Re: update for print/lilypond

[Graeme Lee]

On 23-May 6:53, Matthias Kilian wrote:

[resent to ports@]

Hi,


Based on your diffs, I end up with the diff below. For now, only
test-built and packaged on amd64; I'll do some testing with some
of my old .ly fikes next weekend. Testbuilds (and real functional
tests) on other platforms (at least i386) are welcome, of course.

Some more notes:

- make port-lib-depends-check compains about some extra libraries
   (gmodule-2.0.4000 gthread-2.0.4000 iconv.6), but running ldd(1) on the
   lilypond executable lists them, so I keep them in WANTLIB-main.

- there three warnings about %d format strings used for
   size_t arguments; I did *not* add appropriate patches for now,
   because at least in one case using %zd would be wrong (that check
   against the the SCM object 'max-markup-depth'). I'll have to talk to
   upstream about this.

Ciao,
Kili



I've tested it now on both i386 and amd64 platforms

I ran a 50+ page score and a 30+ page Violin Etude book (among other 
things) on both platforms running current as of the 22nd of May.  Of 
note is that the amd64 platform uses double the memory footprint to 
compile the document and hits the default ulimit on larger works.


Thanks!

Graeme

update for print/lilypond

[Graeme Lee]

Hello

I'd like to submit a (mostly) updated the port for print/lilypond to 
from 2.14.2 to 2.18.2.


This is my first go at ports.  What is the recommended way to submit the 
updated port?  Would a new tarball be ok?  Or are patches to individual 
files preferred?  The port is based on the 2.14.2 Makefile.  I have not 
updated the lilypond-test-baseline yet


tested on amd64

Graeme

Re: update for print/lilypond

[Graeme Lee]

On 13-May 8:33, Matthias Kilian wrote:
Could anyone with enough spare time try to - run make configure on 
i386 - run make configure on whatever platform still uses gcc-3.3? [...]

A cvs diff is preferred; use 'cvs add' and 'cvs rm' to add/remove files,
then 'cvs diff -uNp'. Make sure the maintainer is included in to/cc.


I've attached a diff.

I hope I'm on the right track here...

Almost, but there's no need to package a diff in a .tar.gz ;-)

For ease of use, i'm just pasting your diff plain at the end of
this mail.

I'll give it a try during the next days; for now, just some comments on
*some* parts of your diff...


-REVISION-main =1
+# REVISION-main =  1

Don't comment unneeded lines, just remove them.


-   --with-ncsb-dir=${LOCALBASE}/share/ghostscript/fonts
+   --with-ncsb-dir=${LOCALBASE}/share/ghostscript/fonts \
+   --with-python-include=/usr/local/include/python2.7

In general, don't hard-code /usr/local, use ${LOCALBASE} instead. In
this special case (python), use ${MODPY_INCDIR} which is defined
automagically because lilypond uses MODULES = lang/python.

Some more comments in a few days (I'll also have to look wether the
sprintf-snprintf patch removals are ok).
Oh bother!  I see what I have overlooked.  I will put them back 
(updated, as the files have changed)



Ciao,
Kili

full diff extracted from the lilypond.tar.gz Graeme sent in his
mail:

Re: NEW: security/yersinia

[graeme]

I have been using this on i386. Yersinia works but only if I first bring
down interface pflog0 otherwise yersinia core dumps.

Its not a big issue for me but I was wondering if there was a fix I could
put in for this?

cyrus port request

[Graeme Lee]

Hi ports.

Can we please consider adding --enable-replication to the cyrus build?  
It was considered 'experimental' in 03, and was planned to be deployed 
by CMU by 07.  It's 2009 now.


Thanks,

Graeme

Re: Squid default file descriptors number changed?

[Graeme Lee]

Alexey Suslikov wrote:

Hello [EMAIL PROTECTED]

Looks like default file descriptors number changed with Squid 2.7.


max_filedescriptors 1024 in /etc/squid/squid.conf remedies the problem but
this new default in 2.7 is really strange.


So maybe number of filedescriptors was hardcoded in 2.6 to 1024 but it is
not true for 2.7 any more?
  
Check your login.conf for the daemon class.  openfiles-cur will probably 
be 128.  Maybe

Thanks.

Alexey

  

Re: update: net/net-snmp

[Graeme Lee]

Rui Reis wrote:

here's an update to net/net-snmp. Changelog available here:
http://net-snmp.sourceforge.net/about/ChangeLog.html

I have sent this diff to Tim Kornau (maintainer) but I still didn't
get an answer.

This diff also adds a patch 
(patch-agent_mibgroup_if-mib_data_access_interface_sysctl_c) from claudio@ to 
allow 64bit network counters.


works for me on i386, amd64, macppc and sparc64.

ok?

cheers,
rui
  

All good on amd64.  No longer chewing up 1mb per day on my client.

Net-snmp 5.4 memory leaks

[Graeme Lee]

Are there any plans for updating net-snmp to 5.4.1 in the ports tree?  
5.4 has memory leaks.  Really bad ones.