Re: aarch64 fix for www/webkit

2018-06-01 Thread Landry Breuil
On Fri, Jun 01, 2018 at 08:54:28PM +0200, Peter Hessler wrote: > Ping Well.. i guess i386 and amd64 archs wont care, but what about macppc and sparc64 ? (if they ever build it.. seems not). i see there's a mips64el package for webkit... Did you build it on !arm64 ?

NEW: sysutils/truncate

2018-06-01 Thread Brian Callahan
Hi ports -- Attached is a new port, sysutils/truncate, a utility that allows you to truncate or extend the length of files. It is a port of the truncate(1) utility from DragonFly BSD. Version number comes from the DFly release I pulled the code from. --- pkg/DESCR: The truncate utility adjusts

ruby 2.6.0-preview2

2018-06-01 Thread Jeremy Evans
ruby 2.6.0-preview2 was released yesterday. Release announcement at: http://www.ruby-lang.org/en/news/2018/05/31/ruby-2-6-0-preview2-released/ Attached is a port in case anyone wants to do early testing of the new JIT compiler (enabled with --jit option). Thanks, Jeremy ruby26.tar.gz Descripti

hitch: Fix permissions, add cpath promise for OCSP worker

2018-06-01 Thread Klemens Nanni
Noticed by Maxim Tarasov , /var/hitch/ should be owned by _hitch not root. I failed to spot this as the directory had proper ownership and permissions on my system already during porting and testing. Hitch may also create new OCSP staples, I only tested and run with existing ones or none at all.

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Stuart Henderson
On 2018/06/01 22:07, Stefan Sperling wrote: > On Fri, Jun 01, 2018 at 09:03:08PM +0100, Stuart Henderson wrote: > > On 2018/06/01 15:35, Daniel Jakots wrote: > > > On Fri, 1 Jun 2018 18:46:25 +0100, Stuart Henderson > > > wrote: > > > > > > > Thanks - this is a common problem with many PHP apps r

Re: update x11/stumpwm

2018-06-01 Thread Jeremie Courreges-Anglas
On Thu, May 31 2018, Solene Rapenne wrote: > Klemens Nanni writes: > >> On Thu, May 31, 2018 at 09:48:11AM +0200, Solene Rapenne wrote: >>> A new version of StumpWM has been released, the versioning changed from >>> a regular major.minor to YY.MM so new version is 18.05. It requires the >>> extra

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Stefan Sperling
On Fri, Jun 01, 2018 at 09:03:08PM +0100, Stuart Henderson wrote: > On 2018/06/01 15:35, Daniel Jakots wrote: > > On Fri, 1 Jun 2018 18:46:25 +0100, Stuart Henderson > > wrote: > > > > > Thanks - this is a common problem with many PHP apps rather than being > > > specific to nextcloud/owncloud. R

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Stuart Henderson
On 2018/06/01 15:35, Daniel Jakots wrote: > On Fri, 1 Jun 2018 18:46:25 +0100, Stuart Henderson > wrote: > > > Thanks - this is a common problem with many PHP apps rather than being > > specific to nextcloud/owncloud. Rather than repeating it all over the > > place I've just added a section to th

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Daniel Jakots
On Fri, 1 Jun 2018 21:44:57 +0200, Stefan Sperling wrote: > What I did is put /var/www/dev on a tiny read-only mfs which is > good enough for a random device node. Neat! > fstab: swap /var/www/dev mfs ro,nosuid,-s512K,-P/var/www/dev.tmpl 0 0 > > $ ls -l /var/www/dev.tmpl

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Stefan Sperling
On Fri, Jun 01, 2018 at 03:35:51PM -0400, Daniel Jakots wrote: > Great! Would you mind documenting the creation of *random devices? > Feel free to take the diff and commit on my behalf if you agree (I'm > sure you'll want to change some wordings ;)). I'm don't know if we need > to document for null

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Daniel Jakots
On Fri, 1 Jun 2018 18:46:25 +0100, Stuart Henderson wrote: > Thanks - this is a common problem with many PHP apps rather than being > specific to nextcloud/owncloud. Rather than repeating it all over the > place I've just added a section to the pkg-readme for PHP that > explains this. Great! Wou

Re: aarch64 fix for www/webkit

2018-06-01 Thread Peter Hessler
Ping On 2018 May 27 (Sun) at 00:42:03 +0200 (+0200), Peter Hessler wrote: :This fixes the build on aarch64. aarch64 has libc++.so instead of :libstdc++.so, and libtool fails to find the missing library. : :OK? : :Index: www/webkit/patches/patch-GNUmakefile_in :===

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Bruno Flueckiger
I've learned to bump the REVISION in such diffs. So here is my second try, this time including the bump. As I am new to this, any feedback is appreciated. Cheers, Bruno Index: www/nextcloud//Makefile === RCS file: /cvs/ports/www/next

Re: Nextcloud - cert validataion in chroot

2018-06-01 Thread Stuart Henderson
Thanks - this is a common problem with many PHP apps rather than being specific to nextcloud/owncloud. Rather than repeating it all over the place I've just added a section to the pkg-readme for PHP that explains this. On 2018/06/01 14:48, Bruno Flueckiger wrote: > Hi ports@, > > The News app in

Re: UPDATE: editors/vim

2018-06-01 Thread Rafael Sadowski
On Wed May 30, 2018 at 12:26:18PM +0100, Stuart Henderson wrote: > On 2018/05/29 22:02, Rafael Sadowski wrote: > > Simple update to the latest stable version. > > > > Tested with 8.1.0026-gtk2-perl-python3-ruby on amd64. > > > > :termin runs fine with gvim. > > > > Index: Makefile > > ==

Nextcloud - cert validataion in chroot

2018-06-01 Thread Bruno Flueckiger
Hi ports@, The News app in Nextcloud throws an error message about the server certificate when you try to subscribe to a feed with https. The reason is the missing list of trusted CAs in the chroot(2). The following patch describes the solution for this. Cheers, Bruno Index: www/nextcloud/pkg/R

PHP fix session randomness

2018-06-01 Thread Martijn van Duren
Currently only php5.6 has a session patch and the generated entropy isn't added to the session id. The following patch fixes that and adds support to php7. martijn@ Index: 5.6/patches/patch-ext_session_session_c === RCS file: /cvs

recent update-plist changes

2018-06-01 Thread Marc Espie
It's still churning around. Loads of small recent changes. A summary of what I'm remembering. - the newuser/newgroup checks found quite a few (6 or 7) ports which had forgotten to register their users in user.list. This is actually quite bad from a security standpoint, as it means reusing the sa

Re: PHP cleaner patch for preventing libodbc linking to SAPI

2018-06-01 Thread Stuart Henderson
On 2018/06/01 10:02, Stuart Henderson wrote: > On 2018/06/01 09:24, Martijn van Duren wrote: > > On 05/31/18 15:25, Stuart Henderson wrote: > > > On 2018/05/31 13:53, Martijn van Duren wrote: > > >> The current patch removes the check altogether, this patch adds an > > >> extra check to make sure l

Re: PHP cleaner patch for preventing libodbc linking to SAPI

2018-06-01 Thread Stuart Henderson
On 2018/06/01 09:24, Martijn van Duren wrote: > On 05/31/18 15:25, Stuart Henderson wrote: > > On 2018/05/31 13:53, Martijn van Duren wrote: > >> The current patch removes the check altogether, this patch adds an > >> extra check to make sure libodbc isn't linked to the SAPIs. > >> > >> This is sim

Re: PHP cleaner patch for preventing libodbc linking to SAPI

2018-06-01 Thread Martijn van Duren
On 05/31/18 15:25, Stuart Henderson wrote: > On 2018/05/31 13:53, Martijn van Duren wrote: >> The current patch removes the check altogether, this patch adds an >> extra check to make sure libodbc isn't linked to the SAPIs. >> >> This is similar to how other libraries setup the >> PHP_ADD_LIBRARY_W

Re: [BUG] Inkscape

2018-06-01 Thread Antoine Jacoutot
On Thu, May 31, 2018 at 11:59:52PM -0300, Elias M. Mariani wrote: > Following on this bug, the problem seems to be fixed upstream, but > they didn't update the .xz package of glib2mm, I attach in this mail a > diff with a patch for glib2mm that fixes the problem, I got it from > MacPorts. > Tested

Re: PHP force res_search even stronger

2018-06-01 Thread Martijn van Duren
On 05/31/18 15:24, Stuart Henderson wrote: > On 2018/05/31 14:50, Martijn van Duren wrote: >> Right now if libresolv or libsocket is picked up somewhere they are >> linked in. > > Huh? "If some library that we don't have on OpenBSD is picked up ..." > It reduces the size of the patch, makes it c