Changes in 3.0 include:
* The www-exp module has been renamed webui, and a new web frontend has
been added called paefchen.
* All Tinderbox commands with the exception of tinderd are now performed
using the tc utility.
* The Hosts system has been removed as it was not really used, and not
really suited to the purpose of distributed builds. The ground work has
been laid to support distributed builds, but that is not present in 3.0.
* A DISTFILE_URI config directive has been introduced to allow for
distributed distfile caches.
* The latest port fail reasons and patterns have been synced from
pointyhat.
* The total size taken by a port build (in KB) is now tracked in the
database. This can be obtained using the command ``tc getPortTotalSize
-d DIRECTORY -b BUILD''.
* Port build logs now contain OPTIONS and environment details.
* The ``tc addPort'' command assumes a recursive add by default. This
can be disabled by passing the -R argument.
* Port dependencies are now recorded in the database when ``tc addPort''
is called without -R. The dependency list can be retrieved with ``tc
getDependenciesForPort -d DIRECTORY -b BUILD''.
* The ``tc tbcleanup'' command now supports pruning stale distfiles,
stale packages, and not pruning old error logs. See the README for
command line details.
* The default update type and update host are configurable at Setup and
Upgrade time. The default update type has been changed from CVSUP to
CSUP.
* The last_fail_reason is now exported to the postPortBuild Hook.
* Fix the build time display if a tinderbuild takes longer than 24 hours.
* Add a new command, ``tc rescanPorts'' which updates the properties of
all ports in the datastore (including dependencies).
* Display the remake count in the webui frontends.
* Hack around a potential perl-related leftover false positive.
* A new -O option has been added to addPort/rescanPorts which enables
OPTIONS, but does not run rmconfig before calling config. This way,
existing OPTIONS can be preserved. The default behavior is to call
rmconfig to maintain POLA.
* The webui index page has been enhanced to make it more readable, and
more powerful.
* The number of ports not built due to dependency failures has been
added to the webui's main page.
* tc processLog -v now prints the matching text to aid with false
positive identification.
PLUS:
* In webui:
- Change tinderd queue view. Field for adding new ports is the first now.
- Show target port of current build.
- miwi's RSS module
- an other patch from miwi for inc_tinderbox.php.dist which should make
setting-up the webui easier on non-standard configurations.
* Bug-fix: Add o and O to the list of valid options for addPort.
The new code will silently ignore MASTER_SITE_GOOGLE_CODE and
MASTER_SITE_DEBIAN_POOL unless specifically asked to sort them, in which
case an error message is printed.
Most of the patch came from Ryan Steinmetz <rpsfa@rit.edu> with small
additions by me.
* Add a check for duplicate CATEGORIES [1]
* Be stricter when checking for --mandir and --infodir [2]
* Properly check *_DEPENDS both before and after bsd.port.pre.mk [3]
PR: 125962 [2]
126681 [3]
Requested by: kris [1]
Submitted by: Eygene Ryabinkin <rea-fbsd@codelabs.ru> [3]
Renato Botelho <garga@FreeBSD.org> [2]
Specifically, newer autoconf (> 2.13) has different semantic of the
configure target. In short, one should use --build=CONFIGURE_TARGET
instead of CONFIGURE_TARGET directly. Otherwise, you will get a warning
and the old semantic may be removed in later autoconf releases.
To workaround this issue, many ports hack the CONFIGURE_TARGET variable
so that it contains the ``--build='' prefix.
To solve this issue, under the fact that some ports still have
configure script generated by the old autoconf, we use runtime detection
in the do-configure target so that the proper argument can be used.
Changes to Mk/*:
- Add runtime detection magic in bsd.port.mk
- Remove CONFIGURE_TARGET hack in various bsd.*.mk
- USE_GNOME=gnometarget is now an no-op
Changes to individual ports, other than removing the CONFIGURE_TARGET hack:
= pkg-plist changed (due to the ugly CONFIGURE_TARGET prefix in * executables)
- comms/gnuradio
- science/abinit
- science/elmer-fem
- science/elmer-matc
- science/elmer-meshgen2d
- science/elmerfront
- science/elmerpost
= use x86_64 as ARCH
- devel/g-wrap
= other changes
- print/magicfilter
GNU_CONFIGURE -> HAS_CONFIGURE since it's not generated by autoconf
Total # of ports modified: 1,027
Total # of ports affected: ~7,000 (set GNU_CONFIGURE to yes)
PR: 126524 (obsoletes 52917)
Submitted by: rafan
Tested on: two pointyhat 7-amd64 exp runs (by pav)
Approved by: portmgr (pav)
============
1. You can now specify multiple globs for directories in /var/db/pkg on the
command line (along with full names of directories/paths from /usr/ports
and/or /var/db/pkg, as before)
2. When calling check_interactive or update_port from check_for_updates
tell the user what updated version is available. [1]
Bug Fixes
=========
1. Don't set show_list to "all" for -a unless there is somthing to show
2. Add K (don't postclean) to getopts [2]
Other Improvements
==================
1. In check_dependency_files() don't let ro_opd be empty for the purpose
of the egrep expression. It works now, but it apparently violates the
spec and could case problems down the road (e.g., for bsdgrep).
2. Since the BSDPAN pseudo-ports do not exist in the ports tree there
is nothing for portmaster to do with them. Therefore ignore them when
appropriate, notifying the user when necessary. [3]
Submitted by: Geraint Edwards <gedge@yadn.org> [1]
PR: ports/126006 [2]
Submitted by: ahze [2]
Pointy hat: moi [2]
Problem pointed out by: Bapt <baptiste.daroussin@gmail.com> [3]
PR: ports/125428 [3]
The example in the description of pkg_tree shows old package
names (imake-4, XFree86-libraries).
PR: ports/125990
Submitted by: Carlos Santos <unixmania@gmail.com>
recently, parsing the files /usr/ports/MOVED and /usr/ports/UPDATING. MOVED had
an extra blank line on the end for awhile, and the case of some of the UPDATING
lines have changed (e.g., Author: instead of AUTHOR:). This patch allows
portupdate-scan to tolerate these variances.
PR: ports/125840
Submitted by: Alex Stangl <alex@stangl.us> (maintainer)
* Add some USE_GCC checks [1]
* Check to make sure pkg-descr lines do not end with CRLF [2]
* Warn against using != in assignments [2]
* Check for a valid WWW syntax in pkg-descr [2]
Requested by: gerald [1]
kris [2]
=========
1. Fix -a mode by moving the update of CONFIG_SEEN_LIST to after the port
has been checked for available updates. The old way worked as a side
effect to one of the things I fixed in version 2.4. This is probably
how it should have been done all along, but since the old way worked
I was hesitant to change it.
2. Now that we are using a different format for the INSTALLED_LIST,
fix the bit that always displays the list if we are using -a.
=================
Print a list of ports that will be worked on when doing multiple
ports, either from a glob or listed on the command line.
Bug Fixes
=========
1. When working multiple ports, if a port that is specified on the
command line is reached as a dependency for another port that is on
the command line (either specifically or from a glob) that port
should be rebuilt the first time through, even if it is up to date.
2. By the same token, only rebuild that dependent port once if it
was not up to date. Implement these features by keeping lists of
what needs to be, and has been built specific to the multiport
context.
3. In multiport() declare local variables to be local
4. When reading a glob pattern from the command line do not run
the multiport feature if there is only one match, just pass it on.
Minor Enhancements
==================
1. Save the list of work done using \t and \n and print with echo -e
2. Now that we are properly updating CONFIG_SEEN_LIST at the end of
the config run remove the old update to that list from
dependency_check()
