x11/i3status: Recover from sndiod restart

On occasion I restart sndiod which causes i3status to stop reporting the volume level. The below patch resets initialized after a disconnect so on the next poll i3status will attempt to reconnect to the new sndiod. I opted against resetting initialized in cleanup so a failure in init won't be retri

Re: rust-{clippy,rustfmt} missing dependency on rust

On Thu, Sep 12, 2024 at 12:35:28PM +0100, Stuart Henderson wrote: > On 2024/09/12 08:28, Sebastien Marie wrote: > > Matthew Martin writes: > > > > > I'm on a monthish old snapshot. While there's been a bump to libcrypto > > > and libssl in the meantime

rust-{clippy,rustfmt} missing dependency on rust

I'm on a monthish old snapshot. While there's been a bump to libcrypto and libssl in the meantime, pkg_add -u usually doesn't break anything. On a recent pkg_add -u, Can't install rust-1.81.0 because of libraries Direct dependencies for rust-1.80.1->1.81.0 resolve to libssh2-1.11.0 curl-8.9.1p0

update: shells/zsh-syntax-highlighting

Release today. I've been running the moral equivalent of the release for a month or so, so don't anticipate any issues. diff --git Makefile Makefile index b4e34f0af24..8cabaa8e673 100644 --- Makefile +++ Makefile @@ -1,8 +1,6 @@ COMMENT = fish like syntax highlighting for zsh -GH_ACCOUNT =

Update devel/mercurial to 6.2

Not a mercurial user, but passes a few quick checks. Tests fail in the same way as they did for 5.8. Shell completers are installed correctly now. diff --git Makefile Makefile index 28ad010a64c..022cc404225 100644 --- Makefile +++ Makefile @@ -1,12 +1,10 @@ COMMENT-main = fast, lightweig

login_krb5 build failure

I don't use login_krb5, but the error looks simple enough. cc -o login_krb5 login.o login_krb5.o $(/usr/local/heimdal/bin/krb5-config --deps --libs) -lutil ld: error: duplicate symbol: __progname >>> defined at /usr/lib/crt0.o:(__progname) diff --git patches/patch-login_krb5_logi

Re: Fix build of shells/zsh for sparc64

On Mon, May 30, 2022 at 04:45:16PM -0400, Kurt Mosiejczuk wrote: > The patches for the 5.9 update which works fine for clang, but not for > either base-gcc or ports-gcc. The patch to zpty.c uses #elifdef which > clang knows, but wasn't introduced to gcc until 12.x. > > Better to use #elif defined,

Re: shells/zsh: Update to 5.9

ping; still running without issues here On Mon, May 16, 2022 at 07:46:20AM -0500, Matthew Martin wrote: > zsh 5.9 was just announced[1]. I've been running the below for about > a day and the prior release candidate for about a week before that > without issues. > > This up

shells/zsh: Update to 5.9

ll go away next release. The other test failure is known and because regexec(3) is not unicode aware. - Matthew Martin 1: https://www.zsh.org/mla/announce/msg00134.html diff --git Makefile Makefile index 376e8298c38..b6ee32c4dfc 100644 --- Makefile +++ Makefile @@ -1,6 +1,6 @@ COMMENT=

shells/zsh and _pbuild can't open /dev/ttyxx

zsh is nearing a release and since OpenBSD is too, I thought I'd get some early testing in. (Though at this point the timing won't work out.) A pty test has been failing for a while, but a newly added test is now hanging on [1] ret = read(master, &syncch, 1); I believe the root cause basic

shells/zsh: Update to 5.8.1 for CVE-2021-45444

zsh 5.8.1 was released today. make test fails in the same way it did for 5.8. This is a stable security release with a few bug fixes, including one for CVE-2021-45444, a vulnerability in prompt expansion which could be exploited through e.g. VCS_Info to execute arbitrary shell commands without a u

devel/ninja: Fix shell completion

Install the zsh completion function to site-functions. From other ports share/bash-completion/completions seems to be the right location, but someone should double check me on that. diff --git Makefile Makefile index 7c5972e0079..527dc13ef33 100644 --- Makefile +++ Makefile @@ -10,7 +10,7 @@ COMM

audio/pulseaudio: Fix zsh completion

Third party zsh completers should be installed to zsh/site-functions[1]. Since that's what pulseaudio does by default, just remove the flag. 1: https://github.com/zsh-users/zsh/blob/master/INSTALL#L257 diff --git Makefile Makefile index a1863c1ee46..2eb0f4eccc5 100644 --- Makefile +++ Makefile @

Re: shells/zsh: Keep $OSTYPE up to date

On Sat, Mar 27, 2021 at 09:36:24PM +0100, Christian Weisgerber wrote: > Matthew Martin: > > > A while back [1] it was noted zsh's $OSTYPE is out of sync with the OS > > version. > > > > 1: https://marc.info/?l=openbsd-ports&m=160940571013139&w=2 &g

shells/zsh: Keep $OSTYPE up to date

A while back [1] it was noted zsh's $OSTYPE is out of sync with the OS version. The below patch uses $(OSrev) to automatically bump REVISION when the OS version changes. In the case a normal REVISION bump is needed, the constant can be decremented. When updated to a new release, the constant should

Re: quirks botched commit

On Thu, Feb 25, 2021 at 10:16:03AM +, Mikolaj Kucharski wrote: > On Thu, Feb 25, 2021 at 11:04:06AM +0100, Marc Espie wrote: > > On Thu, Feb 25, 2021 at 09:53:04AM +0100, Marc Espie wrote: > > > next time, please remember to build it. > > > there's a reason do-build checks the syntax of those p

Re: [update] devel/libtermkey

On Sun, May 31, 2020 at 10:12:47PM +0100, Edd Barrett wrote: > Hi Matthew, > > On Sun, May 31, 2020 at 11:11:59AM -0500, Matthew Martin wrote: > > ping > > This and you unibilium diff look fine to me. One small comment below > (applicable to both). > > > &g

New: devel/libtickit

libtermkey is deprecated for new applications in favor of libtickit; however, currently nothing in the ports tree will use it. I'm using libtickit for a personal project, but can maintain it in my tree until neovim or another port uses it. This library provides an abstracted mechanism for building

Re: [update] devel/libtermkey

ping On Sat, May 23, 2020 at 08:12:11PM -0500, Matthew Martin wrote: > No change in symbols and the diff doesn't seem to touch any public > functions. Only bumped minor because upstream did. > > neovim and vis run fine with this. The new patch allows the vis test > suit

Re: [update] devel/unibilium

ping On Sat, May 23, 2020 at 08:11:32PM -0500, Matthew Martin wrote: > Update to unibilium. No change in symbols and the diff doesn't seem to > touch any public functions. Only bumped minor because upstream did. > > Tests pass and libtermkey (up next), neovim, and vis seem ok.

Re: New: zsh-syntax-highlighting

On Fri, May 15, 2020 at 06:57:37AM +0200, Rafael Sadowski wrote: > On Tue Apr 21, 2020 at 06:35:38AM +0200, Rafael Sadowski wrote: > > On Sat Apr 11, 2020 at 08:16:55AM +0200, Rafael Sadowski wrote: > > > On Fri Apr 10, 2020 at 09:11:25AM -0500, Matthew Martin wrote: > >

[update] devel/libtermkey

No change in symbols and the diff doesn't seem to touch any public functions. Only bumped minor because upstream did. neovim and vis run fine with this. The new patch allows the vis test suite to run without the TEST_ENV = TERM=${TERM} hack. I'll ask upstream about adding the patch, but libtermkey

[update] devel/unibilium

Update to unibilium. No change in symbols and the diff doesn't seem to touch any public functions. Only bumped minor because upstream did. Tests pass and libtermkey (up next), neovim, and vis seem ok. 2.1.0 2020-02-11 - Makefile: use "-f" with prove always (to show failed tests) (#13)

Re: github mirror

On Sun, Apr 19, 2020 at 09:32:21PM +0200, Rafael Sadowski wrote: > Use your own *global* gitignore. For more details see core.excludesFile: > https://git-scm.com/docs/gitignore Rather than the global gitignore, you can use the local repo-specific $GIT_DIR/info/exclude

Re: New: zsh-syntax-highlighting

On Fri, Apr 10, 2020 at 09:13:43AM +0200, Rafael Sadowski wrote: > On Thu Apr 09, 2020 at 09:43:36PM -0500, Matthew Martin wrote: > > Small port for zsh-syntax-highlighting. Currently have it in shells, but > > more than happy to move it if someone has a better suggestion. To test

New: zsh-syntax-highlighting

Small port for zsh-syntax-highlighting. Currently have it in shells, but more than happy to move it if someone has a better suggestion. To test source /usr/local/share/zsh-syntax-highlighting/zsh-syntax-highlighting.zsh after installing. DESCR: This package provides syntax highlighting for the

Re: UPDATE: shells/zsh

On Sat, Mar 28, 2020 at 09:46:56AM +0100, Rafael Sadowski wrote: > On Sat Mar 14, 2020 at 04:31:00PM -0500, Matthew Martin wrote: > > On Sat, Mar 14, 2020 at 08:21:10PM +0100, Rafael Sadowski wrote: > > > "Security and bug fix release with a few user visible additions.

Re: UPDATE: shells/zsh

On Sat, Mar 14, 2020 at 08:21:10PM +0100, Rafael Sadowski wrote: > "Security and bug fix release with a few user visible additions." > Changelog: http://zsh.sourceforge.net/releases.html > > This release fixes CVE-2019-20044. (Not tested on OpenBSD) > > OK? Should it go into -stable without the @

Re: git: Adding git-prompt and bash/tcsh/zsh completions

On Mon, Feb 10, 2020 at 07:42:11PM +0100, Johannes Kastl wrote: > Hi Matthew, > > thanks for the feedback. > > On 2020-02-10 19:10 Matthew Martin wrote: > > On Mon, Feb 10, 2020 at 04:56:54PM +0100, Johannes Kastl wrote: > >> Hi all, > >> > >&

Re: git: Adding git-prompt and bash/tcsh/zsh completions

default. The one in git's tree seems to be a shim to let zsh use the bash completer which isn't as detailed. I have no opinion on the completion files for other shells. It may be nice to have a standardized place for bash completions, but I'll leave that to people who use the shell. - Matthew Martin

Update: textproc/zathura

Builds fine. I don't use it, so I'd appreciate if a user could test it still works for them. Requires the girara update just sent. zsh tab completion works without fpath modification. diff --git Makefile Makefile index b85e1d4fdad..71df90927f6 100644 --- Makefile +++ Makefile @@ -1,6 +1,6 @@ # $

Update: x11/girara

Required to update zathura. Reviewing the diff at https://git.pwmt.org/pwmt/girara/compare/0.3.2...0.3.4 there was a symbol addition (girara_xdg_open_with_working_directory) and a few exported symbols had a pointer arg made const. I think a minor bump suffices here. Only consumer seems to be zathu

sysutils/exa: Fix zsh completion

With this exa completes with no fpath changes. diff --git Makefile Makefile index 7bead97a053..83f06082546 100644 --- Makefile +++ Makefile @@ -5,6 +5,7 @@ COMMENT = ls alternative written in Rust GH_ACCOUNT = ogham GH_PROJECT = exa GH_TAGNAME = v0.9.

productivity/khard: Fix zsh completion

With this khard tab completes with no fpath changes. diff --git Makefile Makefile index c5139bd0b9a..d477a35fdcc 100644 --- Makefile +++ Makefile @@ -4,6 +4,7 @@ COMMENT = console CardDAV client MODPY_EGG_VERSION =0.15.1 DISTNAME = khard-${MODPY_EGG_VERSION} +REVI

multimedia/mpv: Fix zsh completion

mpv by default installs its zsh completer to site-functions, so just remove the --zshdir flag. With this change mpv tab completes with no fpath changes. diff --git Makefile Makefile index 9e483bde787..0e4eb3371e7 100644 --- Makefile +++ Makefile @@ -5,6 +5,7 @@ COMMENT = movie playe

Re: sysutils/fzf: install shell support files

On Thu, Jan 16, 2020 at 05:28:49PM +, Stuart Henderson wrote: > On 2020/01/16 18:05, Paco Esteban wrote: > > Hi Edd, > > > > On Thu, 16 Jan 2020, Edd Barrett wrote: > > > > > Hi, > > > > > > This diff packages the necessary support files for for integrating fzf > > > with shells. > > > > >

Re: sysutils/fzf: install shell support files

On Thu, Jan 16, 2020 at 03:49:43PM +, Edd Barrett wrote: > Hi, > > On Thu, Jan 16, 2020 at 08:47:43AM -0600, Matthew Martin wrote: > > On Thu, Jan 16, 2020 at 03:32:36PM +0100, Klemens Nanni wrote: > > > On Thu, Jan 16, 2020 at 02:20:09PM +, Edd Barrett w

Re: sysutils/fzf: install shell support files

On Thu, Jan 16, 2020 at 03:32:36PM +0100, Klemens Nanni wrote: > On Thu, Jan 16, 2020 at 02:20:09PM +, Edd Barrett wrote: > > Hi, > > > > This diff packages the necessary support files for for integrating fzf > > with shells. > > > > With this change, enabling support in (e.g.) zsh is as simp

[patch] net/icbirc: Correct IRC parsing

those commands and I shouldn't break finger memory. I don't use ICB, so this has only been compile tested. Once tested, this should go upstream rather than living in the tree. - Matthew Martin diff --git Makefile Makefile index d7bb3f1160b..a7b292d43da 100644 --- Makefile +++ Makefi

Re: update: shells/zsh 5.7.1

ping On Mon, Feb 4, 2019 at 5:59 PM Matthew Martin wrote: > > Bump zsh to 5.7.1 (5.7 had a bug in vcs_info). Tests pass (just the > usual failure in D07) and it's running fine here. > > Switch to CONFIGURE_STYLE=gnu since there are no autoconf patches > (reverting a

Re: update: shells/zsh 5.7.1

On Thu, Feb 14, 2019 at 10:23 AM Pierre-Emmanuel Andre wrote: > > On Mon, Feb 04, 2019 at 05:59:35PM -0600, Matthew Martin wrote: > > Bump zsh to 5.7.1 (5.7 had a bug in vcs_info). Tests pass (just the > > usual failure in D07) and it's running fine here. > > >

update: shells/zsh 5.7.1

Bump zsh to 5.7.1 (5.7 had a bug in vcs_info). Tests pass (just the usual failure in D07) and it's running fine here. Switch to CONFIGURE_STYLE=gnu since there are no autoconf patches (reverting a change from 2004). The generated configure does indeed match what we'd generate. Drop PORTHOME and T

Re: UPDATE: shells/zsh

E}/share/doc/zsh/html/$V \ I don't see what problem this change fixes; if there isn't a problem, why rearrange the chairs? Also 5.6.2 is due out tomorrow, so should probably wait for that now. - Matthew Martin

Re: UPDATE: shells/zsh

on. This is the upstream default, and I don't see a compelling reason to deviate from it. Patch for 5.6.1 below. It was rolled a bit over an hour ago, so not well tested yet. Obviously we should wait until it's officially announced. There's one test error over what 5.5.1 had whi

Re: [patch] geo/garmindev on aarch64

On Wed, Jul 11, 2018 at 12:43:35PM +0100, Stuart Henderson wrote: > On 2018/07/10 23:16, Matthew Martin wrote: > > Few signed char fixes. > > Was this a build failure before (in which case OK), or did it build but > misbehave (in which case it needs a bump)? Build failure [1].

[patch] geo/garmindev on aarch64

Few signed char fixes. - Matthew Martin diff --git patches/patch-src_EtrexH_CDevice_cpp patches/patch-src_EtrexH_CDevice_cpp new file mode 100644 index 000..4b0c1a0f437 --- /dev/null +++ patches/patch-src_EtrexH_CDevice_cpp @@ -0,0 +1,14 @@ +$OpenBSD$ + +Index: src/EtrexH/CDevice.cpp

[patch] security/libnipper on aarch64

Two signed char fixes. - Matthew Martin diff --git patches/patch-device_crypto_crypto-base64_cpp patches/patch-device_crypto_crypto-base64_cpp index 081341a572d..68abdfc8951 100644 --- patches/patch-device_crypto_crypto-base64_cpp +++ patches/patch-device_crypto_crypto-base64_cpp @@ -1,8 +1,27

[patch] games/spacehulk on aarch64

A signed char fix and fix two obvious typos pointed out by clang. - Matthew Martin diff --git Makefile Makefile index 13dda6c9e91..8cc6733b52f 100644 --- Makefile +++ Makefile @@ -4,7 +4,7 @@ COMMENT=total conversion of the Space Hulk board game DISTNAME= spacehulk-1.5-beta1

Re: [patch] devel/bullet on aarch64

ping On Mon, Jun 25, 2018 at 10:13:10PM -0500, Matthew Martin wrote: > I don't use bullet, but fixes the build with CXXFLAGS=3D-funsigned-char >=20 > - Matthew Martin >=20 >=20 > diff --git patches/patch-examples_ThirdPartyLibs_Gwen_Renderers_Fon

[patch] devel/bullet on aarch64

I don't use bullet, but fixes the build with CXXFLAGS=3D-funsigned-char - Matthew Martin diff --git patches/patch-examples_ThirdPartyLibs_Gwen_Renderers_FontData_h = patches/patch-examples_ThirdPartyLibs_Gwen_Renderers_FontData_h new file mode 100644 index 000..9b91b0d1a3c --- /dev

[patch] editor/ht on aarch64

]. I think the crank can be left out since the binary will be identical on signed char arches and fixes the build on unsigned char arches. - Matthew Martin 0: http://build-failures.rhaalovely.net//aarch64/2018-06-21/editors/ht.log 1: https://github.com/sebastianbiallas/ht/pull/21 diff --git p

update multimedia/mlt 6.6.0

ranks the major. - Matthew Martin diff --git Makefile Makefile index f85a35f6c36..65519303346 100644 --- Makefile +++ Makefile @@ -2,19 +2,17 @@ COMMENT-main = multimedia transformations framework COMMENT-gpl2 = GPLv2-licensed modules for mlt -VERSION =

zsh 5.5.1

Small patch to zsh 5.5.1. Also I noticed the two patches removed for 5.5 are now zero byte files. Anything git users can do to make that easier (or should patch(1) be modified to recognize +++ /dev/null as rm the file)? - Matthew Martin diff --git Makefile Makefile index af6389a94f6

Casting bitmaps (clang 6)

the latter case should all bytes be cast or just the ones >= 0x80? - Matthew Martin

clang 6 fix java/jlint

Fix format specifier, add spaces, and use negative literals when negative numbers are desired (with a cast for -0x8000). - Matthew Martin diff --git patches/patch-jlint_cc patches/patch-jlint_cc index 1bf4f950634..567174b30e4 100644 --- patches/patch-jlint_cc +++ patches/patch-jlint_cc

Re: clang 6 fix devel/jdk/1.8

On Wed, Apr 11, 2018 at 08:27:34AM -0400, Kurt Miller wrote: > On Tue, 2018-04-10 at 21:23 -0500, Matthew Martin wrote: > > Add lots of whitespace and change a few types from signed to > > unsigned. > > Probably should get a REVISION bump, but I'm not sure where tha

Re: clang 6 fix security/steghide

On Wed, Apr 11, 2018 at 01:48:29PM +0200, Jeremie Courreges-Anglas wrote: > On Wed, Apr 11 2018, Giovanni Bechis wrote: > > On 04/11/18 05:16, Matthew Martin wrote: > >> On Tue, Apr 10, 2018 at 04:44:56PM +0200, Giovanni Bechis wrote: > >>> test.log run on 6.3 attach

Re: clang 6 fix security/steghide

On Tue, Apr 10, 2018 at 04:44:56PM +0200, Giovanni Bechis wrote: > test.log run on 6.3 attached, does it differs from yours ? > Cheers > Giovanni Not meaningfully. Thanks for the log. - Matthew Martin

clang 6 fix security/steghide

Seems easier to make the C++11 fixes rather than patch the makefiles to compile for C++03. There's a test failure, but I don't know if it's new. - Matthew Martin diff --git Makefile Makefile index 78a8c328b25..2cc5016be00 100644 --- Makefile +++ Makefile @@ -3,7 +

update: math/z3

Update z3; builds here with clang 6. Enable tests while here. - Matthew Martin diff --git Makefile Makefile index 6fe70abc40a..b32659d5940 100644 --- Makefile +++ Makefile @@ -2,9 +2,8 @@ COMMENT = z3 theorem prover -VERSION = 4.5.0 +VERSION = 4.6.0 DISTNAME = z3

clang 6 fix astro/celestia

Neither of these changes are in the code on github because these chunks of code are no where to be found (even in history). Regardless builds and runs here now. - Matthew Martin diff --git Makefile Makefile index 5d8943d5330..2b18df5a1af 100644 --- Makefile +++ Makefile @@ -3,7 +3,7 @@ COMMENT

update: zsh 5.5

zsh.texi patch seems to be useless: zsh-5.5/fake-amd64 is identical with and without the patch. - Matthew Martin diff --git Makefile Makefile index 36350def5ad..1fc94cb5666 100644 --- Makefile +++ Makefile @@ -2,7 +2,7 @@ COMMENT= Z shell, Bourne shell-compatible -V=

dvd+rw-tools and clang 6

Looks like clang just needs some space. Again only build tested. - Matthew Martin diff --git patches/patch-dvd+rw-mediainfo_cpp patches/patch-dvd+rw-mediainfo_cpp new file mode 100644 index 000..bf6310efc0f --- /dev/null +++ patches/patch-dvd+rw-mediainfo_cpp @@ -0,0 +1,40 @@ +$OpenBSD

apachetop and clang 6

Don't use apachetop, so just build tested. Looks like someone was overzealous with their NULL usage. Also fix what looks like a write past the end of the array. - Matthew Martin diff --git patches/patch-src_apachetop_cc patches/patch-src_apachetop_cc new file mode 100644 index 00

Re: update: shells/zsh 5.4.2

tch-Completion_BSD_Command__bsd_pkg @@ -1,12 +1,7 @@ $OpenBSD: patch-Completion_BSD_Command__bsd_pkg,v 1.4 2015/12/21 19:13:17 jasper Exp $ - -From 3272bdce987a3e5d5448688529ccb8290e148a67 Mon Sep 17 00:00:00 2001 -From: Matthew Martin -Date: Fri, 11 Dec 2015 11:51:09 +0100 -Subject: [PATCH] 37383: add complet

Re: update: shells/zsh 5.4.2

Ping (with pea@ this time and my apologies). Haven't run into issues in the past two months. - Matthew Martin On Sat, Sep 09, 2017 at 09:39:11PM -0500, Matthew Martin wrote: > Been running this for about a week now. The patch is only large because > I'd like to kill the nee

Re: Update to git-2.14.2

nBSD 6.2 (GENERIC.MP) #111: Tue Sep 26 11:09:09 MDT 2017. Similar test results and I see no difference in my usual workflow. - Matthew Martin

update: shells/zsh 5.4.2

Been running this for about a week now. The patch is only large because I'd like to kill the needless whitespace changes in the _bsd_pkg completion patch. I don't have a strong opinion on the matter if someone prefers to keep it as is. - Matthew Martin diff --git Makefile Make

Re: UPDATE: shells/zsh

ight remove the needless whitespace changes in patches/patch-Completion_BSD_Command__bsd_pkg (I have the diff locally if anyone wants it). If anyone knows how to get a list of all packages (installed and not) preferably without having a ports tree, I could fix up the patch to push it upstream. - Matthew Martin

Re: NEW PORT: systemd

On Mon, Jul 31, 2017 at 08:07:13PM +0200, Marc Espie wrote: > Port is attached. Maintainer name and email don't seem to match. - Matthew Martin

[patch] zsh 5.3

Update to zsh 5.3, mostly removing upstreamed patches. A regex test in D07 fails, but upstream believes[1] this is due to OpenBSD's locale support (or lack thereof). Been running for about a day. 1: http://zsh.org/mla/workers/2016/msg02607.html - Matthew Martin Index: Mak

Re: UPDATE: ledger 3.1.1

might > have. > > Regards, > Florian Compiles and works for my ledger file (but I started using ledger after your first mail). Only tests 1, 3, 4, and 5 pass for me; the other 339 fail. Is this expected? - Matthew Martin

[patch] time_t printf in zsh

zsh printf's a time_t as a long. The cast is so that I can send the same patch upstream (assuming that casting to long long is the correct fix). Bug found with clang. - Matthew Martin Index: patches/patch-Src_Builtins_sc

[update] zsh 5.2

Zsh 5.2 was released yesterday. All tests pass on amd64. - Matthew Martin Index: Makefile === RCS file: /cvs/ports/shells/zsh/Makefile,v retrieving revision 1.75 diff -u -p -r1.75 Makefile --- Makefile6 Nov 2015 14:17:03 -

Re: [update] zsh-5.1.1

Ping? I've been running this for over a month with no issues. On Sat, Sep 12, 2015 at 12:04:25AM -0500, Matthew Martin wrote: > zsh has already released a 5.1.1. This is a one line change from the > final diff in the 5.1 thread. > > The failing test (D07) was added between

new: shells/zsh-syntax-highlighting

SEPARATE_BUILD for a port that only copies files? - Matthew Martin zsh-syntax-highlighting.tgz Description: application/tar-gz

[update] zsh-5.1.1

paces after ZSH). I have still yet to see any other failing tests on amd64. - Matthew Martin Index: shells/zsh/Makefile === RCS file: /cvs/ports/shells/zsh/Makefile,v retrieving revision 1.74 diff -u -p -r1.74 Makefile --- shells/zs

Re: [update] zsh-5.1

fail sometimes, but I can't > reproduce that reliably. I haven't run into it yet (even with the many times I reran the tests for the chgrp bug), but I'll keep an eye out. - Matthew Martin

Re: [update] zsh-5.1

On Wed, Sep 02, 2015 at 07:26:35AM +0200, Theo Buehler wrote: > On Tue, Sep 01, 2015 at 08:55:36PM -0500, Matthew Martin wrote: > > Patch to update zsh to 5.1. Fairly boring mostly deleting upstreamed > > patches, deleteing unneeded parts of patches, and adding the pre-test > &

[update] zsh-5.1

the top). It seems that double single quotes are rendered correctly even without the patch (and the patch missed a few anyway). - Matthew Martin Index: Makefile === RCS file: /cvs/ports/shells/zsh/Makefile,v retrieving revision 1.

Re: [patch] Test target PATH

On Fri, Aug 28, 2015 at 08:47:05PM +0100, Stuart Henderson wrote: > On 2015/08/27 22:04, Matthew Martin wrote: > > ${WRKDIR}/bin is only in PATH during extract, configure, build, fake, > > and test. Since fake creates the chown and chgrp symlinks, extract, > > configure, bui

Re: [patch] Test target PATH

On Thu, Aug 27, 2015 at 11:27:44PM -0400, Raul Miller wrote: > On Thu, Aug 27, 2015 at 11:04 PM, Matthew Martin wrote: > > On Thu, Aug 27, 2015 at 11:47:19PM +0200, Juan Francisco Cantero Hurtado > > wrote: > >> >---rm -f ${WRKDIR}/bin/chgrp > > > >

Re: [patch] Test target PATH

l believe that the first patch is correct as any ports it breaks has tests that depend on non-standard install(1) behavior caused by the perl wrapper so including ${WRKDIR}/bin in PATH is just concealing the breakage. - Matthew Martin Index: infrastructure/mk/bsd.port.mk ==

Re: [patch] Test target PATH

/bin/chgrp. When make fake is run, a symlink from ${WRKDIR}/bin/chgrp to /bin/echo is added. This resuls in make fake test using the chgrp in ${WRKDIR}/bin and the test failing since that chgrp doesn't change the group. I am proposing to remove ${WRKDIR}/bin from PATH for tests so that all commands

[patch] Test target PATH

is no longer the perl wrapper in ${WRKDIR}/bin; however, during tests we probably shouldn't be messing with install either. - Matthew Martin Index: infrastructure/mk/bsd.port.mk === RCS file: /cvs/ports/infrastructure/mk/bsd

Re: NEW: x11/libvterm

On Wed, Apr 22, 2015 at 11:37 AM, Matthew Martin wrote: > Attached is libvterm which is required for pangoterm. While it works, > launchpad > seems to tar revisions on the fly in a non-deterministic way making checksums > impossible. I would appreciate if anyone has a better way for

NEW: x11/libvterm

since there are none. - Matthew Martin libvterm.tgz Description: GNU Zip compressed data