Re: CVS: cvs.openbsd.org: ports
On Tue, Jul 16, 2019 at 11:27:33PM -0600, Otto Moerbeek wrote: > CVSROOT: /cvs > Module name: ports > Changes by: o...@cvs.openbsd.org2019/07/16 23:27:33 > > Modified files: > net/powerdns_recursor: Makefile distinfo > net/powerdns_recursor/patches: patch-Makefile_in > Removed files: > net/powerdns_recursor/patches: patch-logger_cc patch-logger_hh > patch-mtasker_context_hh > patch-mystack_allocator_hh > patch-pdns_recursor_cc > patch-rec_channel_cc > > Log message: > Update to PowerDNS Recursor 4.2.0 > That was with some help and ok sthen@
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: o...@cvs.openbsd.org2019/07/16 23:27:33 Modified files: net/powerdns_recursor: Makefile distinfo net/powerdns_recursor/patches: patch-Makefile_in Removed files: net/powerdns_recursor/patches: patch-logger_cc patch-logger_hh patch-mtasker_context_hh patch-mystack_allocator_hh patch-pdns_recursor_cc patch-rec_channel_cc Log message: Update to PowerDNS Recursor 4.2.0
Re: NEW: math/py-cftime and math/py-netcdf4
On Wed, Jul 17, 2019 at 12:50:34AM -0400, Kurt Mosiejczuk wrote:
> On Thu, Jun 27, 2019 at 03:14:31PM +0200, Martin Reindl wrote:
>
> > Here are the two updated tarballs with the ports. py-netcdf4 got a
> > mini-update to 1.5.1.2 in the meantime.
>
> I got back to these two ports.
>
> For py-cftime, I ended u padding a bunch of TEST_DEPENDS including
> adding py-ipaddress for the python2 flavor. I added the package itself
> to the TEST_DEPENDS. With that it passed all tests for python2 and
> all but one for python3 on both amd64 and sparc64. I also tweaked the
> do-test target so it set all environment variables it should.
>
> For py-netcdf4 I tweaked the PKGNAME to be all lowercase. All of its
> similarly named dependencies stay lowercase. I added itself to the
> TEST_DEPENDS. I added TEST_ENV with TMPDIR pointing at ${WRKDIR}/tmp
> since it wants to slap a 4 gigabyte test file in /tmp and my /tmp
> partition is not good enough. I added a pre-test to create the tmp
> directory in ${WRKDIR}. I also tweaked the do-test target just like I
> did for py-cftime. All tests pass on amd64. Well, all pass
> if it has network. A handful of them fail if doing PORTS_PRIVSEP. (I
> temporarily allowed _pbuild network for a second attempt at the tests).
>
> On sparc64, py-netcdf4 fails two tests: test_unsigned and VariablesTestCase.
>
> ==
>
> FAIL: test_unsigned (tst_Unsigned.Test_Unsigned)
>
> --
>
> Traceback (most recent call last):
> File
> "/usr/ports/pobj/py-netcdf4-1.5.1.2-python3/netCDF4-1.5.1.2/test/tst_Unsi
> gned.py", line 31, in test_unsigned
> assert(data1.mask.sum() == data2.mask.sum())
> AssertionError
>
> ==
> FAIL: runTest (tst_compoundvar.VariablesTestCase)
> testing compound variables
> --
> Traceback (most recent call last):
> File
> "/usr/ports/pobj/py-netcdf4-1.5.1.2-python3/netCDF4-1.5.1.2/test/tst_comp
> oundvar.py", line 72, in setUp
> assert (cmptype4 == dtype4a) # data type should be aligned
> AssertionError
>
> I don't know if it is their test code or some other difference on sparc64.
>
Very grateful for your thorough testing and input for these ports.
>From a first glance this is a netcdf-specific code problem on sparc64. dtype4a
is numpy code where I suspect there is no problem on sparc64. Whereas cmdtype4
uses netcdf's createCompoundType method. Need to look deeper into this as time
permits, even though I am not getting a sparc64 (again).
-m
[Update] www/p5-CGI-Lite : Update to 3.02
Hi, Here is a patch to update www/p5-CGI-Lite to 3.02. It build well and passed all tests on myamd64-head system No other ports depends on it. Comments? OK? wen Index: Makefile === RCS file: /cvs/ports/www/p5-CGI-Lite/Makefile,v retrieving revision 1.13 diff -u -p -r1.13 Makefile --- Makefile12 Jul 2019 20:50:50 - 1.13 +++ Makefile17 Jul 2019 01:20:43 - @@ -2,7 +2,7 @@ COMMENT= process and decode WWW forms and cookies -DISTNAME = CGI-Lite-3.00 +DISTNAME = CGI-Lite-3.02 CATEGORIES=www MAINTAINER=Abel Abraham Camarillo Ojeda Index: distinfo === RCS file: /cvs/ports/www/p5-CGI-Lite/distinfo,v retrieving revision 1.4 diff -u -p -r1.4 distinfo --- distinfo10 Jun 2015 09:17:37 - 1.4 +++ distinfo17 Jul 2019 01:20:43 - @@ -1,2 +1,2 @@ -SHA256 (CGI-Lite-3.00.tar.gz) = 5S5ImLsP+XKQBTxyIzpUzUiC2hJG58DoYZNhG2bQsrs= -SIZE (CGI-Lite-3.00.tar.gz) = 36288 +SHA256 (CGI-Lite-3.02.tar.gz) = QeihfNFhX6RZAGYqJUobikQ+BIYR6wq+WpiBsX3Kz30= +SIZE (CGI-Lite-3.02.tar.gz) = 38073
Re: devel/llvm: broken CMake files?
On 2019/07/17 00:51, Jeremie Courreges-Anglas wrote: > On Sun, Jul 14 2019, Brian Callahan wrote: > > Hi ports -- > > > > When trying to update flang, I get this message during `make configure': > > CMake Error at /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake:14 (if): > > Syntax error in cmake code at > > > > /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake:14 > > > > when parsing string > > > > lld-link\.exe > > > > Invalid escape sequence \. > > Call Stack (most recent call first): > > CMakeLists.txt:164 (include) > > > > > > -- Configuring incomplete, errors occurred! > > > > Indeed, if I remove the \ from lld-link\.exe from > > /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake then flang builds > > fine. > > I'm not sure what the problem is, maybe a difference between our cmake > version (3.10.2) and what upstream tends to use? > > Anyway, if you have a patch that removes the backslash and fixes your > problem, ok jca@ > > BTW, regarding cmake, rsadowski@ has an update and he asked for tests: > > https://marc.info/?l=openbsd-ports=156204264710542=2 > > -- > jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE > https://github.com/llvm-mirror/llvm/commit/b8b62917a1566bfc6d8706d06042171aa2306bbc then followed by https://github.com/llvm-mirror/llvm/commit/614d096016667810cc777db37d180accb2390cfb
Re: devel/llvm: broken CMake files?
On Sun, Jul 14 2019, Brian Callahan wrote: > Hi ports -- > > When trying to update flang, I get this message during `make configure': > CMake Error at /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake:14 (if): > Syntax error in cmake code at > > /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake:14 > > when parsing string > > lld-link\.exe > > Invalid escape sequence \. > Call Stack (most recent call first): > CMakeLists.txt:164 (include) > > > -- Configuring incomplete, errors occurred! > > Indeed, if I remove the \ from lld-link\.exe from > /usr/local/lib/cmake/llvm/HandleLLVMOptions.cmake then flang builds > fine. I'm not sure what the problem is, maybe a difference between our cmake version (3.10.2) and what upstream tends to use? Anyway, if you have a patch that removes the backslash and fixes your problem, ok jca@ BTW, regarding cmake, rsadowski@ has an update and he asked for tests: https://marc.info/?l=openbsd-ports=156204264710542=2 -- jca | PGP : 0x1524E7EE / 5135 92C1 AD36 5293 2BDF DDCC 0DFA 74AE 1524 E7EE
NEW: devel/esptool-py 2.6 [6/7]
devel/esptool-py: A Python-based, open source, platform independent, utility to communicate with the ROM bootloader in Espressif ESP8266 & ESP32 chips. The py-serial or py3-serial plugins required. esptool-py.2.6.tar.gz Description: Binary data
NEW: devel/mkspiffs 0.2.3 [7/7]
devel/mkspiffs: Tool to build and unpack SPIFFS images. mkspiffs-0.2.3.tar.gz Description: Binary data
NEW: devel/esptool-ck 0.4.14 [5/7]
devel/esptool-ck: Esptool is a tool to create firmware files for the ESP8266 chip and flash the firmware to the chip over serial port. esptool-ck.0.4.14.tar.gz Description: Binary data
NEW: devel/arduino-esp8266 2.4.1 [4/7]
devel/arduino-esp8266: Framework to program Espressif ESP8266 chipsets via the Arduino environment. arduino-esp8266-2.4.1.tar.gz Description: Binary data
NEW: devel/arduino-esp32 1.0.2 [3/7]
devel/arduino-esp32: Framework to program Espressif ESP32 chipsets via the Arduino environment. arduino-esp32-1.0.2.tar.gz Description: Binary data
NEW: devel/xtensa-lx106-elf-gcc 4.8.5 [2/7]
devel/xtensa-lx106-elf-gcc: GNU cross compiler suite, configured for the xtensa-lx106-elf target. xtensa-lx106-elf-gcc.4.8.5.tar.gz Description: Binary data
NEW: devel/xtensa-esp32-elf-gcc 8.2.0 [1/7]
Hello ports@, I am sending a series of ports for review. These ports are to allow working with Espressif ESP8266 and ESP32 chips in OpenBSD. These ports work with the current Arduino port and the Espressif SDKs themselves. The Espressif way of doing things is a mess, so I'm hosting some of the distribution files, which are conglomerates of patches and overlays that are spread across repositores. This was the easiest way to centralize that I could find. Using their preferred crosstool-NG compiler, I built the cross-compilers in Linux and packaged those patched src files for distribution. If there needs to be a different method, please let me know or if there is a preferred server for self-hosting, it can all be changed. I need some help cleaning up the compiler ports, which are based on compiler ports in the tree. Everything compiles and runs fine and programs both chips. Is there anyone interested in reviewing, testing, okaying and importing? I've been working on these for quite some time and I think I'm going blind to errors as I've been working to make these functional. First port attached is devel/xtensa-esp32-elf-gcc: GNU cross compiler suite, configured for the xtensa-lx106-elf target. Thanks, Tracey xtensa-esp32-elf-gcc.8.2.0.tar.gz Description: Binary data
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/07/16 15:29:41 Modified files: archivers/gtar : Makefile converters/libiconv: Makefile devel/arm-none-eabi/gdb: Makefile devel/bison: Makefile devel/gdb : Makefile devel/gettext : Makefile devel/gpatch : Makefile devel/m4 : Makefile lang/guile2: Makefile math/octave: Makefile math/pspp : Makefile misc/findutils : Makefile print/gv : Makefile security/oath-toolkit: Makefile sysutils/augeas: Makefile sysutils/coreutils: Makefile sysutils/libvirt: Makefile textproc/gsed : Makefile textproc/wdiff : Makefile Log message: Some ports using gnulib previously detected a broken realpath and switched to an internal replacement function. Following the changes to make realpath(3) use the __realpath() syscall these no longer detect broken realpath i.e. produce different code. Bump REVISION to ensure that users get the new version.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/07/16 14:42:26 Removed files: mail/dovecot/patches: patch-src_lib_ostream-file_c Log message: oops, forgot to remove this patch when backing out
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: b...@cvs.openbsd.org2019/07/16 14:38:03 Modified files: math/py-pandas : Makefile distinfo math/py-pandas/patches: patch-pandas_io_formats_terminal_py patch-setup_cfg math/py-pandas/pkg: PLIST Log message: Update to py-pandas-0.24.2. 0.24.x is a major release from 0.23.4, which includes a number of API changes, new features, enhancements, and performance improvements along with a large number of bug fixes. Lists of changes can be found at: - https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.0.html - https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.1.html - https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.2.html OK daniel@
Re: plan9port: win(1) hang on acme(1) window
On Tue, Jul 16, 2019 at 04:14:19PM -0400, Patrick Marchand wrote: > Hello, > > On 07/16, Gleydson Soares wrote: > > I will have a look at it tonight. Even though I’m running acme/win with no > > issues. Will try reproduce it on a newer snapshot. > I had the time to upgrade from the friday snaphot to today's snapshot > and it solved the issue. I'm curious to read the csv commits to find out what > changed over the last few days. Caused by this commit[1] which was later backed out[2]. Sorry for the inconvenience. [1] https://marc.info/?l=openbsd-cvs=156277704122293=2 [2] https://marc.info/?l=openbsd-cvs=156293980006370=2
Re: plan9port: win(1) hang on acme(1) window
Hello, On 07/16, Gleydson Soares wrote: > I will have a look at it tonight. Even though I’m running acme/win with no > issues. Will try reproduce it on a newer snapshot. I had the time to upgrade from the friday snaphot to today's snapshot and it solved the issue. I'm curious to read the csv commits to find out what changed over the last few days. Gleydson, could you take a look at my diff that updates the port to the latest git commit? I'ts my first time contributing to a port and I'd like to know if my diff is okay. Thanks
[SOLVED] plan9port: win(1) hang on acme(1) window
Hi all As gsoares@ said, I tried to upgrade to a newer snapshot. After that, the problem is solved and everything works properly Thanks for the help and sorry for the noise. Regards -- Andrea
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: jas...@cvs.openbsd.org 2019/07/16 13:05:56 Modified files: lang/vala : Makefile distinfo Log message: update to vala-0.44.6
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2019/07/16 12:39:34 Modified files: databases/sqlite3: Makefile distinfo Log message: Upgrade to sqlite 3.29.0. See https://sqlite.org/releaselog/3_29_0.html
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: chr...@cvs.openbsd.org 2019/07/16 12:02:07 Modified files: sysutils/opam : Makefile distinfo Removed files: sysutils/opam/files: aspcud sysutils/opam/patches: patch-m4_ocaml_m4 patch-shell_bootstrap-ocaml_sh Log message: Straightforward upgrade to 2.0.5 OK avsm@
games/zangband patch(es)
ZAangbad goes to (close enough to) 100% CPU due to the use of 64-bit containers on 64-bit systems for what should be a 32-bit container for the Rand_state values over in src/z-rand.c. One way to fix this might be to use stdint.h types for s32b and friends: --- src/h-type.h.orig Fri May 4 18:21:57 2001 +++ src/h-type.hTue Jul 16 08:39:48 2019 @@ -108,19 +108,12 @@ /* Signed/Unsigned 16 bit value */ -typedef signed short s16b; -typedef unsigned short u16b; +typedef int16_t s16b; +typedef uint16_t u16b; /* Signed/Unsigned 32 bit value */ -#ifdef L64 /* 64 bit longs */ -typedef signed int s32b; -typedef unsigned int u32b; -#else -typedef signed long s32b; -typedef unsigned long u32b; -#endif - - +typedef int32_t s32b; +typedef uint32_t u32b; /*** Pointers to all the basic types defined above ***/ A potential issue is the use of time(NULL) to set the seed; this might better be done via arc4random, or not. --- src/bldg.c.orig Mon Aug 27 09:49:18 2001 +++ src/bldg.c Tue Jul 16 08:39:48 2019 @@ -599,7 +599,7 @@ /* Prevent savefile-scumming of the casino */ Rand_quick = TRUE; - Rand_value = time(NULL); + Rand_value = arc4random(); /* Return the amount of gold bet */ return (wager); --- src/dungeon.c.orig Tue Aug 14 05:06:15 2001 +++ src/dungeon.c Tue Jul 16 08:39:48 2019 @@ -3598,14 +3598,7 @@ u32b seed; /* Basic seed */ - seed = (time(NULL)); - -#ifdef SET_UID - - /* Mutate the seed on Unix machines */ - seed = ((seed >> 3) * (getpid() * 2)); - -#endif + seed = arc4random(); /* Use the complex RNG */ Rand_quick = FALSE; --- src/main-dos.c.orig Sun Feb 4 14:11:56 2001 +++ src/main-dos.c Tue Jul 16 08:39:48 2019 @@ -2185,7 +2185,7 @@ /* Initialize the "complex" RNG for the midi-shuffle function */ Rand_quick = FALSE; - Rand_state_init(time(NULL)); + Rand_state_init(arc4random()); /* Set the Angband colors/graphics/sound mode */ Term_xtra_dos_react();
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: rob...@cvs.openbsd.org 2019/07/16 10:59:25 Modified files: www/chromium : Makefile distinfo Log message: update to 75.0.3770.142
Update: sysutils/exa 0.8.0 -> 0.9.0
0.9.0 was released yesterday[1], making it two years since the last
release. More features were added and a couple of bugs were fixed.
I compiled this on -current using lang/rust from ports, haven't tried it
on 6.5.
I had a bit of trouble updating this port. I tried following
port-modules(5) and did this:
- Build the old port/fetch old modules
- Update the tagname, regenerate release checksum and extract
- Regenerate the crate info with modcargo-gen-crates-licenses
- Fetch the new crates, makesum and extract
- cd `make show=WRKSRC`; rm -rf modcargo-crates. If I didn't do this,
update-patches would add a new file for each new crate.
- fix the existing Cargo.toml patch.
- extract crates again.
- build and update-plist
- install package
I tried seaching for docs around updating rust ports and couldn't find
any. Perhaps any other rust porters could chime in?
[1] https://github.com/ogham/exa/releases/tag/v0.9.0
--
Stephen Gregoratto
PGP: 3FC6 3D0E 2801 C348 1C44 2D34 A80C 0F8E 8BAB EC8B
Index: Makefile
===
RCS file: /cvs/ports/sysutils/exa/Makefile,v
retrieving revision 1.5
diff -u -p -r1.5 Makefile
--- Makefile12 Jul 2019 20:49:42 - 1.5
+++ Makefile16 Jul 2019 13:54:46 -
@@ -4,8 +4,7 @@ COMMENT = ls alternative written in Rus
GH_ACCOUNT = ogham
GH_PROJECT = exa
-GH_TAGNAME = v0.8.0
-REVISION = 1
+GH_TAGNAME = v0.9.0
CATEGORIES = sysutils
@@ -18,67 +17,69 @@ DISTFILES +=${DISTNAME}${EXTRACT_SUFX}
MODULES = devel/cargo
BUILD_DEPENDS =lang/rust
-MODCARGO_CRATES += aho-corasick0.5.3 # Unlicense/MIT
-MODCARGO_CRATES += ansi_term 0.8.0 # MIT
-MODCARGO_CRATES += bitflags0.7.0 # MIT/Apache-2.0
-MODCARGO_CRATES += bitflags0.9.1 # MIT/Apache-2.0
-MODCARGO_CRATES += byteorder 0.4.2 # Unlicense/MIT
-MODCARGO_CRATES += cmake 0.1.25 # MIT/Apache-2.0
-MODCARGO_CRATES += conv0.3.3 # MIT
-MODCARGO_CRATES += custom_derive 0.1.7 # MIT/Apache-2.0
-MODCARGO_CRATES += datetime0.4.5 # MIT
-MODCARGO_CRATES += env_logger 0.3.5 # MIT/Apache-2.0
-MODCARGO_CRATES += gcc 0.3.53 # MIT/Apache-2.0
-MODCARGO_CRATES += getopts 0.2.14 # MIT/Apache-2.0
-MODCARGO_CRATES += git20.6.8 # MIT/Apache-2.0
-MODCARGO_CRATES += glob0.2.11 # MIT/Apache-2.0
-MODCARGO_CRATES += idna0.1.4 # MIT/Apache-2.0
-MODCARGO_CRATES += iso8601 0.1.1 # MIT
-MODCARGO_CRATES += kernel32-sys0.2.2 # MIT
-MODCARGO_CRATES += lazy_static 0.2.8 # MIT/Apache-2.0
-MODCARGO_CRATES += libc0.2.43 # MIT/Apache-2.0
-MODCARGO_CRATES += libgit2-sys 0.6.14 # MIT/Apache-2.0
-MODCARGO_CRATES += libz-sys1.0.16 # MIT/Apache-2.0
-MODCARGO_CRATES += locale 0.2.2 # MIT
-MODCARGO_CRATES += log 0.3.8 # MIT/Apache-2.0
-MODCARGO_CRATES += magenta 0.1.1 # BSD-3-Clause
-MODCARGO_CRATES += magenta-sys 0.1.1 # BSD-3-Clause
-MODCARGO_CRATES += matches 0.1.6 # MIT
-MODCARGO_CRATES += memchr 0.1.11 # Unlicense/MIT
-MODCARGO_CRATES += natord 1.0.9 # MIT
-MODCARGO_CRATES += nom 1.2.4 # MIT
-MODCARGO_CRATES += num 0.1.40 # MIT/Apache-2.0
-MODCARGO_CRATES += num-bigint 0.1.40 # MIT/Apache-2.0
-MODCARGO_CRATES += num-complex 0.1.40 # MIT/Apache-2.0
-MODCARGO_CRATES += num-integer 0.1.35 # MIT/Apache-2.0
-MODCARGO_CRATES += num-iter0.1.34 # MIT/Apache-2.0
-MODCARGO_CRATES += num-rational0.1.39 # MIT/Apache-2.0
-MODCARGO_CRATES += num-traits 0.1.40 # MIT/Apache-2.0
-MODCARGO_CRATES += num_cpus1.6.2 # MIT/Apache-2.0
-MODCARGO_CRATES += number_prefix 0.2.7 # MIT
-MODCARGO_CRATES += pad 0.1.4 # MIT
-MODCARGO_CRATES += percent-encoding1.0.0 # MIT/Apache-2.0
-MODCARGO_CRATES += pkg-config 0.3.9 # MIT/Apache-2.0
-MODCARGO_CRATES += rand0.3.16 # MIT/Apache-2.0
-MODCARGO_CRATES += redox_syscall 0.1.31 # MIT
-MODCARGO_CRATES += regex 0.1.80 # MIT/Apache-2.0
-MODCARGO_CRATES += regex-syntax0.3.9 # MIT/Apache-2.0
-MODCARGO_CRATES += rustc-serialize 0.3.24 # MIT/Apache-2.0
-MODCARGO_CRATES += scoped_threadpool 0.1.7 # MIT
-MODCARGO_CRATES += term_grid 0.1.6 # MIT
-MODCARGO_CRATES += term_size 0.3.0 # MIT OR Apache-2.0
-MODCARGO_CRATES += thread-id
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: st...@cvs.openbsd.org 2019/07/16 10:14:57 Modified files: mail/dovecot : Makefile distinfo mail/dovecot/pkg: PLIST-server mail/dovecot-pigeonhole: Makefile distinfo Log message: Revert the Dovecot update for now, there are some problems with replication e.g. https://www.mail-archive.com/dovecot@dovecot.org/msg78030.html
Re: CVS: cvs.openbsd.org: ports
On 2019/07/16 10:14, Stuart Henderson wrote: > CVSROOT: /cvs > Module name: ports > Changes by: st...@cvs.openbsd.org 2019/07/16 10:14:57 > > Modified files: > mail/dovecot : Makefile distinfo > mail/dovecot/pkg: PLIST-server > mail/dovecot-pigeonhole: Makefile distinfo > > Log message: > Revert the Dovecot update for now, there are some problems with replication > e.g. https://www.mail-archive.com/dovecot@dovecot.org/msg78030.html > This was OK Brad.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ki...@cvs.openbsd.org 2019/07/16 09:59:17 Modified files: net/seafile: Makefile.inc net/seafile/client: Makefile distinfo net/seafile/seafile: distinfo Log message: update to seafile-7.0.1
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: lan...@cvs.openbsd.org 2019/07/16 07:14:50 Modified files: geo/mapserver : Makefile distinfo geo/mapserver/pkg: PLIST-perl PLIST-php PLIST-python Log message: Update to mapserver 7.4.1.
Re: plan9port: win(1) hang on acme(1) window
Hi, I will have a look at it tonight. Even though I’m running acme/win with no issues. Will try reproduce it on a newer snapshot. > On 16 Jul 2019, at 04:25, a...@gmx.com wrote: > > Hi > >> I also opened up an issue on the github repository of the project: >> https://github.com/9fans/plan9port/issues/275 > > Great, thanks! I'll follow it > > Regards. >
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: s...@cvs.openbsd.org2019/07/16 03:56:55 Modified files: devel/git : Makefile devel/git/pkg : gitdaemon.rc Log message: Fix gitdaemon rc.d(8) script's mechanism for setting the daemon's UID. UID switching was controlled via git-dameon's --user option in $daemon_flags. This is problematic in case $gitdaemon_flags is unwittingly set to some value which omits the --user option: git-daemon would then run as root. The fix implemented here sets $daemon_user to _gitdaemon by default, and leaves $daemon_flags empty by default. This ensures that git-daemon runs as user _gitdaemon regardless of the value of $gitdaemon_flags. Anyone using git-daemon should check their system. If gitdaemon is currently running as root, set gitdaemon_user=_gitdaemon in rc.conf.local and restart the service. ok kn@
Re: segfault in git using tag verify (patch included)
On Tue, Jul 16, 2019 at 02:00:39AM -0700, Steven Roberts wrote: > This is the patch we created which seems to resolve the issue: > > /usr/ports/devel/git/patches/patch-gpg-interface_c I have not tested this fix, but it does sound like you want to talk with upstream about it: https://git-scm.com/community I suggest checking whether this particular issue has been reported already; if not, submit your fix according to the guidelines found behind above mentioned link. Thanks for tracking it down either ways!
segfault in git using tag verify (patch included)
Hi,
I believe I have found an off-by-one error in git.
It was discovered using sysctl vm.malloc_conf=S attempting to verify a
git tag from:
https://github.com/fenderq/monstermash
When running:
$ git tag -v v1.2
I would get a segfault.
When I removed the malloc option 'S' it did not segfault.
I narrowed it down to segfault using only "More junking"
MALLOC_OPTIONS=J
With the help of a friend (who wants to remain anonymous), we managed to
narrow the segfault cause down.
This is the patch we created which seems to resolve the issue:
/usr/ports/devel/git/patches/patch-gpg-interface_c
diff -ur a/gpg-interface.c b/gpg-interface.c
--- gpg-interface.c Sun Feb 24 08:31:46 2019
+++ gpg-interface.c Mon Jul 15 23:24:03 2019
@@ -116,6 +116,10 @@
for (line = buf; *line; line = strchrnul(line+1, '\n')) {
while (*line == '\n')
line++;
+
+ /* break out of trailing \n */
+ if (!*line) break;
+
/* Skip lines that don't start with GNUPG status */
if (!skip_prefix(line, "[GNUPG:] ", ))
continue;
Re: Have portgen(1) write new DESCR to DESCR.new
On 2019/07/15 09:47, Kurt Mosiejczuk wrote:
> Based on a conversation with afresh1 about the pypi DESCR, this patch
> has portgen write the description to DESCR.new rather than just
> DESCR. This means if one doesn't intervene and deal with DESCR.new,
> portcheck will complain. This is related to my forgetting that I hadn't
> checked DESCR on a new port but thought I had. Perhaps the portgen
> module in question generated a perfect DESCR, but one should make sure
> it did so.
I'm not a fan of this, it adds an extra "mv" step that needs to
be done before installing the software to test and I don't think it will
actually help anything. So the process which is currently
work on port
test/revise/retest
forget to update DESCR
run portcheck, no problem noticed
send port out
will become
work on port
try to test
can't package, mv pkg/DESCR.new pkg/DESCR, "i'll update it when i know
what the software actually does"
test/revise/retest
forget to update DESCR
run portcheck, no problem noticed
send port out
If you want portcheck to trip on it, how about writing an identifiable
string to DESCR ("* Generated by portgen: needs review" or similar) and
having portcheck look for that instead?
(People should try to remember to read every file in the port for a
quick review before they send it to ports@. When I remember - most but
not all of the time! - I just open the tar itself in vim and work through
the files in order).
> Now, this leaves the mechanism that renames the original DESCR to
> DESCR.orig for an updated port (and also does so for Makefile). I figure
> in the case of an update, doing this will also encourage looking at the
> old DESCR versus the new version. I'd be interested to know how others
> feel about that though.
As long as it doesn't change DESCR but just adds a new file this sounds
reasonable.
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2019/07/16 01:47:07 Modified files: mail/evolution : Makefile distinfo Log message: Update to evolution-3.32.4.
[update] math/py-pandas 0.23.4 -> 0.24.2
Enclosed diff brings math/py-pandas to 0.24.2, which is a major release
from 0.23.4, and a minor bug-fix release in the 0.24.x series. The
0.24.x series includes a number of API changes, new features,
enhancements, and performance improvements along with a large number
of bug fixes.
Highlights include:
- Optional Nullable Integer Support
- New APIs for accessing the array backing a Series or Index
- A new top-level pandas.array() method for creating arrays
- Store Interval and Period data in a Series or DataFrame
- Joining with two MultiIndexes
List of all the change from 0.23.4 can be found at
https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.0.html,
https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.1.html,
https://pandas.pydata.org/pandas-docs/version/0.24/whatsnew/v0.24.2.html.
The 0.24.x series of releases will be the last to support Python 2.
Future feature releases will support Python 3 only.
Changes to the port:
- Include py-hypothesis as TDEP, which is required for running tests
Testing:
- Tarball provided by pypi (and by GitHub) misses relevant files for
running the test suite. This results in numerous failures and errors.
Output 'make test':
= 666 failed, 43508 passed, 2423 skipped, 998 xfailed, 23 xpassed, 848
warnings, 163 error in 4239.02 seconds =
- Run tested with several work related tools
Comments/OK?
Index: Makefile
===
RCS file: /cvs/ports/math/py-pandas/Makefile,v
retrieving revision 1.8
diff -u -p -r1.8 Makefile
--- Makefile12 Jul 2019 20:47:46 - 1.8
+++ Makefile16 Jul 2019 07:31:41 -
@@ -2,10 +2,9 @@
COMMENT = data analysis and manipulation library
-MODPY_EGG_VERSION =0.23.4
+MODPY_EGG_VERSION =0.24.2
DISTNAME = pandas-${MODPY_EGG_VERSION}
PKGNAME = py-${DISTNAME}
-REVISION = 3
CATEGORIES = math
@@ -39,6 +38,7 @@ TEST_DEPENDS += databases/py-mysqlclien
databases/py-sqlalchemy${MODPY_FLAVOR} \
devel/ipython${MODPY_FLAVOR} \
devel/py-html5lib${MODPY_FLAVOR} \
+ devel/py-hypothesis${MODPY_FLAVOR} \
devel/py-ipykernel${MODPY_FLAVOR} \
devel/py-pathlib${MODPY_FLAVOR} \
devel/py-xlsxwriter${MODPY_FLAVOR} \
@@ -56,6 +56,8 @@ TEST_DEPENDS += databases/py-mysqlclien
TEST_ENV +=TZ=UTC
+# XXX tarball does not include all files required for the test suite, as such
+# numerous tests fail.
do-test:
cd ${WRKSRC} && ${MODPY_CMD} build_ext --inplace
@${MODPY_TEST_TARGET}
Index: distinfo
===
RCS file: /cvs/ports/math/py-pandas/distinfo,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 distinfo
--- distinfo14 Sep 2018 08:09:05 - 1.1.1.1
+++ distinfo16 Jul 2019 07:31:41 -
@@ -1,2 +1,2 @@
-SHA256 (pandas-0.23.4.tar.gz) = WyTKR6z2kiLoJTDokRHdnRT5uXCrLNOhwsePDE+7pPQ=
-SIZE (pandas-0.23.4.tar.gz) = 10490077
+SHA256 (pandas-0.24.2.tar.gz) = T5GfQJxDNXelAeAjlD5YLFc1XVCnJMWJ54vB1VGlNaI=
+SIZE (pandas-0.24.2.tar.gz) = 11837693
Index: patches/patch-pandas_io_formats_terminal_py
===
RCS file:
/cvs/ports/math/py-pandas/patches/patch-pandas_io_formats_terminal_py,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-pandas_io_formats_terminal_py
--- patches/patch-pandas_io_formats_terminal_py 14 Sep 2018 08:09:05 -
1.1.1.1
+++ patches/patch-pandas_io_formats_terminal_py 16 Jul 2019 07:31:41 -
@@ -3,13 +3,12 @@ $OpenBSD: patch-pandas_io_formats_termin
Index: pandas/io/formats/terminal.py
--- pandas/io/formats/terminal.py.orig
+++ pandas/io/formats/terminal.py
-@@ -42,7 +42,8 @@ def get_terminal_size():
+@@ -42,7 +42,7 @@ def get_terminal_size():
+ tuple_xy = _get_terminal_size_tput()
# needed for window's python in cygwin's xterm!
- if current_os == 'Linux' or \
- current_os == 'Darwin' or \
--current_os.startswith('CYGWIN'):
-+current_os == 'OpenBSD' or \
-+current_os.startswith('CYGWIN'):
+ if (current_os == 'Linux' or current_os == 'Darwin' or
+-current_os.startswith('CYGWIN')):
++current_os.startswith('CYGWIN') or current_os == 'OpenBSD'):
tuple_xy = _get_terminal_size_linux()
if tuple_xy is None:
tuple_xy = (80, 25) # default value
Index: patches/patch-setup_cfg
===
RCS file: /cvs/ports/math/py-pandas/patches/patch-setup_cfg,v
retrieving revision 1.1.1.1
diff -u -p -r1.1.1.1 patch-setup_cfg
--- patches/patch-setup_cfg 14 Sep 2018 08:09:05 - 1.1.1.1
+++ patches/patch-setup_cfg 16 Jul 2019 07:31:41
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: ajacou...@cvs.openbsd.org 2019/07/16 01:46:50 Modified files: databases/evolution-data-server: Makefile distinfo Log message: Update to evolution-data-server-3.32.4.
Re: plan9port: win(1) hang on acme(1) window
Hi > I also opened up an issue on the github repository of the project: > https://github.com/9fans/plan9port/issues/275 Great, thanks! I'll follow it Regards.
[NEW] www/p5-HTTP-Entity-Parser
Hi, ports@: Here is a patch to create new port www/p5-HTTP-Entity-Parser, which is required by the future update of www/p5-Plack. It build well and passed all tests on my amd64-head system. Comments? OK? wen p5-HTTP-Entity-Parser-0.21.tar.gz Description: p5-HTTP-Entity-Parser-0.21.tar.gz
CVS: cvs.openbsd.org: ports
CVSROOT:/cvs Module name:ports Changes by: piro...@cvs.openbsd.org 2019/07/16 01:03:25 Modified files: math/lapack: Tag: OPENBSD_6_5 Makefile distinfo math/lapack/pkg: Tag: OPENBSD_6_5 PLIST Log message: Fix lapack build. The manpages distfile changed and thus affected the packaging breaking other ports in the process. OK espie@, robert@
