July 2007 - macports-users - lists.macosforge.org

updating packages
by Troy Telford 06 Jul '07

06 Jul '07

Normally when I upgrade packages, I use the command: port upgrade outdated Which works fine, however, in the end, I have a bunch of 'inactive' packages on the system. I create RPM packages for Linux professionally; I'm familiar with the dependancies (on Linux, however), and I use Gentoo myself (which has its own ports-like system) I'll readily admit that I don't yet understand the needs & issues with MacPorts; which is why I'm seeking greater wisdom ;) When I try to uninstall many of the recently replaced packages (ie. they've become 'inactive'), I receive a dependancy message that I can't uninstall the port <foo>, as the following ports depend on it (followed by a list of ports). One example I have is 'freetype', which understandibly has a number of dependancies. One of them is cairo. If I run `otool -L libcairo.2.dylib`, I get the list of libraries that it's linked against, and sure enough, libfreetype is there, with the string "(compatibility version 10.0.0, current version 10.15.0)" So it appears that the new 'freetype' package satisfies the dependancies for cairo, and I can remove the 'inactive' package safely. Is there are 'macports approved' method of removing inactive packages -- without messing up the dependancy tree? I generally try to avoid 'forcing' a package removal, and I'm hoping there's a clean way to do it. Any hints/documents I can read for what the 'correct' thing to do is? It seems awfully wasteful to have many inactive versions of a particular package. Thanks, -- Troy Telford

2 1

helpWithError
by Mack Johnson 06 Jul '07

06 Jul '07

Installing "bluefish". ibs/guniprop.o .libs/gutf8.o .libs/gunibreak.o .libs/ gunicollate.o .libs/gunidecomp.o .libs/gutils.o .libs/gprintf.o .libs/ giounix.o .libs/gspawn.o .libs/libglib-2.0.lax/libcharset.a/ localcharset.o -L/opt/local/lib -lc /opt/local/lib/libiconv.dylib / opt/local/lib/libintl.dylib -install_name /opt/local/lib/ libglib-2.0.0.dylib -Wl,-compatibility_version -Wl,1201 -Wl,- current_version -Wl,1201.12 ld64-74.4 failed: bc out of range (394368 max is +/-64K) from _g_atomic_int_exchange_and_add in .libs/gatomic.o to _g_atomic_int_exchange_and_add$stub in .libs/ libglib-2.0.0.1200.12.dylib in _g_atomic_int_exchange_and_add from .libs/gatomic.o collect2: ld returned 1 exit status make[4]: *** [libglib-2.0.la] Error 1 make[3]: *** [all-recursive] Error 1 make[2]: *** [all] Error 2 make[1]: *** [all-recursive] Error 1 make: *** [all] Error 2 Error: The following dependencies failed to build: gtk2 atk glib2 pango Xft2 fontconfig freetype xrender render cairo tiff pcre Error: Status 1 encountered during processing. Mack Johnson m_nez_j(a)mac.com http://macknezjohnson.com

1 0

problem using pkg install : "Package is missing file: COPYING"
by Timothée Brochier 06 Jul '07

06 Jul '07

hello, I want to install the netcdf package for octave (with octave 2.9.9 on an intel mac). Here is what I get : octave:2> pkg install /Users/timbrochier/Desktop/Telechargements/ octave-intel-bin.tar error: Package is missing file: COPYING octave:2> If anyone can help me fixing that, o tell me another way to install the package... Thank you very much, Timothee

1 0

UpdatePortPlease!
by Mack Johnson 06 Jul '07

06 Jul '07

Hi, could someone please update "bluefish-1.0.7" thank you. Mack Johnson m_nez_j(a)mac.com http://macknezjohnson.com

3 3

netcat -L
by Randy 06 Jul '07

06 Jul '07

Trying to run netcat with -L for a persistent listener, but nc complains that -L is an illegal option. Is this something unique to OSX? Have I missed something? Randy

2 3

Draft of a minimalist MacPorts Guide
by markd＠macports.org 06 Jul '07

06 Jul '07

I think the structure of the original guide is fundamentally flawed. I cooked up a draft of a minimalist approach. What do y'all think? http://homepage.mac.com/duling/macports/guide.html I used the stylesheet from another document I had handy, so don't mind the format or the fact that it isn't chunked. The colors and text of the original guide make it look more professional I think but I didn't have time to mess with the stylesheet, especially since I don't know if people will really like this approach anyway. Mark

5 5

Re: UpdatePortPlease!
by Ryan Schmidt 06 Jul '07

06 Jul '07

Michèle, thank you for your response. I'm Cc'ing the macports-users mailing list again to try to direct the discussion back there. On Jul 5, 2007, at 20:52, Michèle Garoche wrote: > Le 5 juil. 2007 à 23:46, Ryan Schmidt a écrit : > >> On Jul 5, 2007, at 15:56, Mack Johnson wrote: >> >>> Hi, could someone please update "bluefish-1.0.7" thank you. >> >> That would typically be the maintainer's job. I'm Cc'ing the >> bluefish maintainer on this email. > > I'm sorry, but I'm not the maintainer of the bluefish macport. I > was only mentionned by the previous maintainer, because I shared > the patches made for the Fink port with him. > > Actually, there seems to be no more maintainer for the bluefish > macport. There is also no more Fink port maintainer and French > translator for bluefish, as I have no more time for both jobs. Well, you were made a comaintainer of bluefish in December 2005: http://trac.macosforge.org/projects/macports/changeset/15530 And the other maintainer dropped maintainership of most (all?) of his ports in March 2006: http://trac.macosforge.org/projects/macports/changeset/17264 Leaving you as the only maintainer of bluefish. But if you don't have time to maintain it and did not want to be its maintainer, then I'll set it to nomaintainer. http://trac.macosforge.org/projects/macports/changeset/26756 > Here a recent reply about the same request: > > <quote> >> Le 3 juil. 2007 à 01:06, Malcolm Fitzgerald a écrit : >>> Hi Michele, >>> >>> I've just downloaded the source code for bluefish v1.0.7 and >>> compiled it. No alterations or special configuration required on >>> my iBook PPC G3 600MHz OSX10.3.9 >>> >>> I wonder if it can be included in MacPorts? >> >> Well, as you want. >> >> What I can say about it after thorough testing, is that it is has >> critical bugs, leading to data loss; that's why I refused to port >> it to Fink. Example: try to open in two different instances of >> bluefish the same file and change one of them. Or try to open two >> windows in the same instance of bluefish, load the same file in >> them, and make changes in one of it. Very weird for translators >> job. Etc... Play extensively with plugins, images, remote >> connection and you will see what I mean. >> >> Moreover, if you dare to inspect the configuration files in depth, >> you will find that instead of using standard process, it reinvents >> the wheel and make drunky-man (to say it mildy) assumption of >> where libraries or binaries are. >> >> And last the UI, though completely internationalized, does not >> load properly when using non English system. Just to give an >> example: about screen, thus leading to ignore translators credits, >> that is not respecting the licence. Severe bug to me. All of these problems you mentioned -- have they just appeared in bluefish 1.0.6 or 1.0.7? or do they exist in 1.0.5 too? If these problems just appeared in recent versions, then sure, that would be a reason that we should not update the port. >> You may think I'm too restrictive, but for me a good-to-be-ported >> software is not only one which compiles (that's a minimal >> requirement, is not it?), but also one which runs correctly under >> any system configuration. >> >> If you look for alternative, screem is far better. Perhaps not so >> polished, but at least it does not crash every five minutes. It >> needs some tweakings in configuration files from cvs, but nothing >> difficult. >> >> Also tea is interesting as having special features I've never >> encountered in any software. It needs some rewriting in graphics >> parts, but nothing especially difficult. > </quote> > > For me the latest stable and good running version of bluefish was > 1.0.5, more recent versions are too buggy for my taste. Obviously, > that should not impede anybody other than me to port it to MacPort > and why not to Fink while at it if so desired :-) Thank you, Michèle, for the detailed explanation. Though I'll admit I'm not reading it very closely since I don't even know what bluefish is; I was just responding to a request on the mailing list to get the port updated, and attempted to forward it on to the responsible party. So, Mack: is there a specific reason why you need bluefish 1.0.7, or would 1.0.5 be enough? Or have you already tried 1.0.7 and found it to work properly?

1 0

Is my MacPorts really sync-ing?
by Mark Hattam 06 Jul '07

06 Jul '07

In that case it's changed what it reports with the -d ... I've always used that because that what the documentation says (or at least said) to do. It used to report the delta, ie the new, the changed and the deleted files. Now it seems to report everything. I was wondering partly because of the several Apache2 Portfile changes I've seen go through on the "Commit Log" list, but haven't apparently been caught by the sudo port -d selfupdate, as the sudo port outdated hasn't brought up any results for Apache2 Mark -- >On 7/5/07, Mark Hattam ><<mailto:mark@dxradio.demon.co.uk>mark(a)dxradio.demon.co.uk> wrote: > >I have MacPorts 1.4.2 on OSX 10.4.10 > >Whenever I've done a port sync recently (and I do it most days), I >seem to be downloading the whole heirarchy every time, and none of my >ports are outdated when I do a port outdated. I used to just get a >listing of the ports that had changed when I did a sudo port -d >selfupdate, I can't believe that everything is updated every day. >Maybe it is subtly changing every day, but it just seems to be the >same long list every day ... > > >It's not that everything changes every day when you see that list: >those are just the things that get checked to see if they've >changed, as far as I know. > >I just did "port -v sync" and here's what I found. > >I get the same list you saw, then these filenames: > >aqua/Growl/files/patch-Growl.xcodeproj__project.pbxproj.i386 > >aqua/Growl/files/patch-Growl.xcodeproj__project.pbxproj.powerpc > >aqua/Growl/files/patch-Makefile > >databases/mysql5/Portfile > >databases/mysql5/files/patch-mysys-base64.c.diff > >databases/sqlite3/Portfile > >databases/sqlite3/files/patch-<http://Makefile.in>Makefile.in > >devel/dialog/Portfile > >devel/libsdl/Portfile > >devel/libsdl/files/patch- AudioFilePlayer.h > >devel/libsdl/files/patch-SDL_coreaudio.c > >editors/vim-app/Portfile > >editors/vim-app/files/patchlist > >editors/vim/Portfile > >editors/vim/files/patchlist > >graphics/cairo/Portfile > >lang/yasm/Portfile > >print/freetype/Portfile > >print/gutenprint/Portfile > >python/py-kid/Portfile > >python/py-yum-metadata-parser/Portfile > >sysutils/repoview/Portfile > >sysutils/repoview/files/patch-repoview.py > >sysutils/rpm/Portfile > >sysutils/rpm/files/patch-configure > >sysutils/rpm/files/patch-<http://macros.in>macros.in > >textproc/libmrss/Portfile > >textproc/libnxml/Portfile > >deleting textproc/libmrss/files/patch-src-mrss_parser.c > >deleting directory textproc/libmrss/files > >deleting databases/mysql5/files/patch-mysys-base64.c > > >so I got some new files, some instructions were run, then this summary: > >sent 10053 bytes received 290465 bytes 2931.88 bytes/sec > >total size is 14699737 speedup is 48.91 > >I sent received more than I sent, so there is some new stuff to be >reviewed with "port outdated." > >I think selfupdate with the debug flag is less useful than sync with >the verbose flag. > >-- >Paul Beard / <http://www.paulbeard.org/>www.paulbeard.org/ ><<http://paulbeard@gmail.com/paulbeard@mac.com>paulbeard@gmail.com/paulbeard@mac.com>

3 4

Re: Is my MacPorts really sync-ing?
by paul beard 05 Jul '07

05 Jul '07

On 7/5/07, Mark Hattam <mark(a)dxradio.demon.co.uk> wrote: > > In that case it's changed what it reports with the -d ... I've > always used that because that what the documentation says (or at > least said) to do. chalk it up to the relentless march of progress. I think quite a few things changed in the 1.3->1.4 cycle. getting better all the time. -- Paul Beard / www.paulbeard.org/ <paulbeard@gmail.com/paulbeard@mac.com>

1 0

Re: Is my MacPorts really sync-ing?
by paul beard 05 Jul '07

05 Jul '07

On 7/5/07, Mark Hattam <mark(a)dxradio.demon.co.uk> wrote: > > I have MacPorts 1.4.2 on OSX 10.4.10 > > Whenever I've done a port sync recently (and I do it most days), I > seem to be downloading the whole heirarchy every time, and none of my > ports are outdated when I do a port outdated. I used to just get a > listing of the ports that had changed when I did a sudo port -d > selfupdate, I can't believe that everything is updated every day. > Maybe it is subtly changing every day, but it just seems to be the > same long list every day ... > It's not that everything changes every day when you see that list: those are just the things that get checked to see if they've changed, as far as I know. I just did "port -v sync" and here's what I found. I get the same list you saw, then these filenames: aqua/Growl/files/patch-Growl.xcodeproj__project.pbxproj.i386 aqua/Growl/files/patch-Growl.xcodeproj__project.pbxproj.powerpc aqua/Growl/files/patch-Makefile databases/mysql5/Portfile databases/mysql5/files/patch-mysys-base64.c.diff databases/sqlite3/Portfile databases/sqlite3/files/patch-Makefile.in devel/dialog/Portfile devel/libsdl/Portfile devel/libsdl/files/patch-AudioFilePlayer.h devel/libsdl/files/patch-SDL_coreaudio.c editors/vim-app/Portfile editors/vim-app/files/patchlist editors/vim/Portfile editors/vim/files/patchlist graphics/cairo/Portfile lang/yasm/Portfile print/freetype/Portfile print/gutenprint/Portfile python/py-kid/Portfile python/py-yum-metadata-parser/Portfile sysutils/repoview/Portfile sysutils/repoview/files/patch-repoview.py sysutils/rpm/Portfile sysutils/rpm/files/patch-configure sysutils/rpm/files/patch-macros.in textproc/libmrss/Portfile textproc/libnxml/Portfile deleting textproc/libmrss/files/patch-src-mrss_parser.c deleting directory textproc/libmrss/files deleting databases/mysql5/files/patch-mysys-base64.c so I got some new files, some instructions were run, then this summary: sent 10053 bytes received 290465 bytes 2931.88 bytes/sec total size is 14699737 speedup is 48.91 I sent received more than I sent, so there is some new stuff to be reviewed with "port outdated." I think selfupdate with the debug flag is less useful than sync with the verbose flag. -- Paul Beard / www.paulbeard.org/ <paulbeard@gmail.com/paulbeard@mac.com>

1 0