in current poky ref manual, the explanation of ALLOW_EMPTY talks about using it with a package name, as in:
ALLOW_EMPTY_${PN} = "1" which makes perfect sense, except that there is the occasional example where something other than "PN" is used, sometimes for good reason, sometimes not clear. for instance, there's this: meta/recipes-qt/qt4/qt4.inc:ALLOW_EMPTY_${PN} = "1" meta/recipes-qt/qt4/qt4.inc:ALLOW_EMPTY_${QT_BASE_NAME}-fonts = "1" which makes sense (i guess) since, in this case, PN=qt4-x11-free, while QT_BASE_NAME=qt4. so there are obviously going to be special cases like that. but here are a few that seem to use BPN unnecessarily: meta/recipes-core/eglibc/eglibc-locale.inc:ALLOW_EMPTY_${BPN}-binaries = "1" meta/recipes-core/eglibc/eglibc-locale.inc:ALLOW_EMPTY_${BPN}-charmaps = "1" meta/recipes-core/eglibc/eglibc-locale.inc:ALLOW_EMPTY_${BPN}-gconvs = "1" meta/recipes-core/eglibc/eglibc-locale.inc:ALLOW_EMPTY_${BPN}-localedatas = "1" meta/recipes-graphics/pango/pango.inc:ALLOW_EMPTY_${BPN}-modules = "1" it seems that, with the above packages, PN = BPN so, yes, either one would have worked, but is there a best practice? i would think that using anything other than PN should be reserved for when it's necessary, so that it's more aesthetically obvious that something different is happening. thoughts? yes, i know it makes no functional difference. again, i'm just trying to codify best practices. rday -- ======================================================================== Robert P. J. Day Ottawa, Ontario, CANADA http://crashcourse.ca Twitter: http://twitter.com/rpjday LinkedIn: http://ca.linkedin.com/in/rpjday ======================================================================== _______________________________________________ Openembedded-core mailing list Openembedded-core@lists.openembedded.org http://lists.linuxtogo.org/cgi-bin/mailman/listinfo/openembedded-core