summaryrefslogtreecommitdiffstats
Commit message (Collapse)AuthorAgeFilesLines
...
* 2.84.0_beta1 releaseRobby Workman2018-12-022-2/+3
| | | | | | | | | | | | | | | | | | | | | | | | - Allow blacklisting of individual packages without collateral damage (e.g. glibc ---> glibc-*). This changes the prior behavior of the blacklist function; previously, adding "glibc" to the blacklist would cause glibc, glibc-profile, glibc-zoneinfo, et al to be ignored by slackpkg. The new behavior is that *only* the glibc package is ignored. If you want to blacklist all packages whose names begin with glibc, you would need to add "glibc.*" to the blacklist now. Also note that any special characters, e.g. "+", will need to be escaped in the blacklist file. (David Woodfall) - Add support for listing .new files without PAGER (David Woodfall) - Remove switch.ch mirrors - Add config option to allow *not* saving .orig configs (Darren Austin) - Mention possible stale mirror if CHECKSUMS.md5 gpg verify fails - Clarify that a press of "Enter" is needed to confirm kernel change (Mario Preksavec) - doinst.sh Don't remove ChangeLog.txt upon upgrade/reinstall. (Patrick Volkerding) - mirrors-x86*.sample: Remove bjtu.edu.cn mirror - Fix for /var/log/packages/ possibly being a symlink to elsewhere - Use CHECKSUMS.md5.asc to determine ChangeLog newness (Patrick Volkerding)
* Note requirement to escape special characters in blacklistRobby Workman2018-12-022-1/+6
|
* Update sample blacklist fileRobby Workman2018-11-221-22/+18
|
* Fix (hopefully) blacklisting of specific packagesDave Woodfall2018-11-222-2/+13
| | | | | | | | | | | | | | In short, it seems impossible to blacklist some packages without catching other undesirable packages. As an example, blacklisting glibc will also catch glibc-zoneinfo (and the other glibc-*) packages. It would be good to have the ability to blacklist only the glibc package without it catching the others. With this change, if sysadmin wants the blacklist to be "greedy," then adding "glibc.*" to the blacklist will do that. Reported-by: Peter Hyman <pete@peterhyman.com> Signed-off-by: Robby Workman <rworkman@slackware.com>
* Minor tweaks to default slackpkg.confRobby Workman2018-11-221-14/+13
|
* Minor grammatical fixupRobby Workman2018-11-201-1/+1
| | | | Thanks to Gerard Lally on LQ.
* Adjust number of .new lines shown based on terminal sizeDave Woodfall2018-11-201-3/+7
| | | | Signed-off-by: Robby Workman <rworkman@slackware.com>
* Update ChangeLog with changes since 2.83.0Robby Workman2018-11-181-0/+15
|
* Add support for listing .new files without PAGERDave Woodfall2018-11-181-7/+33
| | | | Signed-off-by: Robby Workman <rworkman@slackware.com>
* Remove switch.ch mirrorsRobby Workman2018-11-172-12/+0
|
* Add config option to allow *not* saving .orig configsDarren Austin2018-09-194-6/+34
| | | | | | | | | | | | | | When admins choose to allow slackpkg new-config to overwrite currently installed config files with the .new counterparts from an upgraded package, slackpkg has traditionally created backups of the config files first, with the backups being written out with .orig extensions. This commit does not change the default behavior, but it does allow it to be turned off. TLDR: set "ORIG_BACKUPS" to "off" in /etc/slackpkg/slackpkg.conf if you do not wish to have .orig files left over Signed-off-by: Robby Workman <rworkman@slackware.com>
* Mention possible stale mirror if CHECKSUMS.md5 gpg verify failsslackpkg-2.83.0-noarch-4.txzRobby Workman2018-09-051-1/+4
| | | | Signed-off-by: Robby Workman <rworkman@slackware.com>
* Clarify that a press of "Enter" is needed to confirm kernel changeMario Preksavec2018-06-201-1/+1
| | | | Signed-off-by: Robby Workman <rworkman@slackware.com>
* doinst.sh Don't remove ChangeLog.txt upon upgrade/reinstall.Patrick J. Volkerding2018-06-141-1/+0
| | | | | | We're going to stop doing this and see if anything breaks :-) Signed-off-by: Robby Workman <rworkman@slackware.com>
* mirrors-x86*.sample: Remove bjtu.edu.cn mirrorRobby Workman2018-06-122-2/+0
| | | | Thanks to Lockywolf on LQ for the report.
* Fix for /var/log/packages/ possibly being a symlink to elsewhereRobby Workman2018-06-072-7/+7
|
* Use CHECKSUMS.md5.asc to determine ChangeLog newnessPatrick J. Volkerding2018-06-071-0/+27
| | | | | | Thanks to drgibbon on LQ for the suggestion. Signed-off-by: Robby Workman <rworkman@slackware.com>
* 2.83.0 release2.83.0Robby Workman2018-05-291-0/+5
|
* Bugfixes in SlackBuild scriptPatrick J. Volkerding2018-05-291-6/+6
| | | | Signed-off-by: Robby Workman <rworkman@slackware.com>
* slackpkg.SlackBuild: tweaks to support PRINT_PACKAGE_NAME et alRobby Workman2018-05-241-3/+12
|
* slack-desc: Sync volkerdi's changes in -currentRobby Workman2018-04-291-11/+11
|
* 2.83.0beta2Robby Workman2018-04-042-1/+3
|
* post-functions.sh: Tweak handling of new kernel promptsRobby Workman2018-04-042-5/+5
|
* files/slackpkg.8: mandoc lint fixesorbea2018-04-032-55/+58
|
* 2.83.0beta1Robby Workman2018-04-032-1/+13
|
* post-functions.sh: Fix stray stuff in vimdiff promptRobby Workman2018-04-031-1/+1
|
* post-functions.sh: Sort .new files list from looknew()Robby Workman2018-04-031-1/+1
| | | | Thanks to Eduard Rozenberg.
* post-functions.sh: Make sure user confirms kernel updateRobby Workman2018-04-031-6/+6
|
* post-functions.sh: Respect $ROOT in looknew()Robby Workman2018-04-031-1/+1
| | | | Thanks again to burdi01 on LQ.
* post-functions.sh: Don't offer to run lilo if initrd is in lilo.confRobby Workman2018-01-071-1/+8
|
* Make sure user knows that the warning about -current only shows onceRobby Workman2018-01-061-0/+5
|
* 2.82.3beta5Robby Workman2018-01-062-2/+17
|
* files/slackpkg: SOURCE handling - land softer if no trailing /Didier Spaier2018-01-061-2/+15
|
* core-functions.sh: Warn (first time) if -current mirror is selectedDidier Spaier2018-01-061-7/+25
| | | | Require confirmation the first time a -current mirror is selected.
* post-functions.sh: Invert test for "not no" to "is yes"Robby Workman2018-01-061-1/+1
| | | | Essentially, don't process a "not yes" as an "is yes"
* post-functions.sh: check for /etc/lilo.conf too for lilo offerRobby Workman2018-01-061-4/+4
| | | | | | | | Existence of /etc/lilo.conf coupled with executable /sbin/lilo probably indicates lilo usage on the system, so that seems like a reasonable compromise for everyone fussing about slackpkg's offer to run /sbin/lilo for them after kernel upgrades even though they use grub or whatever else.
* slack-desc: Minor tweaksRobby Workman2018-01-041-4/+4
|
* 2.82.3beta4Robby Workman2018-01-042-1/+5
|
* slackpkg: Abort if system date is near epochRobby Workman2018-01-031-0/+17
| | | | This merges slackpkg-quit-on-epoch.patch from Slackware ARM
* 2.82.3beta3Robby Workman2018-01-022-1/+9
|
* pkglist.awk: Handle the (hopefully rare) case of $ROOT usageRobby Workman2018-01-021-1/+1
|
* Add $ROOT to $WORKDIR, $TEMP, and $TMPDIR instancesRobby Workman2018-01-022-34/+35
| | | | | | | | This *should* have no effect if $ROOT is blank Files modified: files/core-functions.sh files/slackpkg
* core-functions.sh: Fix Slackware ARM gpg keyRobby Workman2018-01-021-1/+1
|
* mirrors-arm.sample: Update mirror listRobby Workman2017-12-311-0/+2
|
* core-functions.sh: Create $WORKDIR before using itRobby Workman2017-12-301-0/+3
| | | | | | This should never be an issue on a properly installed system (assuming slackpkg is installed on the system), but it doesn't hurt anything either to be safe.
* slackpkg.SlackBuild: Make VERSION able to be overriddenRobby Workman2017-12-301-1/+1
|
* 2.82.3beta2Robby Workman2017-11-082-1/+7
|
* post-functions.sh: Fix vimdiff patchDavid Allen2017-11-081-1/+1
| | | | | | | | ** This was my mistake; my mail client munged the patch, so I had to apply it all by hand. The extra (old) line of 'showmenu' options was what I'd copied to compare the new options, and I forgot to delete it when committing. Mea culpa. --rworkman
* Increment version to 2.82.3beta1Robby Workman2017-11-072-1/+7
|
* Build script: Fixup expectations for gzipped filesRobby Workman2017-11-071-32/+26
| | | | | The files in the git repo are *not* gzipped because this makes tracking changes, well, impossible.