The kdeplasma-addons-4.6.2-optional-akonadi patch adds a condition for adding the #include of akonadi/contact/contactdefaultactions.h as follows: #ifdef LANCELOT_DATAMODELS_HAS_PIMLIBS #include <akonadi/contact/contactdefaultactions.h> #endif but then Akonadi::CotactDefaultAction is used on line 183 without any conditions. I'd submit a patch but I have no idea of what should be the behavior in that case. Any ideas? David E. Narváez Reproducible: Always Steps to Reproduce: 1. emerge --unmerge akonadi 2. emerge kdeplasma-addons Actual Results: Compilation fails because Akonadi was not declared Expected Results: Ebuild should compile fine
I need more info. In your step 1, you mention akonadi (I assume kde-base/akonadi) which is wrong, maybe you mean app-office/akonadi-server? Which package exactly fails? attach a build.log and emerge --info please.
(In reply to comment #1) > I need more info. In your step 1, you mention akonadi (I assume > kde-base/akonadi) which is wrong, maybe you mean app-office/akonadi-server? > Which package exactly fails? attach a build.log and emerge --info please. I just noticed I forgot to mention which was the problematic source file, sorry for that. Well, I should have confirmed step 1 because I'm not sure of how to get LANCELOT_DATAMODELS_HAS_PMLIBS undefined, but the point of the bug is that if that symbol is undefined, line 183 of the file kdeplasma-addons-4.6.2/runners/contacts/contactsrunner.cpp still uses Akonadi::ContactDefaultAction without any conditions. I'd say that's enough info, but anyhow, I'll be attaching the requested information. David E. Narváez
Created attachment 270281 [details] emerge info Adding emerge --info
Created attachment 270283 [details] build.log Here's the build log, grep for the word Akonadi to see the error.
Created attachment 270285 [details] contactsrunner.cpp This is how the contactsrunner.cpp file looks after the patch is applied. Clearly, the use of Akonadi in line 183 is not possible if the LANCELOT_DATAMODELS_HAS_PMLIBS symbol is undefined and should be conditioned somehow.
see comment 9 of the dupe *** This bug has been marked as a duplicate of bug 363585 ***