the different SF permissions correspond to. I've noticed the problem of attaching files to tracker issues has been submitted to SF support a few times. Apparently, only the admins and the original issue submitters can attach files.
I have added you as a project member. I want to start by giving you the ability to attach files (patches) to bug/RFE issues but not full CVS access (yet). Could you check if you see the "attach file" box in the tracker now?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Ryan, I tried your patch to add the red/green bars... Works fine. Thanks, it helps me move some users off of another coverage tool onto emma(they just had to have those colored bars)..
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have attached a patch to http://sourceforge.net/tracker/index.php?func=detail&aid=962015&group_id=108932&atid=651900
This adds red and green bars to the html report. It shouldn't make the report any wider than it already is.
I still can't attach files to bug reports, so I am continuing to post patches to my web site for you to download. Please let me know as the patches are applied to the tree, so that I can remove them from my site.
Ryan
Thanks. I have downloaded both this and the patch for excluding empty constructors. Sorry about being slow here: it's a challenge to keep this project going along with my day job.
Incidentally, the reason the empty constructors contains the aload_0/invokespecial/return insructions sequence is that they are never truly empty: every constructor in Java chains to the super constructor (except for java.lang.Object). Thus, I think this exclusion should be triggered in a different way.
I hope the stream of bugs I consider reasonably high priority has dried up now and we can start new feature development in 2.1 branch. Because the new exclusion rules (empty constructors etc) will change the coverage stats, I think this should go into 2.1, not 2.0.
Incidentally, I sent an invite to join EMMA dev team to your SF alias. The offer is still open if you are interested.
I hope to find some time to publish instructions on the build and dev process for the project.
I understand how hard it is to keep an open source project going and work a day job. No worries about how fast you get to my patches.
Personally, I would like to be able to exclude methods based on some metadata that the developer includes in their source file, but that can't be done easily until java 1.5. But, for private constructors, using the logic I posted actually makes sense. If you think about what must be done in a method, if you just have those three byte-code instructions in a private constructor, it must be an empty no-arg constructor. Even simply passing a different argument to the parent's constructor will cause more byte-code instructions to load the variable into a register.
I haven't received the invitation, but I would like to join the project. I will dig through the SF site to try to figure out how to do so. If I can't find it though, can you invite rbb@rkbloom.net?
Ryan
I haven't been able to find the invitation to join the project, and SF doesn't seem to have a page I can go to to accept the invitation. Can you send another one?
Ryan
I am still figuring out what the different SF permissions correspond to. I've noticed the problem of attaching files to tracker issues has been submitted to SF support a few times. Apparently, only the admins and the original issue submitters can attach files.
I have added you as a project member. I want to start by giving you the ability to attach files (patches) to bug/RFE issues but not full CVS access (yet). Could you check if you see the "attach file" box in the tracker now?
I can add files to bugs now
Ryan
Ryan, I tried your patch to add the red/green bars... Works fine. Thanks, it helps me move some users off of another coverage tool onto emma(they just had to have those colored bars)..