Red Hat Bugzilla – Bug 222191
Merge Review: eclipse - An open, extensible IDE
Last modified: 2015-02-12 04:18:05 EST
I'd like to get the eclipse spec file reviewed in preparation for the core and extras merge. I won't be able to commit this to the extras repository because java-gcj-compat uses the eclipse-ecj sub-package but I would still like to go through the process. Spec URL: http://people.redhat.com/bkonrath/eclipse/eclipse.spec SRPM URL: http://people.redhat.com/bkonrath/eclipse/eclipse-3.2.1-28.fc7.src.rpm Note that this spec file is already in Fedora cvs and I will be committing changes as I go through this review process. Description: The Eclipse Platform is designed for building integrated development environments (IDEs) that can be used to create applications as diverse as web sites, embedded Java(tm) programs, C++ programs, and Enterprise JavaBeans(tm).
I wonder if it would be a conflict of interest if I took this one? Probably not.
only if you dont give it a fair review
I'll give it a fair review. So I guess I can take this one. I'm intimately familiar with the spec so it'll probably be easier for me.
Updated files are located here: http://www.bagu.org/eclipse/eclipse.spec http://www.bagu.org/eclipse/eclipse-3.2.1-30.fc7.src.rpm
Hey Andrew... if you are reviewing this I assume it should be blocking FE- REVIEW instead of FE-NEW. I am going to change it to do so. If I am in error, feel free to change it back.
You are correct about the blocker but Tom's actually going to review it. I've re-assigned to him. Thanks.
The files referenced in comment #0 are out of date. Please let me know when somebody is ready to do the review so that I can post updated files. Thanks, Ben
According to the assignment, Thomas is up for this review. But he hasn't made any comments in this ticket, so I'm not really sure what's up.
I can't speak for Tom, but I'd bet he's completely forgotten about this. If someone else can review, I'll work with them on changes as necessary.
Is this bug still actual ?
(In reply to comment #10) > Is this bug still actual ? Definitely, it is. Nobody made full review so far.
(In reply to comment #11) > Definitely, it is. Nobody made full review so far. Right now we have F11/Eclipse 3.4.2, initial request was for F7/Eclipse 3.2.1. And I can't see spec file actual for today given somewhere in comments. Link to spec file on comment #4 is dead. So I still wonder...
It should perhaps be renamed Merge Review, not Review Request anymore.
what is going on here? But is assigned to rruss but fedora-review ? flad is set by tibbs. If there is no response I'll remove both the assigned and fedora-review flag so someone can take the bug and do the review.
This is assigned to Rodney because it was assigned to Lillian and he was her manager when she left Red Hat. I think it can safely be re-assigned and I'll do so now.
Reset to new.
This bug is definitely not actual: Core and Extras have been merged thousands of years ago, Eclipse is a part of the default Fedora repo. Why should this bug still be open ?
(In reply to comment #18) > This bug is definitely not actual: Core and Extras have been merged > thousands of years ago, Eclipse is a part of the default Fedora repo. Why > should this bug still be open ? Because nobody ever made that review. This is merge review from times of merging Fedora Core and Fedora Extras (that’s Fedora 7) which nobody ever finished. Do you volunteer to do it?
I spoke with Mathieu Bridon (bochecha) on QA IRC channel today. It seems I am not a proper person to perform the review as I have no rights to approve the spec neither am I skilled enough for such a complicated package. We agreed, however, that I will do the 'technical' review - check if all formal 'must-have's are in place and will put it here. Afterwards it will be easier for a real reviewer to finalize the review.
(In reply to comment #20) > I spoke with Mathieu Bridon (bochecha) on QA IRC channel today. It seems I > am not a proper person to perform the review as I have no rights to approve > the spec neither am I skilled enough for such a complicated package. > > We agreed, however, that I will do the 'technical' review - check if all > formal 'must-have's are in place and will put it here. Afterwards it will be > easier for a real reviewer to finalize the review. Right, anybody who can write a comment to bugzilla, can create a review. Even if it is not binding, it will greatly help somebody who has the rights (me, for example) to just run through the review, check it, and make the review swiftly and with less effort spent. Also, it greatly counts towards you abilities (and qualification to be packager with proper rights).
As the maintainer I would be really thankfull if someone does the review. We always wanted to finalize it but it was just not right someone of the current maintainers to do it.
@Alexander Kurtakov: will you please attach actual spec and SPRM to this bug ? Just to follow the formalities of the Review process: https://fedoraproject.org/wiki/Package_Review_Process#Contributor
Andrey, this is not the procedure for merge reviews. See https://fedoraproject.org/wiki/Merge_Reviews (Citation: You should be able to review the package against the current HEAD revision in rawhide git. You can use the current rawhide package to check against. ) The justification behind that is that packages already in fedora are updated while the review is going so no static spec/srpm can be attached.
To make it easier: * Link to git repo http://pkgs.fedoraproject.org/cgit/eclipse.git/ - contains all patches and spec * Link to koji page http://koji.fedoraproject.org/koji/packageinfo?packageID=183 - contains all builds of eclipse
Created attachment 677579 [details] rplmint output rpmlint output attached. Small summary of findings is below. I suggest biggest portion of those should be ignored due to specifics of java packages, but I am at the moment just making a summary for further, more experienced reviewers (some warnings can be really important). 0. eclipse-src 0.1. invalid-url (2 times) 1. eclipse-swt.i686 1.1. unstripped-binary-or-object (6 times) 1.2. no-documentation 1.3. hidden-file-or-dir (3 times) 2. eclipse-equinox-osgi.i686 2.1. only-non-binary-in-usr-lib 2.2. no-documentation 3. eclipse-platform.i686 3.1. unstripped-binary-or-object (2 times) 3.2. non-etc-or-var-file-marked-as-conffile 3.3. conffile-without-noreplace-flag (2 times) 3.4. no-documentation 3.5. dangling-symlink (65 times) 3.6. hidden-file-or-dir (13 times) 3.7. incorrect-fsf-address (6 times) 3.8. non-conffile-in-etc 3.9. zero-length 3.10 no-manual-page-for-binary 4. eclipse-jdt 4.1. only-non-binary-in-usr-lib 4.2. no-documentation 4.3. hidden-file-or-dir 4.4. dangling-symlink (3 times) 4.5. dangling-relative-symlink 4.6. no-manual-page-for-binary 4.7. dangerous-command-in-%pre 5. eclipse-pde 5.1. only-non-binary-in-usr-lib 5.2. no-documentation 5.3. dangling-symlink (2 times) 5.4. hidden-file-or-dir 5.5. no-manual-page-for-binary 5.6. invalid-url 6. eclipse-tests 6.1. no-documentation 6.2. script-without-shebang (2 times) 6.3. dangling-symlink 6.4. no-manual-page-for-binary
Mass reassigning all merge reviews to their component. For more details, see this FESCO ticket: https://fedorahosted.org/fesco/ticket/1269 If you don't know what merge reviews are about, please see: https://fedoraproject.org/wiki/Merge_Reviews How to handle this bug is left to the discretion of the package maintainer.
Eclipse package is going through heavy inspection by current maintainers for every major release (once per year) and there was no interest from anybody stepping up to do official review - thus closing.