New Ticket     Tickets     Wiki     Browse Source     Timeline     Roadmap     Ticket Reports     Search

Custom Query (25206 matches)

Filters
 
Or
 
  
 
Columns

Show under each result:


Results (6001 - 6100 of 25206)

Ticket Summary Cc Owner Type Status Priority
#18241 RFE: have doxygen always use configure.cxx css@… enhancement closed Normal
Description

Attached is a proposed change to ensue doxygen links using configure.cxx
instead of c++.

#5041 RFE: graphviz-2.6 needs explicit dependency on guile and java gwright@… defect closed Normal
Description

The graphviz configure will happily build with or without scheme/java components depending on whether the appropriate ports are installed or not.

How graphviz builds should be tracked explicitly through dependencies (or variants?) for reproducuibility.

#12027 RFE: gnustep portgroup jberry@… enhancement closed Low
Description

gnustep portgroup ready for gnustep-make 2.0

#6354 RFE: gmake-3.80: use bzip2 waqar@… defect closed Normal
Description

The attached patch enables the use of bzip2 in the gmake-3.80 port.

#14482 RFE: Global variant descriptions (per source) raimue@… enhancement closed Normal
Description

There are variants which are used on multiple ports. To provide a consistent experience, these should get only one place for their description.

Proposal:

  • Add a file ${sourcepath}/.config/variant_descriptions to each source
  • The file should be formatted in Tcl style as follows:
    name1 {The description of name1 variant}
    name2 {Insert description here}
    
  • If no description was given in the Portfile itself, port variants will read this file (if present) and display the variant description from there
  • port lint should be tweaked to check for existance of global variant descriptions before spitting out warnings
#3605 RFE: gimp2 with altivec, print, gtkhtml and rsvg jmpp@… defect closed Normal
Description

I am willing to maintain this port

#17107 RFE: ghostscript changes takanori@… enhancement closed Normal
Description

Attached are proposed enhancements to ghostscript:

  • I do not think patch-src_unix-aux.mak.diff is needed.
  • pkgconfig is only needed in the build process.
  • ${prefix}/bin/fixmswrd.pl is a perl script, so use the MacPorts perl.
  • zlib automatically get linked in (with or without framework option), so make is a dependency.
  • Use the MacPorts 1.7 variable frameworks_dir.
  • Use the x11prefix variable.
#15670 RFE: gcc43 variant for py25-numpy ram@… enhancement closed Normal
Description

Can we please get back the gcc43 variant for py25-numpy?

I would like to build my complete toolchain, including numy and scipy, as well as my own code, using gfortran from GCC-4.3. It worked nicely until recently. Why were the already working variants removed?

r37408 | ram@macports.org | 2008-06-06 15:24:41 +0200 (Fri, 06 Jun 2008) | 2 lines
Changed paths:
   M /trunk/dports/python/py-numpy/Portfile
   M /trunk/dports/python/py25-numpy/Portfile

py-numpy/py25-numpy: remove unnecessary fortran variants

Don't know why they are supposed to be unnecessary? I want them back;-)

#14712 RFE: gcc43: Add support for gcc_select mww@… defect closed Normal
Description

Here is a patch that adds a select file for support by gcc_select. It was adapted from the select file for gcc42.

#4106 RFE: gcc34 and g77 gwright@… enhancement closed Normal
Description

since gcc34 now provides g77, it would be nice that the symlink be done

--- Portfile.orig       2005-07-21 21:22:14.000000000 -0400
+++ Portfile    2005-07-21 16:20:04.000000000 -0400
@@ -63,5 +63,7 @@
 post-destroot   { cd ${destroot}${prefix}
                  file delete -force share/man/man7
                  file delete -force share/info
+                 cd ${destroot}${prefix}/bin
+                 system "ln -s g77-dp-${major} g77"
                }
#19527 RFE for lzo2: Remove '--disable-static' from configure.args yunzheng.hu@… enhancement closed Normal
Description

I've attached a trivial patch that removes '--disable-static' from configure.args.

Is there any reason why static libraries are disabled by default?

#3105 RFE: fftw-3 gwright@… defect closed Normal
Description

I had problems with fftw and made an update without noticing that a fftw-3 port existed.

However, I see that my version has some features that yours don't, like altivec and fortran, with fma declared as a ppc variant, so I submit it anyway

#17310 RFE: ffmpeg: change port version number to reflect current svn revision acho@… enhancement closed Normal
Description

On their download page http://www.ffmpeg.org/download.html, the ffmpeg developers say

FFmpeg has always been a very experimental and developer-driven
project. It is a key component in many multimedia projects and has 
new features added constantly. New, official "releases" are few and 
far between. In short, if you want to work with FFmpeg, you are 
advised to go along with SVN development rather than relying on 
formal releases. SVN snapshots work really well 99% of the time so 
people are not afraid to use them. 

and

If you are looking for a formal release, stop now, there are none. 
Maybe we will have some in the future, but don't hold your breath. 

The ffmpeg port version is 0.4.9pre1_12. This is the number of the last official release of ffmpeg which corresponds to svn revision 3298, Fri Jul 9 20:08:33 2004 UTC, over 4 years ago.

The current ffmpeg port svn revision is 14381 and ffmpeg itself reports its version as

$ ffmpeg --version
FFmpeg version SVN-r14381, Copyright (c) 2000-2008 Fabrice Bellard, et al.

Based on this information, I propose that we change the ffmpeg port version to 0.4.9-${svn_rev} where ${svn_rev} is the current svn revision used in the port.

I think this would yield a more informative and accurate version number that is consistent with the version the program itself reports but allows for possible future release numbers.

I have attached a patch to make this change but in light of the fact that this change is a relatively significant one, I am asking for concurrence and/or comments from the port maintainer (even though the port is marked openmaintainer) before any such change is made.

#10618 RFE: feature to replicate gentoo use flags functionality macports-tickets@… enhancement closed Normal
Description

Hey, It would be really handy if:

on the configure.args line of each Portfile there was a {$UserOptions} var that was filled with the contents of a /opt/local/etc/ports/configure/{Category}/{PackageName}.{Version}|{Global} file. I was thinking the options possible for each package could be listed in a {PackageName}.options file for gui tools/control panels to use.

the default build options being specified there. It would allow us to make permanent config.args.opt changes that would be preserved (not clobbered) when we sync the main ports tree. You could then put a sync action sync-configs instead of just 'sync' to sync only config files if you really want to reset config.opts the to defaults.

This would give you the functionality of gentoo use flags without a complicated use-flag system. If you merge Version-specific options over the top of Global opts you have just a two layer flattening to do to result in the final config options.

I think this would provide a straight forward way of managing package options that would not have to be re-applied while keeping the overall system up to date.

Any comments, or oh,ba'shaws! to be directed to mrgibson@… thanks. :)

#17097 RFE: ensure that latexdiff uses MacPorts perl aschenke@… enhancement closed Normal
Description

Attached is a proposed change to latexdiff:

  • Corrects lint errors.
  • Installs documentation.
  • Ensures the the system perl is not used.
  • Makes perl5 the default perl (see #16830).
#17060 RFE: Ensure that automake uses MacPorts perl mww@… enhancement closed Normal
Description

Attached is a proposal to ensure that automake uses the MacPorts perl.

See #16830 for the reasons for using perl5 instead of perl5.8.

#4209 RFE: emacs lacks mac-command-key-is-meta macports-tickets@… defect closed High
Description

variable (or its effect), as described in the emacs info page (on the mac version). This prevents one (apparently) from binding the meta key to something other than escape (i.e. to command or option, the normal defaults on the mac platform)

#15433 RFE: Eliminate duplicates in portlist raimue@… enhancement closed Normal
Description

See the following output:

$ port echo name:vim or description:vim
p5-text-vimcolor                
vim                             
vim-app                         
vim2html                        
p5-text-vimcolor                
vim-app                         
vim2html

It should be avoided to list ports twice or more times here.

#19139 RFE: Do not display revision 0 in composite version macports-tickets@… enhancement new Normal
Description

At many places port uses the composite version format to display version information. This consists of the version string as given in the Portfile and an additional underscore with the revision. As the default revision is 0, there is no need to display it.

So @1.2_0 would become @1.2 instead. This would increase the readability of port's output as most ports never increase the revision.

If specified in input as for port uninstall foo @1.2 it should be assumed to be revision 0 only if unambiguous.

#5506 RFE: dhcp 3.0.2 port's startup scripts are missing cms@… defect closed Normal
Description

net/dhcp has the runtime dependency sysutils/DarwinPortsStartup, but doesn't include the necessary ${prefix}/etc/rc.d/ scripts

#6771 RFE: dbus needs to be upgraded macports-tickets@… defect closed Normal
Description

The dbus port is old and there have been ABI and API changes to dbus that prevent newer or upgraded applications from using the version currently in dports.

#554 RFE: darwinports has no method to designate empty directories as useful landonf@… enhancement closed Normal
Description

Some ports, including apache create empty directories for things like .pid files or log files that are created at runtime, there should be a way to avoid dports removing these dir other than putting .turd files in there.

#590 RFE: cvsync port macports-tickets@… enhancement closed Normal
Description

Here is a Portfile for cvsync. Jason

#19096 RFE: create system_x11 variant for mkfontdir jeremyhu@… enhancement closed Normal
Description

Attached is a proposed change to mkfontdir since XQuartz seems to provide
/usr/X11/bin/mkfontdir.

#650 RFE: coreutils port macports-tickets@… enhancement closed Normal
Description

coreutils port. Should obsolete fileutils, shellutils, and textutils.

#14557 RFE: Commands for installing .app bundles ryandesign@… enhancement closed Normal
Description

Commands for installing .app bundles would be helpful. See for example how the tuxmath port does it at the moment.

post-destroot {
    xinstall -m 755 -d ${destroot}/Applications/MacPorts/TuxMath.app/Contents/MacOS
    ln -s ${prefix}/bin/tuxmath ${destroot}/Applications/MacPorts/TuxMath.app/Contents/MacOS/TuxMath
}

This could be done like this:

app.install    yes
app.name       TuxMath
app.binary     ${prefix}/bin/tuxmath

Additional directives would be needed for inclusion of Resources etc. Also, a Info.plist with version information should be generated.

#147 RFE:cmd_name depspec {#do stuff}, tcl primitive to provide conditionals macports-tickets@… enhancement closed High
Description

A number of ports nned to be able to test the environment (test installed libs/files/location of installed items) to be able to do the right thing (set variables, install locations, etc).

I propose a syntax like :cmd_name depspec {#do stuff}

where we reuse depspec syntax to test for files/libs/etc and then are able to set variables in a block, for example change the default variant in response to what is providing a dependancy

#2899 RFE: cmake-2.0.6 syntax files for emacs and vim cssdev@… enhancement closed Normal
Description

This patch for devel/cmake/Portfile downloads Kitware's CMake emacs and vim syntax files for CMakeLists.txt and *.cmake files. These three files are downloaded from the CMake website, and they are installed into emacs and vim directories during the destroot process. These files are not included with the source, but they are provided directly on the download page. It might be preferrable to add these as a variant, but these three files probably are not significant enough to constitute a full variant. The diff below assumes BUG #2898 resolution, bringing the CMake version to 2.0.6. Should the emacs file be byte-compiled prior to installation?

--- Portfile.206        Thu Apr 14 12:53:19 2005
+++ Portfile    Thu Apr 14 13:40:34 2005
@@ -13,6 +13,32 @@
        used in conjunction with the native build environment.
 homepage       http://www.cmake.org/
 platforms      darwin freebsd
-master_sites   http://www.cmake.org/files/v2.0/
-checksums      md5 840ea86d7f1a56b469052c9cabbeac54
+master_sites   http://www.cmake.org/files/v2.0/:source \
+               http://www.cmake.org/CMakeDocs/:syntax
+
+distfiles      ${name}-${version}.tar.gz:source \
+               ${name}-mode.el:syntax \
+               ${name}-indent.vim:syntax \
+               ${name}-syntax.vim:syntax
+
+extract.only   ${name}-${version}.tar.gz
+
+checksums      ${name}-${version}.tar.gz \
+                       md5 840ea86d7f1a56b469052c9cabbeac54 \
+               ${name}-mode.el md5 9e84a1509d5ea5b0c669e0eb016c9b41 \
+               ${name}-indent.vim md5 9ed4a0a1cec2957f2e839fd4b904d55c \
+               ${name}-syntax.vim md5 7337be86138adee979c0e9df216995cd
+
 configure.args --mandir=/share/man
+
+post-destroot  {
+       xinstall -d -m 0755 ${destroot}${prefix}/share/emacs/site-lisp
+       xinstall -m 0644 ${distpath}/${name}-mode.el \
+               ${destroot}${prefix}/share/emacs/site-lisp
+
+       xinstall -d -m 0755 ${destroot}${prefix}/share/vim/syntax
+       xinstall -m 0644 ${distpath}/${name}-indent.vim \
+               ${destroot}${prefix}/share/vim/syntax
+       xinstall -m 0644 ${distpath}/${name}-syntax.vim \
+               ${destroot}${prefix}/share/vim/syntax
+}
#18102 RFE: Change perl dependency of xmlwrapp stechert@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of xmlwrapp. See #16830 for the reasons.

#17513 RFE: change perl dependency of torrentsniff eridius@… enhancement closed Normal
Description

Attached is a proposed change in the perl dependency of torrentsniff.
Since torrentsniff uses the perl5 PortGroup, I do not think perl dependency is needed.
It also resolves issues brought up in #17182 and #16830.

#17514 RFE: Change perl dependency of subversion-perlbindings dluke@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of subversion-perlbindings.
See #16830 and #17182 for the reasons.

#17940 RFE: Change perl dependency of sleuthkit md@… enhancement closed Normal
Description

Attaches is a proposed change to the perl dependency of sleuthkit.
See #16830 for the reasons.

#17640 RFE: Change perl dependency of sendemail mww@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of sendemail.
See #16830 for the reasons.

#17440 RFE: change perl dependency of rrdtool markd@… enhancement closed Normal
Description

Attached is a proposal to change the perl dependency of rrdtool.
See #16830 for the reasons.

#17801 RFE: Change perl dependency of redland-bindings arto.bendiken@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of redland-bindings.
See #16830 for the reasons.

#17441 RFE: change perl dependency of rancid markd@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of rancid.
See #16830 for the reasons.

#17641 RFE: Change perl dependency of p5-file-temp-svkonly dluke@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of p5-file-temp-svkonly.
See #16830 for the reasons.

#18100 RFE: Change perl dependency of openldap landonf@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of openldap.
See #16830 for the reasons.

#17443 RFE: change perl dependency of nedi markd@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of nedi.
See #16830 for the reasons.

#17448 RFE: change perl dependency of nagios-plugins markd@… enhancement closed Normal
Description

Since nagios-plugins uses the perl5 portgroup, I believe that the perl
dependency is already taken care of.

This also fixes the issues raised in #16830.

#17575 RFE: change perl dependency of mod_perl2 ryan@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of mod_perl2.
See #16830 for the reasons.

#17444 RFE: change perl dependency of mediaserv ecronin@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of mediaserv.
See #16830 for the reasons.

#17445 RFE: change perl dependency of jlj mstevens@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of jlj.
See #16830 for the reasons.

#17814 RFE: Change perl dependency of ikiwiki tommyd@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of ikiwiki.
Since ikiwiki already uses the perl5 PortGroup, the perl5.8 dependency is not
needed.

The change would go towards solving #16830.

#17815 RFE: change perl dependency of gallery ckieffer@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of gallery.
See #16830 for the reasons.

#17944 RFE: Change perl dependency of frozenbubble2 pmq@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of frozenbubble2.
See #16830 for the reasons.

#17571 RFE: change perl dependency of clearsilver dluke@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of clearsilver.
See #16830 for the reasons.

#17446 RFE: change perl dependency of bro pmq@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of bro. See #16830 for the reasons.

#17574 RFE: change perl dependency of boxbackup ecronin@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of boxbackup.
See #16830 for the reasons.

#17573 RFE: Change perl dependency of autopsy ecronin@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of autopsy.
See #16830 for the reasons.

#18101 RFE: Change perl dependency of automake17 springer.jonathan@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of automake17. See #16830 for the reasons.

#17447 RFE: change perl dependency of argus-monitor markd@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of argus-monitor. See #16830 for the reasons.

#17132 RFE: change perl dependency in udunits takeshi@… enhancement closed Normal
Description

Attached is a proposal to change the perl dependency in udunits.
I believe that perl is only needed during the build phase.
See #16830 for the change to perl5.

#17137 RFE: change perl dependency in tablet-encode ecronin@… enhancement closed Normal
Description

Attached is a proposal to modify the perl dependency in tablet-encode.
See #16830 for the reasoning.

#17439 RFE: change perl dependency in sclj kayos@… enhancement closed Normal
Description

Attached is a patch to change the perl dependency in sclj.
See #16830 for the reasons.

#17394 RFE: change perl dependency in imapsync mww@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency in imapsync.
See #16830 for the reasons.

#17136 RFE: change perl dependency in gnupod pmq@… enhancement closed Normal
Description

Attached is a proposal to modify the perl dependency in gnupod.
See #16830 for the reasoning.

#17512 RFE: Change perl dependency for wakeonlan mww@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of wakeonlan.

#17570 RFE: Change perl dependencies of vim and vim-app raimue@… enhancement closed Normal
Description

Attached are proposed changes to the perl dependencies of vim and vim-app.
See #16830 for the reasons.

#17395 RFE: change perl dependencies of terminus-font rlonstein@… enhancement closed Normal
Description

Attached is a proposed update to the dependencies of terminus-font.

As far as I can tell, perl is not needed once the fonts are installed.
I also changed the perl due to the issues in #16830.

#17572 RFE: change perl dependencies of spidereyeballs mstevens@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependencies of spidereyeballs.
See #16830 for the reasons.

#17442 RFE: change perl dependencies of nfsen markd@… enhancement closed Normal
Description

Attached is a proposed update to the perl dependency of nfsen.
See #16830 for the reasons.

#17817 RFE: Change perl dependencies of mecab-jumandic-* rsky0711@… enhancement closed Normal
Description

Attaches is a proposed change to the perl dependencies of mecab-jumandic-*.
See #16830 for the reasons.

#17267 RFE: change perl dependencies in pspp jeremy@… enhancement closed Normal
Description

Attached is a proposed change in the perl dependencies of pspp.
See #16830 for the reasons.

#17147 RFE: change perl dependencies in metasploit2 opendarwin.org@… enhancement closed Normal
Description

Attached is a proposal to change the perl dependencies in metasploit2.
See #16830 for the reasoning.

#17128 RFE: change faad2 depends_lib to depends_build pguyot@… enhancement closed Normal
Description

I believe that faad2 only depends on libtool, autoconf, and automake
during the build stage.

I don't think it depends on perl at all.

Attached is the proposed change.

#17938 RFE: change default perl used in dpkg landonf@… enhancement closed Normal
Description

Attaches is a proposed change of dpkg. See #16830 for the reasons.

#17937 RFE: change default perl used in chsuf mww@… enhancement closed Normal
Description

Attaches is a proposed change of chsuf.
See #16830 for the reasons.

#17642 RFE: Change default perl port of ps2eps aschenke@… enhancement closed Normal
Description

Attached is a proposed change to the perl dependency of ps2eps.
See #16830 for the reasons.

#14667 RFE: ccache: Move symlinks to avoid mtree violation nox@… defect closed Normal
Description

Currently, ccache violates the mtree by using ${prefix}/ccache. The symlinks inside this directory should be moved to ${prefix}/libexec/ccache where this would be more appropriate.

An ui_msg should inform users about the change.

If done, the ccache HOWTO needs an update.

#19455 RFE: Build SuiteSparse shared libraries stechert@… enhancement closed Normal
Description

Attached is a proposed change to SuiteSparse to build shared libraries.
It also allows universal builds.

#12280 RFE: buildmakejobs setting (simultaneous make) macports-tickets@… enhancement closed Normal
Description

When building with Make and multiple CPUs (or slow Makefile targets), sometimes things go a lot faster when using multiple commands at once with "make -j2". Support for that MacPorts is good.

Note that setting the number of jobs higher than e.g. "# CPU + 1" is a good way of using too much memory and making the build slow to a crawl due to lack of available memory and swapping instead.

#19173 RFE: build libxml2 with muniversal PortGroup cedric.luthi@… enhancement closed Normal
Description

Attached is a proposed change to the universal build of libxml2.

#14914 RFE: build fftw-3 shared libraries takeshi@… enhancement closed Normal
Description

As noted in the mailing list, fftw-3 does not build shared libraries.
Attached is a proposal to allow shared libraries to be built.
http://lists.macosforge.org/pipermail/macports-users/2008-April/009783.html

I also removed configure arguments which have no effect.

#25770 RFE: bitstream-vera does not install architecture dependent files jbenninghoff@… enhancement closed Normal
Description

As far as I can tell, bitstream-vera does not install architecture dependent files.
Attached is a proposed change to let bitstream-vera be a better dependency of a universal port.

#6353 RFE: bison-2.1: use_bzip2 toby@… defect closed Normal
Description

The attached patch changes the bison-2.1 port to use the bzip2 archive instead of the gzipped tarball.

#37921 RFE: binutils's "notes" entry should specify which ports exactly it causes to fail to build macports-tickets@… enhancement new Normal
Description

Right now binutils comes with the following "notes" entry in its portfile:

"Having ${name} installed will cause some other ports to\
fail to build. Consider uninstalling ${name}."

I'd like to know which ports exactly these are, or at least have a link to a Trac ticket or something, so I can be better informed about whether I should actually follow this advice to uninstall it or not.

#11824 RFE: Base - Add startupitems for Inetdcompatibility macports-tickets@… enhancement closed Low
Description

This needs a little fine tuning probably, but here are modified portstartupitem.tcl and portdestroot.tcl (complete and patches) that add:

startupitem.inetdcompat startupitem.inetdcompat.socketsvcname startupitem.inetdcompat.nowait

#38120 RFE: automatically remove .DS_Store files from build directories macports-tickets@… enhancement new Normal
Description

Often times I'll stop a port's build part way through to check on stuff in Finder, and when I do that, the port's build path will get littered with .DS_Store files, causing the port's build folder to not get deleted during the cleaning step, because it won't be empty. These files should be deleted so that build directories get cleaned away properly.

#272 RFE: authorized sendmail should be in ports macports-tickets@… enhancement closed Normal
Description

It would be nice to be able to enable authorized sendmail from ports. This is probably best by first completing a sendmail port, then adding a sasl variant, but it could be done any number of ways.

Here are the instructions for building the software. Hopefully patching dlcompat is not needed, if it is then it should either be verified that the patch is compatible with everything else that uses dlcompat or a custom version should be installed with a different name. Hopefully authorization can also be enabled for a released version of sendmail as well as sendmail from cvs.

$ cvs -d cvs.opendarwin.org:/cvs/od co dlcompat
$ cd dlcompat; sudo make install

$ curl -O ftp://ftp.andrew.cmu.edu/pub/cyrus-mail/cyrus-sasl-1.5.27.tar.gz
$ tar xzf cyrus-sasl-1.5.27.tar.gz
$ cd cyrus-sasl-1.5.27/config/
$ cp /usr/libexec/config* ./
$ curl -O http://fink.sourceforge.net/files/ltconfig
$ curl -O http://fink.sourceforge.net/files/ltmain.sh
$ cd ..
$ ./configure --with-plugindir=/usr/local/lib/sasl --libdir=/usr/local/lib --mandir=/usr/share/man --disable-krb4 --disable-gssapi --enable-login

$ vi plugins/Makefile
--
-plugins/Makefile: LDFLAGS = -L/usr/local/lib -export-dynamic -rpath $(plugindir)
+plugins/Makefile: LDFLAGS = -L/usr/local/lib -export-dynamic -module -rpath $(plugindir)
--

$ vi lib/dlopen.c lib/saslutil.c

----- begin patch -----
--- cyrus-sasl-1.5.27/lib/dlopen.c      Sun Aug 13 18:44:02 2000
+++ cyrus-sasl-1.5.27-modified/lib/dlopen.c     Mon May 13 16:06:01
2002
@@ -158,7 +158,7 @@
     int flag;
     void *library;
     void *entry_point;
-#if __OpenBSD__
+#if __OpenBSD__ || __APPLE__
     char adj_entryname[1024];
 #else
 #define adj_entryname entryname
@@ -179,7 +179,7 @@
        return SASL_FAIL;
     }

-#if __OpenBSD__
+#if __OpenBSD__ || __APPLE__
     snprintf(adj_entryname, sizeof adj_entryname, "_%s", entryname);
 #endif

---- end patch -----
---- begin patch -----
--- cyrus-sasl-1.5.27/lib/saslutil.c    Wed Dec 20 22:03:43 2000
+++ cyrus-sasl-1.5.27-modified/lib/saslutil.c   Mon May 13 16:18:12
2002
@@ -358,6 +358,12 @@
     if (!rpool->initialized) {
        getranddata(rpool->pool);
        rpool->initialized = 1;
+#ifdef __APPLE__
+      {
+       long *foo = (long *)rpool->pool;
+       srandom(*foo);
+      }
+#endif /* __APPLE__ */
     }
 }

@@ -370,6 +376,11 @@
     /* init if necessary */
     randinit(rpool);

+#ifdef __APPLE__
+    for (lup=0;lup<len;lup++) {
+        buf[lup] = (char) (random() >> 8);
+    }
+#else /* __APPLE__ */
 #ifdef WIN32
     for (lup=0;lup<len;lup++)
        buf[lup] = (char) (rand() >> 8);
@@ -377,6 +388,7 @@
     for (lup=0; lup<len; lup++)
        buf[lup] = (char) (jrand48(rpool->pool) >> 8);
 #endif /* WIN32 */
+#endif /* __APPLE__ */
 }

 /* this function is just a bad idea all around, since we're not trying to
----- end patch -----

$ sudo make install
$ vi /usr/local/lib/sasl/Sendmail.conf
----- Use the system password database for authentication ------
pwcheck_method: passwd
-----

$ cvs -d cvs.opendarwin.org:/cvs/Apple co sendmail

$ vi sendmail/devtools/Site/site.config.m4
------
APPENDDEF(`confENVDEF', `-DSASL')
APPENDDEF(`conf_sendmail_LIBS', `-lsasl -ldl')
------

$ mkdir /tmp/sendmail.obj
$ mkdir /tmp/sendmail.dst
$ sudo make DSTROOT=/tmp/sendmail.dst OBJROOT=/tmp/sendmail.obj RC_ARCHS=ppc install

$ cp -Rp /tmp/sendmail.dst/* /


-- edit sendmail.cf m4 -- trust authenticated users for relaying
define(`confAUTH_MECHANISMS', `PLAIN LOGIN')
TRUST_AUTH_MECH(`PLAIN LOGIN')

[references]
http://www.sendmail.org/~ca/email/auth.html
http://asg.web.cmu.edu/archive/message.php?mailbox=archive.cyrus-sasl&msg=1001
#349 RFE: a tcl primitive to populate destdir kvv@… enhancement closed Normal
Description

A fair number of ports don't populate desdir before trying to install, and fal over if you don't do it for them, as such many ports have a lot of install lines making the directories I propose a tcl primitive that takes a list and makes directories )named as specified in the list) in /destroot/prefix/

use something like

make-dir bin man lib

#5736 RFE: apache2 port only allows prefork MPM when worker MPM is Apple recommended james@… enhancement closed Normal
Description

The apache2 port builds using the default MPM for "unix" distributions which is the prefork MPM. This MPM is a non-threaded pre-forking implementation. However, in recent Apple documentation ( http:// developer.apple.com/tools/subversionxcode.html ) it is recommeneded that apache2 is built using the worker MPM (the multi-process, multi-threaded MPM). Accoring to Apache, its virute lies in the fact that "by using threads to serve requests, it is able to serve a large number of requests with less system resources than a process-based server. Yet it retains much of the stability of a process-based server by keeping multiple processes available, each with many threads."

I would recommend that in the least there is a variant that adds --with-mpm=work to the configure.args, but believe that it should be enabled by default and that the prefork MPM be a variant for those that can't/ don't want to use it.

#583 RFE: apache2 port macports-tickets@… enhancement closed Normal
Description

Here are the various files needed for an apache2 port. It installs into /opt/local/apache2 to avoid clobbering an existing apache 1 installation. The issues that I think exist (others may find more) are: a) No mod_perl or mod_python. I really don't know these too well so I thought I'd submit the port and someone who does know them can add them. b) No php update. Again I don't really know it. c) No suexec. I could add this, but without mod_perl or mod_python I didn't see the point. In order to be able to create modules with this I symlinked /opt/local/apache2/build/libtool to /usr/bin/glibtool, which places a requirement on the Apple Developer Tools, but since DarwinPorts itself requires that I didn't think it was a problem. The patchfile makes the default install use the www user that exists on all systems (no way to pass that arg to configure as best as I could tell). The apache2.sh is only installed if you install the variant +activate_server, just like the apache 1 port. Jason

#18189 RFE: alternate universal build of libiconv ryandesign@… enhancement closed Normal
Description

Attached is an alternate build mechanism for libiconv which,
hopefully, is a little easier to maintain.

#18190 RFE: alternate universal build of libiconv ryandesign@… enhancement closed Normal
Description

Attached is an alternate build mechanism for libiconv which,
hopefully, is a little easier to maintain.

#18191 RFE: alternate universal build of gettext ryandesign@… defect closed Normal
Description

Attached is an alternate build mechanism for gettext which should
make it a little easier to maintain.

#12310 RFE: Allow packages in base/src to use different libs sfiera@… enhancement closed Normal
Description

There are a number of different dylibs that are compiled in base/src/*: macports; pextlib; darwintracelib; tclobjc; and in the future, cflib and registry2.0. All of these need different libraries to function, but there's not currently a way to link different libraries into the different dylibs (well, cflib and tclobjc sort of do it, but they still link a lot of unnecessary stuff). I've prepared a patch that will allow these different packages to have full control over the libs they link to.

While I don't think it's all that detrimental to link libz, libcurl, libcrypto, sqlite3, and readline into e.g. darwintracelib, I'm pretty sure it's not necessary--I've installed the results of my patch and port -t still works just fine--and it strikes me as safer/cleaner.

#25066 RFE: Allow arpack to build universal shared libraries mmoll@… enhancement closed Normal
Description

Attached is a proposed patch to arpack to allow shared libraries and a 32/64-bit universal build.

The g95 variant has been removed to facilitate this change.

#5065 RFE: a fix for X11/app-defaults recurrent problems macports-tickets@… enhancement closed Normal
Description

Many ports use a post-destroot procedure to fix X11/app-defaults which will break the activation if port tries to overwrite the symlink in /usr/X11R6/lib/X11 with a directory, which is the case for Apple's X11 for instance.

However, if /usr/X11R6/lib/X11 really is a directory, which is the case for XFree86 4.5.0, app-defaults placed in /etc/X11 are boldly ignored.

So here is a little addition to the destroot_finish procedure which - hopefully - cope with the issue by adapting the app-defaults location to the user's setup, thus making app-defaults fiddlings in post- destroot obsolete.

#38508 RFE: add version number support to `platforms` field in portfiles larryv@… enhancement closed Normal
Description

Thought of this is #38506; not sure if there's already a ticket open for it...

Anyway, currently, if a port is incompatible with a certain OS version, the portfile author has to write something like this:

pre-fetch {
    if {${os.platform} == "darwin" && ${os.major} < 9} {
        ui_error "${name} is only supported on OS X 10.5 Leopard or later."
        return -code error "unsupported platform version"
    }
}

It'd be much easier to just be able to write something likeplatforms darwin 9+ or something like that instead.

#11563 RFE: adduser should be done the same way as startupitems macports-tickets@… enhancement closed Normal
Description

set to yes and define required variables instead of adding in pre-destroot / pre-install in Portfiles.

#14777 RFE: Add Tcl variant to db44 macports-tickets@… enhancement closed Normal
Description

Attached is a proposal to include a Tcl variant in db44.

I can not figure out why the static libraries were deleted in post-destroot
so I removed them as well.

#10633 RFE: add PDO::sqlite driver to PHP 5.1 port enhancement closed Low
Description

Hi,

After discussion with Ryan Schmidt, to whom this ticket is destined, he decided that maybe it would be a good idea to include sqlite driver to PDO in PHP 5.1 port.

To do so Ryan, you just have to add '--with-pdo-sqlite' to the configure line. Everything is included in the PHP 5.1 source, so there is no need for a variant or another port, you could enable it by default, the overhead is minimal.

To test if the driver was enabled after the "make install", you can run this : $ /opt/local/bin/php -r "phpinfo();" | grep "PDO drivers.*sqlite.*sqlite" | wc -l

Which will give you 0 if the driver was not installed, 1 if it was. I'm not sure this is something you can do, and maybe it's not the most surefire way to test for the availability, but that's how I do it ;)

Anyway, thanks for the help.

#5525 RFE: add "package.resources" Portfile key macports-tickets@… enhancement new Normal
Description

At long last, I've modified the self-building DarwinPorts port to include a path-patching postflight script, as well as a ReadMe and License. Note that this requires the packages.resources patch in "Depends On", which may mean it needs to wait for 1.2. But, hopefully this will help the DP maintainers build that!

#17082 RFE: Add livecheck to mpfr vincent-opdarw@… enhancement closed Normal
Description

Attached is a proposal to add livecheck to mpfr.

#24123 RFE: Add license pseudo-portname macports-tickets@… enhancement closed Normal
Description

I've attached a patch that adds a license pseudo-portname.

#14913 RFE: add fortran variant to hdf5 mcalhoun@… enhancement closed Normal
Description

In the recent update to hdf5, r35665, a variant was added which builds the Fortran interface.
In the same variant are optimization options, which do not work yet.
Attached is a patch which would allow the fortran interface to be built (without the problematic optimizations).

#17809 RFE: Add correct dependencies for use_* in configure raimue@… enhancement closed Normal
Description

There are some options available to run configuration script generators before configure.

  • use_autoconf
  • use_autoreconf
  • use_automake
  • use_xmkmf

All of them require tools to be provided by the system or to be installed by MacPorts.

Currently, the dependencies are added in each Portfile using such an option, but it would be better to do it automatically.

#17972 RFE: Add a PortGroup with a different universal build mechanism mcalhoun@… enhancement closed Normal
Description

To facilitate the building of more 32/64-bit universal binaries,
I would like to propose a PortGroup which builds the packages
separately and then merges them together.

It is intended to be an improvement of the merge function.

#37808 RFE: Add a "minor" variable to the perl portgroup macports-tickets@… enhancement closed Normal
Description

So that in a portfile, I can use something like this:

${perl5.major.minor}

to get an output of something like this:

5.12.4
Note: See TracQuery for help on using queries.