grade port: 1
> Error: Unable to execute port: upgrade fontconfig failed
Did you recently upgrade to Mavericks? If so, read
https://trac.macports.org/wiki/MavericksProblems and
https://trac.macports.org/wiki/Migration.
--
Clemens Lang
__
asked to put it back in, especially
because it can be used to install older versions from the binaries
archives we publish on packages.macports.org.
If you run
port notes https://neverpanic.de/documents/mpstats.tar
it'll work just fine.
--
Clemens Lang
branches/gsoc11-statistics/stats-server/db/schema.rb
Sure, I can provide a full dump – I'd rather not mail it to the list,
though, so please contact me off-list if you want a dump.
--
Clemens Lang
___
macports-users mailing list
macp
ut the statistics data (you can generate some for
your own local tests) but just the data from the categories and ports
tables.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
sted committer as a step toward improving the
> system...
What about people who are not committers? Plus, development can be done with
random auto-generated data and there even is a script to generate some in the
source tree.
--
Clemens Lang
___
macports
Hi,
> On 22 Mar 2014, at 20:55 , Clemens Lang wrote:
> > I do not log any post data along with the IP address, so I cannot tie IPs
> > to UUIDs, except maybe by correlating the last-update timestamp and the
> > log file.
>
> We are talking NSA-like meta-data, whic
n on the list you will see that. So, I am not against this great
> feature at all.
Sorry, I didn't go back to re-read the archives on that -- I probably should
have instead of accusing you out of the blue.
So, let's get this matter behind
;t contain any IPs.
> I would like to suggest to eventually (not necessary now) copy or
> move/rename "branches/gsoc11-statistics" to something that makes more
> sense now when GSOC 11 is over and the project lives.
Yeah, that makes sense. Probably somewhere in contrib, where mpa
Hi,
> - Outright negative advertising (“MacPorts driving you to drink?”)
As a quick follow-up:
https://github.com/Homebrew/homebrew.github.io/commit/b0e71dddbfca348a9f19dadf1a5196abe4c9207e
--
Clemens Lang
___
macports-users mailing list
macpo
Hi,
> I have attached the config.log file, any suggestions?
You didn't agree to the Xcode license agreement.
Run sudo xcodebuild -license in a terminal.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.o
Hi,
On April 2, 2014 7:43:01 AM CEST, Horst Simon wrote:
>Is it possible to build;d it with TLS and SASL support?
Yes, it is. See the output of `port variants postfix` and install postfix with
the variants you need.
--
Clemens Lang
___
macpo
andbox hiding /usr/local, and that
should make it a lot easier for MacPorts to work in presence of homebrew. That
doesn't necessarily work the other way round, but it's much easier for homebrew
to ignore stuff installed by MacPorts than vice-versa.
--
Clemens Lang
___
oring at that version and that date?
[1] has the answer for that: "mongodb" is not distributable because its license
"agpl" conflicts with license "OpenSSL" of dependency "openssl"
[1]
https://build.macports.org/builders/buildports-mtln-x86_64/bui
e.
[1] https://trac.macports.org/changeset/119239
[2] https://trac.macports.org/changeset/119240
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
grep '_SSL'
[3] http://trac.macports.org/ticket/43436
[4] http://trac.macports.org/changeset/119239
[5] unless you replaced your version of /usr/lib/lib{crypto,ssl}* with a
vulnerable version manually or have a vulnerable version in /usr/local/lib.
--
Clemens Lang
_
ain five lines with the version number
at 1.0.1g.
HTH,
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
pt/local/sbin/dovecot -F
and checking that it is the one listening on the port you test using
$> lsof -i tcp:$port
You could also try running another SSL-enabled service using the same
OpenSSL version (e.g. apache2 from MacPorts) and check if that is
vulnerable to heartbleed.
--
Clemens
k I can help any further at this point. I'd suggest you take this
upstream to the developers of dovecot2 -- maybe they know something about how
openssl is used in their software that might shed some light on the problem.
--
Clemens Lang
___
macports-
Hi,
> Is there any interest in supporting a "relative" variant for Ruby? The
> portfile edit is trivial, cf below:
If that doesn't add too much runtime overhead or other unforseen
side-effects, why not enable it by default for all installatio
-license
to take care of that.
MacPorts 2.3 will contain a check that will detect this.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
ing in
macports.conf using `sandbox_enable no' to see if that helps.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
a Portfile for GitX-dev. I wanted to do it for a
while,
but I'm apparently not using GitX often enough to care about the update, or I
would have done it by now.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
htt
y?
I've CC'd somebody more knowledgeable on sandboxing than I am. Joshua, any
idea what's wrong here or is this expected behavior?
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
sudo port rev-upgrade. Rev-upgrade will automatically find binaries like
the one that's broken on your system, automatically determine the necessary
actions to fix the problem and execute them.
--
Clemens Lang
___
macports-users mailing l
)
You are using a version of MacPorts that is >= 2.3.0. Please update the
port_binary_distributable.tcl script (the one from the source tree you used to
install MacPorts should be fine).
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
(iirc that was /usr/mactex/bin or something?).
For obvious reasons if MacTeX installs into /usr/local/bin (which I
doubt it does, but just for future reference and similar endeavors)
you should avoid doing that.
--
Clemens Lang
___
macports-users mailing
enDNS servers are probably not
the fastest to ping, though.
Also note that you can disable this behavior in OpenDNS' control panel.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
t with these
> possibilities more easily.
Exactly.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
ails because of the proxy?
The rsync download for the base code will fail, but that won't abort the
selfupdate. The port tree sync will use your configured source, i.e. svn,
which will succeed. You shouldn't have any problems.
HTH,
--
Clemens Lang
___
.log (possibly compressed) or
open a ticket, attach main.log there and post the link to the ticket.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
ron/Software/port-whatsnew.tclsh" line 58)
This problem will occur for all contrib scripts that use MacPorts Tcl with
2.3.0.
I have fixed port-whatsnew in r120482 [1]. Please update to the latest version
from svn and it should start working again.
[1] https:
ports you have, very likely on the
scale of gigabytes). To do that, run sudo port uninstall inactive.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
blemHotlist#xmlwf
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
Hi,
> > Please notify the developers of z88 Aurora about this, if you can.
> Why not make a port for z88 Aurora?
That's a good idea, please contribute a Portfile if you can. I don't
have the time to port it myself at the moment
here? If mkvtoolnix only compiles with GCC, that might be an option, and it's
certainly better than having a binary that always crashes.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosfor
acPorts ruby if it's there and system ruby if it isn't.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
;s the case using
otool -L `which mkvmerge`,
then select a library from your MacPorts prefix in the output of this
command (e.g. one of the boost libraries) and do the same. If one of them
contains libc++.dylib and a different one libstdc++.dylib, that's your
problem (and for us to fix, and w
the case of mkvtoolnix since somebody
already did the hard work of porting extra copies of ICU and boost, but
then again mkvtoolnix doesn't have a maintainer, and that's a job for
somebody who is interested enough in mkvtoolnix to volunteer as
maintainer.
--
Clemens Lang
___
3.0? Does it warn you about missing Command Line Tools?
There's a db.h in /usr/include, installed by the Command Line Tools, which
should be picked up by configure. Since that doesn't happen for you, I assume
the file is missing. Please run xcode-select --instal
se quiet mode: port -q
- Redirect the output of port into a pipe: port … | cat
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
that supports keeping build dependencies.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
istence and/or file modification date on a remote server) do not use this.
As such, I don't think "computing dependencies" can cause this even with the
new code (since it doesn't have any new code), but downloads might.
Nevertheless we
f curl_easy_reset() every ~10 calls:
Might defeat the original purpose of the change, which was enabling re-use
of HTTP connections.
The original question remains, though: Why aren't your connections closed, or
re-used?
--
Clemens Lang
_
dle. It needs
to be an Apple property list file and contain the key CFBundleIconFile at the
top dict level, where the corresponding value is the path of the icon file
relative to the Contents/Resources directory. You can use the app PortGroup to
build it
ermore, there's the depends_skip_archcheck variable that can be used
to achieve this. See src/macports1.0/macports.tcl, line 2157.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
rt given
manually or one of its dependencies.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
uld say where the problem is.
It was a berlios problem, see https://trac.macports.org/ticket/44252.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
on the
grounds that their copy of openssl is system-provided. In that sense, we could
distribute a copy of git that uses Apple's OpenSSL, but not one that uses
MacPorts OpenSSL.
The issue is well-covered, though - just google for openssl gpl.
Remember IANAL and this isn't lega
be fixed
IMO. It's weird that GnuTLS needs OpenSSL to build.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
On July 8, 2014 1:41:14 AM CEST, Jerry wrote:
>
>On Jul 6, 2014, at 7:03 AM, Clemens Lang wrote:
>
>> Hi,
>>
>>> Suggest you look at that file to see what is interfering with your
>network
>>> connection. Wildcard DNS A records and proxies are common c
wser/branches/release_2_3/base/src/port/port-help.tcl
for the message you are looking for. I'm working on getting those manpages
written. Help very welcome.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https:
nt says: an utterly ugly hack. It will
break horribly if upstream suddenly decides to do 100 commits on top of their
current head so the value you list in ${git.branch} is no longer cloned.
This isn't a problem for mplayer2, because there's only a few commits per
ted. I never got that to work correctly – which doesn't
necessarily mean it can't be done, I just didn't bother and used a remote
virt-manager instead.
> By the way does any body know any alternative ?
You should be able to use virsh, the shell of libvirt. It should be
ava
. You'd have to ask Cal.
No. In fact, a recent update fixed that configuration, since it was
previously broken.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
the
purpose of the user. Using a separate unprivileged user here makes sure no
build system can modify (or even read) your files. Your configuration is
insecure and should be avoided.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.m
again,
sandboxing is there for exactly this reason, to reduce the risk of running
make install DESTDIR=$destroot with root privileges. And again, looking at
your requirements makes me think you should really use a non-root installation
of MacPorts.
--
Clemens Lang
___
versions of a port active.
If you have the new version installed but deactivated, the "outdated"
pseudo-port will not contain the port. "port outdated", however, will.
--
Clemens Lang
___
macports-users mailing list
macports-users@
after that and re-install them.
Since we don't have a 10.10 buildbot there are no binary archives and
re-installing from source will take a while. Expect delays.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.maco
commands just print an error message and require the user to more
> laboriously figure out how to manually build from source.
Maybe we should make it possible to run `sudo port -f selfupdate` despite the
incorrect platform?
Or even outright provide a "
any time.
This is usually caused by one of our blacklist entries. In your case, we had a
blacklist entry for yandex.ru because of the amount of spam tickets we were
getting from accounts registered with yandex.ru. I've removed the blacklist
entry and you should now be able to file ti
available open DNS server like
> 8.8.8.8
> should do the trick just as well.
I'd suggest using an address in 127./8 with the exception of 127.0.0.1 instead.
The whole 127./8 is reserved for loopback use.
--
Clemens Lang
___
macports-users ma
struct::list equal [lrange $fileComponents 0 $prefixLen]
$normalizedPrefix} {
set prefixComponents [file split $prefix]
set fileComponents [lreplace 0 $prefixLen-1] $prefixComponents
}
set filename [file join {*}$fileComponents]
--
Clemens Lang
currently already implied for all other paths, because we don't normalize
(which is exactly why symlinking prefix even works at all), but port provides
breaks in this situation, because it does in fact normalize.
Not sure what you mean with `unreadable pathnames in
/opt/local/var/mac
rts_kde_kdepim4-runtime/kdepim4-runtime/work/kdepim-runtime-4.12.5/`
Not what this thread is about, not likely to change, and not an issue IMO.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosfor
+bar +baz will install
A
B +foo
C +bar
D +baz,
but we actually don't know which variants exist in a Portfile until we've run
it. To avoid running a Portfile twice, MacPorts doesn't check whether you
actually specified a variant that exists.
--
Clemens Lang
__
Hi,
please take your conversation off-list. We don't want to be on Lennart's
next list of hostile open source projects.
--
Clemens Lang
(I admit, I couldn't resist that one. But please, go arguing elsewhere.)
___
macports-users mailin
mes, this causes crash recovery and fixes the issue.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
he
end of your installation. For this to work correctly if you're pressing ^C
we're also implementing signal handling, which is taking us a while. If
you'd like to test that, use trunk (but don't complain about breakage by
other new features).
--
Clemens Lang
___
/private/opt, then placed
> a symlink at /opt.
I've suggested a simple addition to port provides in
https://lists.macosforge.org/pipermail/macports-users/2014-October/036559.html
that would fix problems like these where MacPorts' prefix contains a symlink. We
should implemen
installed
ports.
> P.S. Note that the migration instructions may be clearer if they are organized
> in chronological order, i.e., the 'port -qv installed', etc. might come before
> upgrading the OS and reinstalling Xcode.
Sure, you cou
o, if you're, for example, updating boost and
qt4-mac, I'd expect it to take a while.
That being said, I've seen sub-par sqlite performance recently, so maybe you're
hit by that? It usually fixes itself with the next "port" process you start,
though.
--
Clemens Lang
_
ing this also. Good to hear that I'm not imagining it.
I think our SQLite database isn't performing very will in large transactions
(such as this one, or activating boost).
--
Clemens Lang
___
macports-users mailing list
macports-users
crypted FileVault2-CoreStorage setup ever since it was
available and have seen good performance meanwhile.
I still think the issue is somewhere in SQLite and how we use it.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
) as
backup target. In fact, due to OS X' somewhat weird choices on UTF-8
normalization that might in fact not be a bad idea.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
ll all ports for which images_to_archives.tcl
is triggered.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
have failed.
So,
apparently something or somebody deleted this file without removing the metadata
about it in the MacPorts registry, and _that's_ the problem.
So, yes, it is wrong and dangerous to remove these files, since they are, in
fact,
not unused as you noticed.
--
Clemens Lang
en't done the conversion yet. If the
script would have problems with symlinks I think we would have heard about it
earlier.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
one that automatically when it
changed the database schema (and the corresponding handling code) accordingly.
I'm not sure what that hasn't happened for you. Did you at some point jump
between
MacPorts versions, restore the database from a backup or something related?
--
Clemens Lang
__
/src/cregistry/sql.c?marks=333-360#L330
[2]
http://trac.macports.org/browser/branches/release_2_3/base/src/dedup_portfiles.tcl.in
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
ly updated to the MacPorts version that has this change?
Which version are you actually running? What does
$> sqlite3 /opt/local/var/macports/registry/registry.db 'SELECT * FROM
metadata'
print?
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
date_needed` being present and set
to one, which means the conversion didn't run or successfully finish on your
system. You will have to run it again, e.g. by running
sudo port -df selfupdate.
If it fails again, you should at least see some error output.
--
Clemens Lang
something cryptic.
> Anyway, as I said, running it a second time didn't do squat in this case. A
> forced selfupdate was required, apparently.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
uld
>wait for it to complete.
Correction: The process' name in Activity Monitor will be tclsh8.5.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
f the size, is opt/local/share/mystonline
What kind of sandbox issues are you seeing? In general, it should be possible
to symlink the whole /opt/local to somewhere else (at least that's what I
tested the last time I touched the code).
--
Clemens Lang
___
s, this
causes the database software to do a scan across 1000 items 1000 times (once
for each file we're activating). Of course now with boost, which installs some
35k files, that's 35k^2. You see where this is going…
--
Clemens Lang
___
mac
re how SQLite does
things, as long as it's fast.
> I wonder if this also explains the memory exhaustion problems I've seen occur
> during lengthy port upgrades (i.e. many ports at once).
That's very unlikely. I also haven't seen any problems like these.
--
Clemens
orts base somewhere.
It's possible that there is a memory leak in the parts of MacPorts written in
C. A lot
of the code is old and hasn't seen maintenance or re-factoring in a while. I'm
pretty
confident the parts I've looked at (registry, darwintrace, tracelib and curl
parts of
pextlib) are leak-free, but that doesn't mean that other parts are.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
mode.
All of my 500+ ports either built fine, or the problems were already fixed.
Of these ports, about 5 needed an Xcode installation and thus failed. Those
were, for example, osxfuse, textmate2, wireshark and a few others.
1 didn't build in trace mode, but that was a known
t using
it to automatically ignore wrong resolve results. This has been discussed
before and there were some strong opinions against fixing this behavior
which isn't our bug and for telling users to shout at their providers for
ignoring the RFC.
--
Clemens Lang
__
guration
are definitely not the right flags - something went wrong at the configure
level.
> Is there a way to skip using those bundled packages (I also have tcl-thread
> installed, same version as the one bundled)?
No, and it's not supported.
--
Clemens Lang
___
using an incompatible Tcl version (8.4 and
8.6 don't work)
3. It wouldn't work because the MacPorts Tcl packages wouldn't be installed
in a location automatically found by your Tcl interpreter when you try to
run package require.
--
Clemens Lang
__
Hi,
- On 28 Nov, 2014, at 12:08, René J.V. Bertin rjvber...@gmail.com wrote:
>> > What would be best to troubleshoot, the terminal output from configure, or
>> > config.log ?
>>
>> Configure output and config.log.
>
> Both attached in a zip archive. Thanks.
Problem is
checking for Apple O
isn't. Since this failed on your system,
the Makefile adds the -framework flag which breaks your build.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
from outside MacOSForge's
network.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
/onlinedocs/gcc/Traditional-GNU-Objective-C-runtime-API.html
> What I meant is what part of MacPorts code uses gnustep.
The Pextlib library (which is a Tcl extension we load into MacPorts' Tcl
interpreter) uses it for some tasks, e.g. to get the cu
See also
http://stackoverflow.com/questions/26316037/git-doesnt-work-after-a-clean-installation-using-macports-due-to-libcurl-4-dyli/
http://superuser.com/questions/736210/how-do-i-remove-an-old-version-of-iconv-on-which-other-software-depends-and-ot/
HTH,
--
Clemens Lang
t,
given the current data protection laws and stuff that happens to your data in
the US…), before we work on that, though.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
out this?
Are you actually trying to compile software using MacPorts on Linux? TBH, I
don't think you will get that to work.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
t prevent these things, so
nobody bothered fixing them.
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
nd all our URLs change...
--
Clemens Lang
___
macports-users mailing list
macports-users@lists.macosforge.org
https://lists.macosforge.org/mailman/listinfo/macports-users
101 - 200 of 343 matches
Mail list logo