summaryrefslogtreecommitdiff
path: root/usr.sbin/pkg_add
AgeCommit message (Expand)Author
2006-02-10reset $state->{dir} to the correct value after deleting the replaced package.Marc Espie
2006-02-10Use the ProgressMeter engine to display the update messages. This makesBernd Ahlers
2006-02-10If the length of the message in OpenBSD::ProgressMeter::message() is shorterBernd Ahlers
2006-02-09Turn validate_plist() into a real visitor that calls validate() for eachMarc Espie
2006-02-09use new interface (*close*)Marc Espie
2006-02-09Add some nice interfaces instead of the hint() argument to close().Marc Espie
2006-02-07synchMarc Espie
2006-02-07minor independent tweaks:Marc Espie
2006-02-07look at noexec as wellMarc Espie
2006-02-07Don't push undef into the todo list. This avoids some perl warnings.Bernd Ahlers
2006-02-06tell user about collision much earlier, so that they do not wonder whatMarc Espie
2006-01-30Clarify pkg_info -e example; ok espie@Miod Vallat
2006-01-24protect against weird situations where dependencies don't matchMarc Espie
2006-01-17special-case packages that use old-libs: always update them, even if theMarc Espie
2006-01-13nitMarc Espie
2006-01-12be way more explicit in pointing people to packages-specs(7)Marc Espie
2006-01-09avoid anything that isn't a library in warning for SHARED_LIBS.Marc Espie
2006-01-06tweak the descriptions of -r and -u: make the distinctionJason McIntyre
2005-12-26have pkg_create warn about shared libs which have a version number hardcodedMarc Espie
2005-12-05make sure set_modes always get called for a `real' extraction.Marc Espie
2005-11-29- Zap duplicate semicolon. From Alexander Hall via PR #4913. Thanks!Bernd Ahlers
2005-11-19magically complete fragment with `committed' names, solve the issue thatMarc Espie
2005-11-18restructure some text slightly;Jason McIntyre
2005-11-17finally get around to documenting the alternate call to pkg_create.Marc Espie
2005-11-05same fix as Add.pm: make sure we parse the whole package.Marc Espie
2005-11-05-F conflicts does not really exist, so don't document itNikolay Sturm
2005-11-03kludge: avoid creating non-empty files.Marc Espie
2005-11-03exit with 1 if we can't install shit thanks to dependencies issues.Marc Espie
2005-11-02In case we are pretending, make sure we skip to the next archive member.Marc Espie
2005-11-02Simplify OpenBSD::PackingElement::Sysctl::install().Bernd Ahlers
2005-11-01Add -P option. This shows the pkgpath for each package. Nice for buildingBernd Ahlers
2005-10-28fix error code path for pkg_info -e pouetMarc Espie
2005-10-27Fix a perl warning during package deletion.Bernd Ahlers
2005-10-26add checks in skip for walking off the file.Marc Espie
2005-10-26minor clean-up, unhardcode test for type, use a table instead.Marc Espie
2005-10-25zap 229 (frantisek holop)Marc Espie
2005-10-24more robust deletion: rebuild item from file system instead of usingMarc Espie
2005-10-24use a visitor('prepare_for_deletion') to validate plist for removal,Marc Espie
2005-10-24factor out a bit of code, and be more aware of symlinks.Marc Espie
2005-10-24filter out .libs-* later, so that compute_closures() packages do notMarc Espie
2005-10-24do not try to find a match for .libs-* packages.Marc Espie
2005-10-23if resumed from ^Z, display meter without waiting for changes.Marc Espie
2005-10-22add a hint to close that says how the fd got closed (real error, premature,Marc Espie
2005-10-22always display the list of packages we can't update.Marc Espie
2005-10-22more non-errors from ftp.Marc Espie
2005-10-22I haven't received any feedback at all about this patch, so everyoneMarc Espie
2005-10-22be more explicit about what's going on.Marc Espie
2005-10-19mark links being extracted with `zap', so that they do not get recordedMarc Espie
2005-10-10less verbose listing of librariesMarc Espie
2005-10-10several paths lead to has_new_sig, cache the result.Marc Espie