| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
arguments. Repair...
|
|
- mark all regexps I can with /o if they can be compiled once.
- turn $o->method() into $o->method
- remove unneeded prototypes
- reduce split /re/ into split "string" where possible.
|
|
Create a common state class for delete and add, put the progressmeter
creation in there.
neat effect: we don't have a progressmeter, we use a stub class instead
and *never* pull the code in at all.
|
|
really_add).
move common delete_installed() code inside Delete.pm
|
|
|
|
ensuring we don't reload fontdirs twice with weird fontdirs case.
|
|
common to Add and Delete, for now).
Use it to have common code that records items to remove / items remaining.
Stash it into pkg_add/pkg_delete state, and abstract the cleaning up
operation into the SharedItemsRecorder (so that pkg_add and pkg_delete no
longer have to know about the internals of SharedItems at all).
|
|
write shortcuts for common behavior, and start hiding some details.
|
|
|
|
|
|
|
|
much shorter.
|
|
reword the introduction to the stuff originally from FreeBSD to clarify
slightly.
|
|
Fix a really fun bug in pkg_delete: solve_installed_names has to
normalize the list of names so that comparing its size vs. its
transitive closure is meaningful. Use the classical hash + list to
preserve the order. While we're there, use the hash to avoid asking
the same questions again and again.
After some back&forth with bernd. okay bernd@
|
|
is finding some dependency problems, explain which package depend on what.
(with -v, -i or '-F verbosedeps')
idea & ok espie@
|
|
pkg_* bits;
ok espie@
|
|
|
|
* move the signature code to a packing-list method, prepend the pkgname.
* depend on distant listings, cache these systematically.
* cut the PKG_PATH at :/ systematically, much simpler than parsing URLs.
People should get used to trailing /s.
* let pkg_add -u actually run the update.
* check for signature in more places, for instance, let pkg_add installed_stuff
be a void operation.
* grab enough stuff during updates to check signatures, so that pkg_add -u
can report updates that are not needed right away, instead of waiting
for the more confusing `trying to update'
* beginning of @module support.
* @pkgpath support, including updates.
* systematically expand stems when applicable.
* use visitors systematically in pkg_info.
* pkg_info -S reports signature.
* remove hardcoding of categories in PackingList, so that future modules
will be able to expand them.
* add @updateset and @incompatibility keywords.
|
|
|
|
ditch this test, pkg_delete -n should no longer loop indefinitely.
|
|
put an # XXX comment so I won't be tempted to clean it again.
|
|
to pretend cleaning it up).
|
|
design checked with millert@, relying on process termination for unlock
at his suggestion.
Perf optimization: don't try to reinstall installed packages in kitchensink
mode if !-r.
|
|
okay pvalchev@, naddy@, marco@, deraadt@
(keep -f around for the transition, will scrap it soon)
|
|
some dependency got badly adjusted.
|
|
|
|
don't want to load RequiredBy all the time).
Remove a few $state->{not} tests that are no longer needed.
pkg_add -n / pkg_delete -n should work much better when they need to
handle multiple dependencies.
(todo: kill plist caches, finish replacing with specialized caches like
libraries).
|
|
|
|
|
|
|
|
|
|
document the new stuff, partial reorder of options.
|
|
directly. Optimize for !wantarray, by telling whether the list is empty or not.
prepare for a second similar list (REQUIRING) that will replace pkgdep, since
pkg_add -r spends a hell of a lot of its time reading/writing long
packing-lists...
Use the simplified interface.
Remove quite a few tests that the RequiredBy list is non-empty that do
nothing but obfuscate the code: just deal with the damn list already.
|
|
|
|
|
|
change prototype of validate_plist, ditch globals cdrom_only and ftp_only
(stash this into state).
Use Warn properly.
|
|
|
|
use these to simplify all those packing lists manipulations.
demote non-root detection to a warning in -n mode:
fix a couple of minor bugs, of stuff that was run in -n mode and should not.
Namely, manpages were indexed/unindexed (ouch) and tempfiles were creating
during updates.
|
|
don't enable the progress meter if a very verbose mode is on.
|
|
end exclusively now.
|
|
|
|
reruns ldconfig -R (okay, we don't check that -m would be enough, so what ?)
First hooks for -r: check that replacement should work.
|
|
be usable by pkg_add -r.
|
|
|
|
|
|
- It will be used after deletion/adding during updates.
- We might avoid loading it if a package contains no new directories.
Actually a lot of packages don't use it.
|
|
(and use it).
Optimisation: don't clone() directories
|
|
|
