kpriv/pkgsrc
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity
148086 commits 166 branches 0 tags 1.5 GiB
Commit graph

548 commits

Author SHA1 Message Date
wiz
4e9930c4bb Match changes on NetBSD: mount /usr/X11R7 in sandbox by default instead of 
/usr/X11R6.
 2009-09-06 19:43:18 +00:00
he
cc977aac3f With the most recent pkg_install, directories are removed when the 
last file in a directory is removed.  This might cause common and
expected-to-be-existing directories in /usr/pkg to be removed, and
a subsequent attempt at installing a file to the now non-existent
directory will instead create a file with the name of the expected
directory.

This will create PLIST errors for the package in question, but also will
cause the erroneously named file to not be removed on package removal.

This can cause cascading bulk build failures for subsequent packages.
To prevent this, after each package is done, check if some of the
common top-level expected-to-be directories are now files, and
remove them and emit an error message if so.

I *think* I already fixed the single package which had the unfortunate
problem of not declaring "include" as an installation directory, but
this should prevent the problem from re-occurring in the future as well.

Thanks to joerg@ for the hint for pulling in the value of PREFIX.
 2009-07-09 12:46:49 +00:00
sketch
d2e20bcaf4 Sync bootstrap file list with reality. On the plus side, mk/bulk works again. 2009-07-02 22:34:16 +00:00
sketch
cc9b709684 Make the bulk scripts which use 'set -eu' actually work. 2008-11-19 16:27:14 +00:00
sketch
a61cd42ace Make sure we pick up pax via USE_TOOLS. 2008-06-13 21:52:16 +00:00
sketch
1c77803f61 Mark shells/pdksh as broken so it doesn't get removed part-way through the 
build and kill bmake.  This isn't really a great solution as it means it
won't get built on any platform, maybe we should just make pdksh a bmake
dependancy on every OS.
 2008-06-10 08:17:47 +00:00
sketch
18dffc798c For some reason, devel/bmake now requires shells/pdksh on Solaris. Update 
the save-files list to accommodate.
 2008-06-09 12:51:41 +00:00
he
a8e2184877 Commit two fixes which at least for me appear to be required when 
doing an old-style bulk build on 4.0 or older systems:

 - in mk/bulk/build, do "make update" instead of "make install + clean"
   when installing a new pkg_install
 - in pkgtools/pkg_install/Makefile, don't try to use our own
   executables (${WORKDIR}/pkg_add/pkg_add etc.) if they don't exist

Discussed with joerg, and even though he's not entirely happy
with the latter change, he didn't appear to have a better suggestion.

This, and putting /usr/pkg/sbin earlier than /usr/sbin in the $PATH
appears to be required to get an old-style bulk build going.
 2008-06-08 21:04:30 +00:00
martin
d86d72404e Drop clauses 3 and 4 from TNF licenses 2008-04-29 05:46:08 +00:00
jlam
5c253acc19 + Skip mounting X11 directories if --without-x is specified. 
+ Add a "chroot" option, stolen from pkg_comp(8) that starts a root
  shell with a better environment setup.
 2008-04-01 16:30:19 +00:00
jlam
e0dde0f751 Add missing ] 2008-04-01 16:06:35 +00:00
joerg
1375bd1199 Remove remaining compat code for old pkg_install versions. 2008-03-15 16:27:42 +00:00
jlam
a38826c59e Add a --without-x flag to the mksandbox script. 2008-02-26 05:06:32 +00:00
rillig
66736dde43 Replaced _PKG_SILENT and _PKG_DEBUG with RUN. 2008-02-07 21:36:13 +00:00
rillig
7fca192e2e Replaced all references to pkglint with lintpkgsrc, which has its own 
package since a few days.
 2008-01-04 15:49:07 +00:00
wiz
7512dd9d36 Revert unintended commit. 2007-12-02 12:58:25 +00:00
wiz
6bfb4a5067 Fix typo in error message. From Sergey Svishchev. 2007-12-02 12:57:32 +00:00
rillig
9fab8079e7 Removed _INTERACTIVE_COOKIE, after asking on tech-pkg if anyone needs 
it. It had only been used to change the "nerrors" variable in bulk
builds, but this variable is not used further anyway.
 2007-12-01 11:11:53 +00:00
martti
bd986edbc7 Remove trailing spaces. 2007-10-09 19:19:08 +00:00
kano
abdb54b351 fix obsolete URLs for www.NetBSD.org 
close PR pkg/37071
reviewd by xtraeme@
 2007-10-07 12:59:11 +00:00
adrianp
ccc8f33d94 The new audit-packages code goes to stdout now. 2007-08-27 15:38:08 +00:00
adrianp
d176003e13 audit-packages prints vulnerabilities to stdout so compensate for this. 2007-08-06 02:43:34 +00:00
adrianp
35f7e128c2 Add --config <file> from build script 2007-07-25 12:00:21 +00:00
adrianp
554cf5daf1 Make sure PKG_INFO is set 2007-07-25 10:57:44 +00:00
adrianp
b61d8add4a Define ${PKG_INFO} before using it - from wiz@ 2007-07-21 12:48:31 +00:00
adrianp
9b5c38d79d Add support for the upcoming pkg_install-20070714 which now includes 
audit-packages.
 2007-07-14 17:17:45 +00:00
joerg
409b183fc0 Remove USE_MTREE support. Keep one copy (the NetBSD version) for 
the purpose of print-PLIST in plist/common-dirs.mtree.

Discussed with wiz@, no objections on tech-pkg@.
 2007-07-02 14:54:09 +00:00
wiz
e57a27b67c Fix PR 34139 for bulk builds too. 
Patch from Sergey Svishchev.
 2007-06-15 22:28:53 +00:00
rillig
3a6deb3539 The option --prepare only generates the bulk build database. 2007-06-15 15:52:22 +00:00
rillig
be572cf6c1 If a package is already installed, bulk-install should not try to build 
it again.
 2007-05-31 11:49:09 +00:00
rillig
4a63eaf9f3 When removing an already installed package, the correct package version 
is reported.

Fixed the output by not nestedly redirecting the output of "bmake debug".
 2007-05-23 10:50:44 +00:00
kristerw
c54aa3660b Add all packages that are used by pkgsrc/bootstrap to the list of 
packages that should not be built on non-NetBSD systems.
 2007-05-20 23:33:06 +00:00
wiz
43948ca294 Revert previous (1.140), committed by accident. 2007-05-19 14:10:35 +00:00
wiz
daff9198ed On Linux (at least Red Hat Enterprise 3), openssl needs libraries 
from /usr/kerberos and /dev/random in sandbox.  On FreeBSD, use
devfs to populate /dev in sandbox.

From Sergey Svishchev.
 2007-05-19 14:09:54 +00:00
rillig
6dd050be15 It makes more sense to run "bmake debug" when the WRKDIR is still 
populated with all sorts of files, since they provide the most useful
information.
 2007-04-15 22:14:44 +00:00
rillig
b425a8ed66 The egrep line can be removed when audit-packages becomes fast enough. 2007-03-22 11:43:18 +00:00
rillig
5e2cc53c82 A hint to the real problem is a nice addition to an error message. 2007-03-07 23:32:49 +00:00
rillig
f25aeae6f1 Added the BULK_ID to BULK_BUILD_ID_FILE, so that multiple bulk builds can 
share the same pkgsrc tree in read-write mode. Fixes PR 35939.
 2007-03-07 22:13:38 +00:00
rillig
b1deb83ac9 The first few lines of a bulk build log don't need to be printed on the 
terminal.

The wrapper warnings are included in the build log, no matter if it
succeeds or not.
 2007-03-06 15:54:54 +00:00
rillig
07da582827 Removed the unused "arch" variable. 2007-03-06 10:57:11 +00:00
rillig
ffa8f5f575 The warnings from the wrapper log are added to the report file. 2007-03-06 10:56:34 +00:00
rillig
3d61247309 The post-build-only option also needs to load the variables. 2007-02-23 03:43:53 +00:00
rillig
d1b2c5f7f4 At the very beginning, the variables are extracted by running bmake in 
the pkglint directory. This part had been more time-consuming than
necessary. Now it only needs one call to bmake instead of n.
 2007-02-21 14:25:20 +00:00
jschauma
b9211a69aa - also find possible listings in {whatever}foo variations 
- use PKGVULNDIR from env for correct location of pkg-vulnerabilities
 2007-02-21 05:32:30 +00:00
jschauma
853b673667 Only invoke audit-packages if we do find the package-name prefix in 
the vulnerabilities file.  This cuts down significantly on the time
spent running sort-packages:
On 3228 packages before:
20259.43s real  9888.82s user 13138.60s system

With this change:
 6595.97s real  3215.58s user  4265.13s system
 2007-02-20 06:46:20 +00:00
rillig
56ae0e7a97 The error output of "make debug" may be useful to know, too. 2007-02-19 12:10:43 +00:00
wiz
a9be4d8d01 Remove some legacy variable handling after 2-3 branches. 2007-01-30 07:11:19 +00:00
kristerw
d96a551412 Remove old .bulk-not_available (NOT_AVAILABLE_FILE) files that are left 
from previous run.
 2007-01-28 19:23:22 +00:00
rillig
492c03802f Empty files are so ambiguous, so better leave a friendly message in the 
STARTFILE.
 2007-01-24 02:48:51 +00:00
rillig
2d13644250 Quick fix: The main code of the bulk build cannot be run in the shell's 
"set -e" mode, since it would fill up the temporary directory. Reverted
that change.
 2007-01-24 01:37:20 +00:00