Commit graph

174 commits

Author SHA1 Message Date
wiz
d766856cda Add FILEBASE support for choosing the file name for a binary package.
From Aleksey Cheusov in relation to PR 44698.
2011-03-19 09:45:24 +00:00
reed
e42ac49fb2 Better match library names to improve REQUIRES/PROVIDES.
This is from cheusov via PR #42907.
(I have been using this for months, as I had noticed some of the same
problems.)
2010-12-02 11:15:10 +00:00
chs
bb3b230a5b add EMULDIR and EMULSUBDIR to MESSAGE_SUBST list. 2010-09-24 01:30:54 +00:00
gdt
0060a6312b Use new "pkg-add -U -D" for make replace in DESTDIR case.
Removes overbroad use of -f to override the
depending-packages-have-satisfied-dependencies check, replacing with
the narrow override of -D.

Discussed on tech-pkg@.
OK pkgsrc-pmc@.
2010-09-14 22:30:42 +00:00
gdt
b4484b0d85 In "make replace" case with DESTDIR other than no, require 20100914
for -D flag in "pkg_add -U -D".

Discussed on tech-pkg@.
OK pkgsrc-pmc@.
2010-09-14 22:29:07 +00:00
gdt
d0d03a2d17 Adjust workaround for make replace in destdir case. Use of -f is of
course a too-large hammer, but in addition to overriding checks it
appears to change behavior in some cases when no overrides are
necessary.  Therefore, use pkg_add -U as before first, and only try -f
if that fails.  (This is temporary and should be replaced by -D to
omit only the exact depends check as soon as that's in tree.)
2010-06-16 15:15:05 +00:00
gdt
7ce50f298b Repair regression in "make replace" in DESTDIR case.
"make replace" is defined to replace a package with a newer version,
and update depdending packages to depend on the new version.  It has
long been understood that this is not always safe, with the responses
being "tell people to be careful" and the unsafe_depends variable
scheme and pkg_rolling-replace.  In the DESTDIR case, make replace is
implemented by pkg_add -U.  Usually, this is fine - even if the
ABI/shlib majors have changed, the package is replaced, and then a
later make replace of unsafe_depends=YES packages, either manually or
via pkg_rolling-replace, will bring the system to where it should be.
However, there are pinned dependencies on osabi where the depending
package will not accept the new version, and that causes pkg_add -U to
fail.  This is incorrect, as a) those packages don't depend on the
osabi exact version any more than packages depending on jpeg depeend
on the particular shlib major, yet jpeg dependencies aren't pinned.
And, osabi changing version is not necessarily an ABI change -
consider 5.0_STABLE just before 5.1RC1 and just after, where only the
version string changed.

Therefore, add -f to pkg_add -U so that the update will succeed.
2010-06-12 00:53:43 +00:00
reed
8799ab98f0 From Aleksey Cheusov
http://mail-index.netbsd.org/tech-pkg/2010/03/05/msg005088.html
http://mail-index.netbsd.org/tech-pkg/2010/03/05/msg005090.html

This fixes problem where some packages didn't list correct
REQUIRES or PROVIDES.
Symlinks are removed from PROVIDES.
Libraries provided by same package are removed from REQUIRES.

Also see pkg PR #42907
2010-05-03 16:46:36 +00:00
joerg
84f968d786 Use the real root account for MESSAGES, not the fake user for
unprivileged builds.
2010-02-19 19:11:08 +00:00
joerg
23b2e199d6 Garbage collect support for old pkg_install versions. 2010-02-19 14:27:36 +00:00
joerg
0d6fef1741 Start removal of PKG_SUFX as option by making selection of compression
format a separate new option (PKG_COMPRESSION).
2010-02-09 23:02:13 +00:00
joerg
6582b66b8c Revamp the replace logic for the destdir case. Don't bother messing up
the +CONTENTS files, just set the various variables on the dependencies.
Tested by tnn@
2010-01-30 21:07:29 +00:00
joerg
13557d430c Use new pkg_add -U for replace in the destdir case. 2010-01-30 20:13:19 +00:00
joerg
150c72bb14 Don't bother with pkg_create -L. @src is essentially unused and the
combination of -p and -I provides all the information needed.
2009-11-05 15:39:42 +00:00
asau
3e1ef65ec4 Avoid echoing commands at bin-install. 2009-10-31 12:22:59 +00:00
obache
6d0f1f47ee Fix typo in comment. 2009-10-02 12:35:54 +00:00
joerg
5386a38a9d Compute +SIZE_ALL as documented in pkg_info(1). Addresses PR 41767. 2009-07-24 06:54:37 +00:00
agc
7b77ea0de4 Apply the fixes from Jukka Salmi in pkg PR/41621 to fix smart message
display.
2009-07-17 23:24:57 +00:00
joerg
5fc98f11b5 Make sure that _UPDATE_RUNNING=YES is passed down to su-deinstall for
PKG_PRESERVE packages (e.g. pkg_install). Otherwise an old version of
pkg_install is not removed from pkgdb.
2009-06-17 12:54:57 +00:00
joerg
b6a759d181 Bump PKGTOOLS_REQD version:
- @dirrm is now obsolete and should most conditional mkdir/rmdir lines
  in PLIST should be replaced by @pkgdir
- license handling can use pkg_admin
2009-06-14 15:12:56 +00:00
joerg
4e1bd80afd Add support for using USE_DESTDIR=yes and replace together.
Based on the patch send to tech-pkg in 2008, but less intrusive.
2009-06-09 08:40:27 +00:00
joerg
6437125994 Remove preserve packages for make update with -N -f -f and therefore do
not depend on the behavior of older pkg_delete to allow the removal with
simple force.
2009-05-16 01:34:22 +00:00
reed
93688a0e21 Add new metadata: PREV_PKGPATH and SUPERSEDES.
These will be available from pkg_summary(5) database.
To be used to find previous packages when a package is moved
or renamed. (Even though that is not recommended.)

Documented in pkgsrc Guide and in the pkg_summary(5) manual page.

Upcoming commit will have some packages modified to add them.
2009-05-02 16:14:37 +00:00
joerg
7b5175ce9a Add support for @pkgdir for non-destdir installation. 2009-04-24 14:02:41 +00:00
joerg
ed1f7dd692 Tell the user the right config file and option for vulnerabilities based
on the pkg_install version.  Reported by Nicolas Joly.
2009-04-04 19:07:58 +00:00
joerg
7cb919b410 Skip vulnerability check for packages that set NO_PKGTOOLS_REQD_CHECK.
Set AUDIT_PACKAGES and friends correctly though. Fixes bootstrap on
platforms with pdksh.
2009-03-20 16:52:40 +00:00
rillig
b1bf3ebab8 Replaced _PKG_SILENT and _PKG_DEBUG with RUN. 2009-03-17 22:13:36 +00:00
joerg
8601a121bb package-install with DESTDIR should pass down -A as needed to get
automatic installed handling correct.
2009-02-13 11:30:07 +00:00
joerg
2fcbca4813 Use pkg_admin for audit processing if pkg_install >= 20090201 is
installed.
2009-02-11 14:34:36 +00:00
joerg
4be66048ad Fix circular dependency detection for the case of non-leading circles.
For the dependency chain A->B->B the old code created:
_PKGSRC_DEPS=", B-1.0, A-1.0", which obviously can't match the PKGNAME
of B.
Skip the _flavor-bootstrap-depends processing if PKG_FAIL_REASON is
already set and move the check for circular dependencies right after the
block that defines PKGNAME and friends.
2009-02-06 15:27:00 +00:00
rillig
57fbd43ec4 Replaced _PKG_SILENT and _PKG_DEBUG with RUN. 2008-11-29 13:54:45 +00:00
tron
06c20fecf5 Fix broken shell code:
You must *not* use "test ... && ..." when you use "set -e". Because if the
first expression fails your shell script will abort.

This should fix problem with NetBSD-current's (correctly behaving)
"/bin/sh" report on current-users by David Holland and Kurt Schreiner.
2008-05-26 14:21:43 +00:00
abs
081369412a Include OWNER (if set) in BUILD_INFO - from dillo 2008-05-21 20:39:52 +00:00
joerg
eed07ff045 Simplify logic as audit-packages and download-vulnerabilities always
comes from pkg_install now.
2008-04-07 13:18:25 +00:00
joerg
5dae2194ab Sync with reality: audit-packages does exist, the vulnerability file
itself might not.
2008-04-07 13:16:26 +00:00
adrianp
8e7149e8cb Remove old comment - this now does call audit-packages directly 2008-04-05 16:59:10 +00:00
joerg
caa9b2eebf Move handling of pkg_install version into flavor. Add an explicit
phase pkg_install-depends before bootstrap-depends that just tries to
install a new pkg_install if the current version is too old. Still
keep it as bootstrap dependency for the bulk build code.
For NetBSD, PKG_TOOLS_BIN has to be computed in shell code due to a make
bug.

OK: jlam@
2008-04-03 14:07:51 +00:00
agc
4a96928ff3 Fix from David Holland for PR 38134 - fix a typo which was causing
problems which manifested in "make update" failing.
2008-03-20 20:02:29 +00:00
joerg
d69e33e15f Remove support for separate audit-packages, PKGTOOLS_REQD has been
bumped in the mean time.
2008-03-15 16:12:27 +00:00
joerg
a783161f5f Explicitly "" around $1, so that e.g. csh-like expansion of {} doesn't
happen.
2008-03-11 17:44:03 +00:00
joerg
053f4aa0ea Replace "pkg_admin -S lsbest" usage with pkg_info -E. The base stripping
in bsd.buildlink3.mk was broken with pkg_install-20080309 was it
returned a relative path. It would have failed before e.g. with symbolic
links in the path. pkg_info -E is simpler and was added exactly for this
purpose. Fixes PR 38213 and PR 38211.
2008-03-10 20:05:59 +00:00
obache
0491e08655 Move show-depends-options target from bsd.options.mk to flavor/pkg/utility.mk
to work with option-less packages.
2008-02-18 14:30:40 +00:00
gdt
8c1132d309 Add better error message when failing to tar up old package. 2008-01-23 14:48:50 +00:00
rillig
fe783d7116 Don't run pkg_create with the -v option. It prints a temporary package
name which isn't correct, and the "Value of SrcDir" that is used is not
important to any pkgsrc user. Instead, let the pkgsrc infrastructure
print the package name.
2008-01-23 14:07:07 +00:00
rillig
f185f4e34c Explicitly record the PKGNAME via the @name command. That way, it is
possible to create the package file using a temporary file first, and if
everything has succeeded, to rename it to the real name. This time, I
tested it creating various binary packages and installing them
afterwards, so I'm pretty sure it works now.
2008-01-05 22:06:20 +00:00
rillig
c564dfd5db Reverted the change that tried to make binary packages more sane because
it had severe consequences: pkg_create gets lots of information from the
filename into which the package is written. The extension decides what
compression to apply, and the basename gets recorded as the @name. This
part needs more work.

Noticed by stoned@.
2008-01-04 14:22:06 +00:00
rillig
d3624fdf10 Made the .include directives simpler, since the directory of the
including file is always the first in the search path.
2008-01-04 01:46:24 +00:00
rillig
8df3cdc4f6 When creating the binary package, first create a temporary file, and if
everything went well, rename it to the real name. That way, it is less
likely that broken binary packages are created. It is a common
assumption that binary package files, if they exist, are usable.

An example for a broken binary package is security/sudo-1.6.9p10, in
which sbin/visudo wasn't readable when creating the package as an
unprivileged user.
2008-01-03 23:21:48 +00:00
adrianp
8c8f2165fb * Rename AP var to AUDIT_PACKAGES as the former was a bit obscure
* Add DOWNLOAD_VULN_LIST var
* Get AUDIT_PACKAGES and DOWNLOAD_VULN_LIST to point to the right binaries
  taking into consideration OPSYS and pkg_install version
* Replace all references to audit-packages and download-vulnerability-list
  with AUDIT_PACKAGES and DOWNLOAD_VULN_LIST
2008-01-03 20:51:21 +00:00
adrianp
b919017b3c audit-packages logic for its location depends on a variety of factors
including OS, pkg_install version and NetBSD version.  The following
should pick the correct version to run.
2007-12-16 01:49:08 +00:00