summaryrefslogtreecommitdiff
path: root/usr.sbin/pkg_add
AgeCommit message (Expand)Author
2018-07-09do cleanup after tags, as noticed by aja@Marc Espie
2018-07-09explain what @define-tag mktexlsr supersedes mktexlsr-localMarc Espie
2018-07-08document what happens when the @define-tag defining package gets removedMarc Espie
2018-07-07- move everything tag-related into state->{tags}Marc Espie
2018-07-07what was I thinking. No need to iterate, just use the keyMarc Espie
2018-07-03add "superseded" scaffold.Marc Espie
2018-07-02remove useless Pp;Jason McIntyre
2018-07-02explicitly document some common pkg_add features that also make senseMarc Espie
2018-07-01ahah. complex tag changes show I need to use the same algorithm as for libsMarc Espie
2018-06-30over-eager refactor means fw_update goes thru delete_handle, so teachMarc Espie
2018-06-28fix tag external representation, avoids weird space at end of plistMarc Espie
2018-06-26tags are cutting their first teeth, but it can be a normal debug flagMarc Espie
2018-06-26kill the solve_tags part of pkg_create for now, as it doesn't work at allMarc Espie
2018-06-26make it possible to remove old packages with erroneous tags...Marc Espie
2018-06-25fix ::lookup interface return, oopsMarc Espie
2018-06-25so tags need some love to print out correctlyMarc Espie
2018-06-25OO version of atendMarc Espie
2018-06-25special case: @tag found in self may have NO parameters and just be usedMarc Espie
2018-06-24tags should work more like other errors. In particular, avoid infiniteMarc Espie
2018-06-24document that %D is localbase in define-tagMarc Espie
2018-06-24get ready to reuse verify_tagMarc Espie
2018-06-24implement tag checking during normal solvingMarc Espie
2018-06-24fix need_paramsMarc Espie
2018-06-24fw_update not fwupdate, from weerd@Stuart Henderson
2018-06-24styleMarc Espie
2018-06-23define-tag can't use normal destate, because it migrates before any cwdMarc Espie
2018-06-23solve_tags not yet functional, I'll reactivate the checks during creationMarc Espie
2018-06-23well, when we read the plist from the ports tree, we need to registerMarc Espie
2018-06-23prelink only also needs define-tag, unbreaks pkg_create with run-dependsMarc Espie
2018-06-23PkgCreate has no business knowing details of define-tagMarc Espie
2018-06-23do some more basic checking of tags vs define-tag so that parameter usageMarc Espie
2018-06-23document @define-tag and @tag.Marc Espie
2018-06-23have a default tag value of empty, because.Marc Espie
2018-06-22move the tag base code into SolverBase, and use it in PkgCreate, so thatMarc Espie
2018-06-22as discussed with aja@, some commands must be run once for every distinctMarc Espie
2018-06-22activate tag solver (won't do anything if plists don't have actual tags)Marc Espie
2018-06-22in the presence of multiple tags, tag definition must be "unique"Marc Espie
2018-06-22hook atend tag runningMarc Espie
2018-06-22special case for a tag in self, which should be run appropriately duringMarc Espie
2018-06-22let the tag logic migrate into state for running atend,Marc Espie
2018-06-21split the common half of dependencies handling into SolverBase, as PkgCreateMarc Espie
2018-06-20make it possible to do @tag sthg in the same package that doesMarc Espie
2018-06-20so tags will need to match OLD tag definitions as well as new ones.Marc Espie
2018-06-20tags from changed_handles will be relevant, from nowhere else...Marc Espie
2018-06-20use delete_handle here as wellMarc Espie
2018-06-20modernize pkg_delete code, we have DeleteSets, which means handleMarc Espie
2018-06-20move the solver autoloader into UpdateSet so that pkg_delete will beMarc Espie
2018-06-20prepare to split half of the dependencies code into a manageable subclassMarc Espie
2018-06-19Pass the set thru to deletion operations (we may need to solve_tags!)Marc Espie
2018-06-19fix solve_tags to deal with the actual list of tags (match tag to tagMarc Espie