May 2009 - macports-tickets - lists.macosforge.org

[MacPorts] #16262: netcdf update 3.6.2 -> 4.0
by MacPorts 22 May '09

22 May '09

#16262: netcdf update 3.6.2 -> 4.0 --------------------------------+------------------------------------------- Reporter: mmoll(a)macports.org | Owner: fulton(a)cmu.edu Type: enhancement | Status: new Priority: Normal | Milestone: Port Updates Component: ports | Version: 1.6.0 Keywords: | --------------------------------+------------------------------------------- Patched Portfile for netcdf upgrade -- Ticket URL: <http://trac.macports.org/ticket/16262> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 27

[MacPorts] #15772: ccache support sets incompatible compiler
by MacPorts 22 May '09

22 May '09

#15772: ccache support sets incompatible compiler ----------------------------------+----------------------------------------- Reporter: landonf(a)macports.org | Owner: macports-tickets(a)lists.macosforge.org Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 1.6.0 Keywords: | ----------------------------------+----------------------------------------- Prior to the introduction of ccache-specific functionality, ccache could be enabled by setting the binpath in macports.conf. This would result in standard compiler names being found in /opt/local/libexec/ccache. With the introduction of ccache-specific functionality -- in of itself a layering violation -- the compiler is now munged at configure-time: {{{ CC=ccache /usr/bin/gcc-4.0 }}} This "two word" CC command breaks some otherwise valid Makefile that expect CC to be a single command. An example: {{{ ifndef $(CXX) CXX = c++ endif }}} Results in: {{{ Makefile:8: *** invalid syntax in conditional. Stop. }}} -- Ticket URL: <http://trac.macports.org/ticket/15772> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 2

Re: [MacPorts] #14846: Couldn't sync the ports tree: Synchronization of the local ports tree failed doing rsync
by MacPorts 22 May '09

22 May '09

#14846: Couldn't sync the ports tree: Synchronization of the local ports tree failed doing rsync ---------------------------------------+------------------------------------ Reporter: macports.users@… | Owner: macports-tickets@… Type: defect | Status: closed Priority: Normal | Milestone: MacPorts Future Component: base | Version: 1.7.0 Resolution: worksforme | Keywords: Port: | ---------------------------------------+------------------------------------ Changes (by jmr@…): * status: new => closed * resolution: => worksforme * component: ports => base -- Ticket URL: <http://trac.macports.org/ticket/14846#comment:5> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 0

[MacPorts] #19219: port uninstall says that libbonobo depends on p5-getopt-long
by MacPorts 22 May '09

22 May '09

#19219: port uninstall says that libbonobo depends on p5-getopt-long ---------------------------------+------------------------------------------ Reporter: vinc17@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: MacPorts 1.8.0 Component: base | Version: 1.7.1 Keywords: | Port: ---------------------------------+------------------------------------------ I've just installed p5-getopt-long for testing (it wasn't required by other ports). Then I wanted to uninstall it, but I get the following error: {{{ $ prunille:~> sudo port -v uninstall p5-getopt-long ---> Unable to uninstall p5-getopt-long 2.37_1, the following ports depend on it: ---> libbonobo Error: port uninstall failed: Please uninstall the ports that depend on p5 -getopt-long first. }}} There seems to be something broken in the "port uninstall" logic. -- Ticket URL: <http://trac.macports.org/ticket/19219> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 4

[MacPorts] #19702: Webalizer MacPorts Port stalls on libpng
by MacPorts 22 May '09

22 May '09

#19702: Webalizer MacPorts Port stalls on libpng --------------------------------------------+------------------------------- Reporter: reedjackson@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Port Binaries Component: ports | Version: 1.7.0 Keywords: | Port: webalizer --------------------------------------------+------------------------------- Attempting to fetch libpng-1.2.35.tar.bz2 from http://internap.dl.sourceforge.net/libpng This port stalls here, most likely because the libpng version at http://internap.dl.sourceforge.net is now 1.2.36 -- Ticket URL: <http://trac.macports.org/ticket/19702> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 3

Re: [MacPorts] #13887: libstreams 24.1 does not build on OSX 10.5.1 + Xcode 3.0 (was: libstreams 24.1 does not build on OSX 10.5.1 + XCode 3.0)
by MacPorts 22 May '09

22 May '09

#13887: libstreams 24.1 does not build on OSX 10.5.1 + Xcode 3.0 -------------------------------+-------------------------------------------- Reporter: ralf@… | Owner: pguyot@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: Keywords: | Port: Libstreams -------------------------------+-------------------------------------------- -- Ticket URL: <http://trac.macports.org/ticket/13887#comment:5> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 0

[MacPorts] #19200: bastet: don't create directory called "755"
by MacPorts 22 May '09

22 May '09

#19200: bastet: don't create directory called "755" -------------------------------------+-------------------------------------- Reporter: ryandesign@… | Owner: sfiera@… Type: defect | Status: new Priority: Normal | Milestone: Port Bugs Component: ports | Version: 1.7.1 Keywords: | Port: bastet -------------------------------------+-------------------------------------- bastet creates a directory called "755" in the current directory when it's installed. There's an incorrect xinstall command in the destroot phase. The simplest fix is to remove that line entirely since it's not needed since MacPorts base creates ${destroot}${prefix}/bin for you. -- Ticket URL: <http://trac.macports.org/ticket/19200> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 1

[MacPorts] #19189: ATLAS 3.8.2 gets compiled for 64bit on some architectures
by MacPorts 22 May '09

22 May '09

#19189: ATLAS 3.8.2 gets compiled for 64bit on some architectures -------------------------------+-------------------------------------------- Reporter: max@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Port Bugs Component: ports | Version: 1.7.1 Keywords: atlas 64bit 32bit | Port: atlas -------------------------------+-------------------------------------------- On architectures where the ATLAS build script detects 64bit support it compiles ATLAS as a 64bit library. This results in linking errors when linked to from other ports, as those are compiled in 32bit. When configure is called with the '-b 32' flag, a 32bit compilation is forced. I'm not sure what the way to proceed is, my simple patch just adds the -b 32 flag to configure.args -- Ticket URL: <http://trac.macports.org/ticket/19189> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 2

[MacPorts] #17427: MacPorts 1.6 installation failure after Leopard hard drive replacement
by MacPorts 22 May '09

22 May '09

#17427: MacPorts 1.6 installation failure after Leopard hard drive replacement ------------------------------------+--------------------------------------- Reporter: equipoise1@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Port Bugs Component: ports | Version: 1.6.0 Keywords: | Port: ------------------------------------+--------------------------------------- I recently replaced a hard drive on my Core2Duo MacBook Pro. I had previously installed MacPorts on the old drive. After replacing the old drive and installing leopard, I had all of my Mac OS X user settings copied onto the new drive. I then installed Xcode and X11. Next I installed MacPorts on the new drive. Then I tried: port variants octave and port install octave This chugged away for a while but produced error msgs. The gist of these concerned Perl 5.8.8.something and the fact that /some/path/to/a2p is being used (is active?) by an earlier version of perl5. I've googled to some stuff about this, including the same error msg. This msg includes a suggestion about using the "-f" option. Question 1: Where do I read about the "-f" option, which I believe pertains to the "port" command, and information explaining other options? An hour of googling on this has been largely unproductive. I've also been unable to bring up a man page on "port". I believe these problems have to do with vestigial registry information from my earlier installation on the old drive. Now I'd like to revert my OS to where it stood after I installed it on the new drive but before I installed MacPorts. I'm concerned about losing things like gcc, which existed independent of what I've tried with MacPorts. Question 2: How can I remove everything related to MacPorts, including anything that it did install, while not endangering things that would have been come with the Leopard and Xcode installation I did? Question 3: Before removing things installed under MacPorts, is there a way to determine which of these components may be usable? Question 4: When I've purged the old things, should I be able to reinstall MacPorts according to the instructions in the guide, or will some alternative approach be required? Thanks. -- Ticket URL: <http://trac.macports.org/ticket/17427> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 2

[MacPorts] #17332: checksum error when installing jdom-1.6.1
by MacPorts 22 May '09

22 May '09

#17332: checksum error when installing jdom-1.6.1 ---------------------------------------+------------------------------------ Reporter: sgendler(a)appliedminds.com | Owner: macports-tickets(a)lists.macosforge.org Type: defect | Status: new Priority: Normal | Milestone: Port Bugs Component: ports | Version: 1.6.0 Keywords: dom4j | Port: dom4j ---------------------------------------+------------------------------------ I got repeated checksum errors, despite clean --all and other remedies. I compared Portfile to checksum of file retrieved from sourceforge URL in port output and the checksums matched. The checksum for the file downloaded by port, however, was utterly different: INCORRECT: [96] ...distfiles/dom4j %>openssl md5 dom4j-1.6.1.tar.gz MD5(dom4j-1.6.1.tar.gz)= 520b2f0cbfcabca66786d8da1d033ae5 CORRECT: ~/Downloads %>openssl md5 dom4j-1.6.1.tar.gz MD5(dom4j-1.6.1.tar.gz)= 1e7ef6d20939315714de4a8502f27b2d Both files were downloaded simultaneously - one via Safari and the other via port. I'd be inclined to blame some sort of caching between my computer and sourceforge, except that it appears to be dependent upon user-agent. I just checked, and got the correct file via mozilla, too. I'll probably simulate the port user-agent in a browser and sees what happens in a bit, but I need to get back to work for a while, first. I've been battling this problem for well over an hour already. It'd help if someone were to post the user-agent so that I don't have to go sniffing for it on the wire. -- Ticket URL: <http://trac.macports.org/ticket/17332> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 4