Pike-devel November 2005

pike-devel@lists.lysator.liu.se

7 participants
22 discussions

Pike on the Gentoo P4 Computer Language Shootout
by Isaac Gouy 20 Nov '05

20 Nov '05

Although we've had Pike installed on the Computer Language Shootout for a few months, the default Debian install isn't configured for double-precision so several of the Pike programs show Error. http://shootout.alioth.debian.org/sandbox/pike.php Recently I started running the shootout scripts on a Gentoo P4 box and installed the latest release of Pike configured for double-precision - which gives a much better impression. http://shootout.alioth.debian.org/gp4/pike.php Please contribute faster more elegant Pike programs to the Computer Language Shootout - they'll certainly be published on GP4 - and having more programs makes it more likely that we'll re-install Pike on Debian with an appropriate configuration. http://shootout.alioth.debian.org/gp4/faq.php#contribute best wishes, Isaac __________________________________ Yahoo! FareChase: Search multiple travel sites in one click. http://farechase.yahoo.com

1 0

NoValue
by Martin Nilsson (Opera) ＠ Pike (-) developers forum 15 Nov '05

15 Nov '05

What is the NoValue stuff in the master good for? Is it used at all?

2 1

Note to self: How to build Nettle on windows the lazy way
by Peter Bortas ＠ Pike developers forum 13 Nov '05

13 Nov '05

This is mostly a note to myself* before I forget: libnettle builds fine with rntcl if configured with --disable-assembler --build=x86-windows. This is how the offcial windows builds of Pike from me are configured. Obviously this makes for a slow libnettle, but there is no extra work required to get it to work with VC6. * And possibly to grubba. It would be nice if Roxens Xenofarm machines got a libnettle.

3 2

Thanks for 7.6.50
by Xavier Beaudouin 13 Nov '05

13 Nov '05

Hello, I wanted just to tell you thanks for 7.6.50, it is far more stable than 7.6.24 on caudium... I have still hard lookup and problems with process.popen on freebsd, but Caudium.net website is mutch stable than it was on 7.6.24 (it was not able to be on line more than 1 or 5 minutes..., now uptime is about 24h, less than with 7.2... but the idea is there). Anybody that can help us with FreeBSD specifics eg process.popen will be welcomed to help the growing freebsd/pike community (there is lots of people that use freebsd with pike that are not subscribed on caudium / roxen / pike mailing lists). /Xavier

1 0

Re: 64bit pike on Solaris 10
by Henrik Grubbstr�m (Lysator) ＠ Pike (-) developers forum 08 Nov '05

08 Nov '05

>Just assumed it wasn't a developer issue but rather me doing something >wrong. Moving discussion to developer list. > >Some improvement! > >Parse argument 3:"-m/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/master.pike"... [...] >Init master... >There's no master to handle the error. Dumping it raw: >Error.BadArgument("Subtract on different types.\n") Ok, seems the change had the intended effect. Now try updating interpret.c.

2 10

Re: 64bit pike on Solaris 10
by Marcus Agehall (PacketFront) ＠ Pike (-) developers forum 08 Nov '05

08 Nov '05

Just assumed it wasn't a developer issue but rather me doing something wrong. Moving discussion to developer list. Some improvement! Parse argument 3:"-m/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/master.pike"... Parse argument 4:"/sw/pkg/pike/src/Pike-v7.7.21-snapshot/src/post_modules/Shuffler/make_sources.pike"... init_pike() Init CPU lib... Main init... Init shared nodes... Parse argument 1:"-DNOT_INSTALLED"... Parse argument 2:"-DPRECOMPILED_SEARCH_MORE"... Parse argument 3:"-m/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/master.pike"... Parse argument 4:"/sw/pkg/pike/src/Pike-v7.7.21-snapshot/src/post_modules/Shuffler/make_sources.pike"... Init C stack... Init time... Init threads... Init strings... Init interpreter... Init types... Init opcodes... Init programs... Init objects... Init modules... Initializing static module #0: "Builtin"... Initializing static module #1: "_math"... Initializing static module #2: "files"... Initializing static module #3: "sprintf"... Initializing static module #4: "_system"... Init master... There's no master to handle the error. Dumping it raw: Error.BadArgument("Subtract on different types.\n") Exit 10... Exiting static module #4: "_system"... Exiting static module #3: "sprintf"... Exiting static module #2: "files"... Exiting static module #1: "_math"... Exiting static module #0: "Builtin"... make[5]: *** [override] Error 10 make[5]: Leaving directory `/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/post_modules/Shuffler' make[4]: *** [Shuffler] Error 1 make[4]: Leaving directory `/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/post_modules' make[3]: *** [post_module_objects] Error 1 make[3]: Leaving directory `/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u' make[2]: *** [all] Error 2 make[2]: Leaving directory `/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u' make[1]: *** [_make_in_builddir] Error 2 make[1]: Leaving directory `/sw/pkg/pike/src/Pike-v7.7.21-snapshot' make: *** [compile] Error 2 / Marcus Agehall (PacketFront) Previous text: >13832556 2005-11-08 12:42 /29 lines/ Brevbäraren >Recipients: Pike (-) importmöte för mailinglistan >Comment to 13832479 >Subject: Re: 64bit pike on Solaris 10 >-------------------------------------------------------------------- >On Tue, 8 Nov 2005, Marcus Agehall (PacketFront) @ Pike (-) importmöte för mailinglistan wrote: > >> Enabling the trace macros gave some traces of what is going on: >> >> /sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/pike -DNOT_INSTALLED -DPRECOMPILED_SEARCH_MORE -m/sw/pkg/pike/src/Pike-v7.7.21-snapshot/build/sunos-5.10-sun4u/master.pike /sw/pkg/pike/src/Pike-v7.7.21-snapshot/src/post_modules/Shuffler/make_sources.pike /sw/pkg/pike/src/Pike-v7.7.21-snapshot/src/post_modules/Shuffler sources.h sources_to_compile >> Init master... >[...] > >Looks normal until it actually starts to execute pike code. > >> Init master... >> There's no master to handle the error. Dumping it raw: >> object(src/errors.h:107) >> Exit 10... > >errors.h line 107 would be a bad argument error. I've improved the output >from svalue.c:describe_svalue() for this case. Please update svalue.c and >see if you get better diagnostics. > >> However, it doesn't make more sense to me than what gdb already gave >> me. :( It fails somewhere in the low_clone_program() call when >> attempting to clone the master program. > >BTW: Shouldn't this discussion be on the developer list? > >-- >Henrik Grubbström grubba(a)roxen.com >Roxen Internet Software AB > > >(13832556) /Brevbäraren/-------------------------------------------- >

1 0

modules/_Protocols_DNS_SD on Pike 7.6 ?
by Xavier Beaudouin 07 Nov '05

07 Nov '05

Hello, Is there any chances that modules/_Protocols_DNS_SD can be included in next pike 7.6 release if possible? /Xavier -- Quand on essaye continuellement, on finit par y arriver. Donc, plus ca rate, plus on a de chance que ca marche... (Proverbe Shadok)

1 0

New string flags
by Martin Nilsson (Opera) ＠ Pike (-) developers forum 03 Nov '05

03 Nov '05

Would it be an idea to add a CONTAINS_ZERO flag to the new string flag field, so we can rid the code of all O(n) if(strlen(s->str)!=s->len)?

2 2

Pike 7.6.40
by Martin Nilsson (Opera) ＠ Pike (-) developers forum 03 Nov '05

03 Nov '05

I've released Pike 7.6.50 as latest official release on the Pike site. Please contribute binary builds (make bin_export) for the missing platforms (right now all). Remember to include the config.info file. I'll make no annoucements besides this one, so please do your best to spread the word.

2 1

RC x
by Martin Nilsson (Opera) ＠ Pike (-) developers forum 02 Nov '05

02 Nov '05

I've uploaded 7.6.50 for review.

1 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

Pike-devel November 2005