wise-2.2.0.ebuild requests a specific release of hmmer (-r1) which creates conflicts with other packages (e.g. qrna) when both are installed. I suspect that the "-r1" should be removed or the "=" should be changed to ">=". The problem arises because hmmer is up to "-r2" now, and if you install it emerge starts to complain about conflicts. I strongly suspect that there should be some sort of a dependecy check confirmation process because one rarely wants "=" with "-rX" -- one presumably always wants ">=" or "<=" or no "-rX". Reproducible: Always Steps to Reproduce: 1. emerge wise; emerge qrna 2. emerge --pretend --update --verbose --deep --newuse world Actual Results: Output should list conflice over hmmer requirements (-r1) vs. (-r2). Expected Results: The ebuilds should be constructed so there are not dependent on other specific package states unless absolutely necessary.
Currently I am getting on ~x86: # emerge -uNDpv world These are the packages that would be merged, in order: Calculating dependencies... done! !!! Multiple package instances within a single package slot have been pulled !!! into the dependency graph, resulting in a slot conflict: sci-biology/hmmer:0 ('ebuild', '/', 'sci-biology/hmmer-2.3.2-r2', 'merge') pulled in by sci-biology/hmmer required by world ~sci-biology/hmmer-2.3.2 required by ('installed', '/', 'sci-biology/embassy-hmmer-2.3.2-r2', 'nomerge') ('installed', '/', 'sci-biology/hmmer-3.0_alpha2', 'nomerge') pulled in by sci-biology/hmmer required by world sci-biology/hmmer required by ('installed', '/', 'sci-biology/qrna-2.0.3c', 'nomerge') [cut] # emerge --unmerge hmmer embassy-hmmer qrna sci-biology/hmmer selected: 3.0_alpha2 protected: none omitted: none sci-biology/embassy-hmmer selected: 2.3.2-r2 protected: none omitted: none sci-biology/qrna selected: 2.0.3c protected: none omitted: none >>> 'Selected' packages are slated for removal. >>> 'Protected' and 'omitted' packages will not be removed. >>> Waiting 5 seconds before starting... >>> (Control-C to abort)... >>> Unmerging in: 5 4 3 2 1 >>> Unmerging sci-biology/hmmer-3.0_alpha2... >>> Unmerging sci-biology/embassy-hmmer-2.3.2-r2... >>> Unmerging sci-biology/qrna-2.0.3c... * GNU info directory index is up-to-date. #
There is an inherent conflict over the hmmer package, in that wise, qrna and embassy-hmmer *all* want it and they *all* seem to specify different versions. Wise wants hmmer-2.3.2-r1. Embassy-hmmer wants hmmer-2.3.2-r3. Qrna seems to be content with any version, including hmmer-3.0_beta2. Unfortunately there don't appear to be any standardized "test" components with any of these packages so it is difficult to know whether the dependency is real or an illusion. We ought to be able to get at least the embassy authors to verify that the embassy packages will work with hmmer-3.0_beta2.
So does anyone actually still know why wise requests this specific ebuild? I cannot find anything anywhere in source or ChangeLog... The change in hmmer between 2.3.2-r1 and 2.3.2-r2 is exactly src_compile() { - ./configure \ + econf \ --host=${CHOST} \ So, I'm going to stick my neck out and change the dependency to ~sci-biology/hmmer-2.3.2. Please file bugs if anything explodes.