Pike-devel

pike-devel@lists.lysator.liu.se

3044 discussions

Calendar question
by David Gourdelier 30 Mar '04

30 Mar '04

Hi again, It seems to me that the autodoc of the Calendar module is incorrect or there is a problem in the Calendar.Time* classes. The doc say you can use Calendar.TimeRange("unix", int unixtime) but when I try with either pike 7.5 or 7.2, I have: > Calendar.TimeRange("unix", time()); Calling undefined function. ISO.TimeRange: TimeRange()->create_unixtime_default(1080599245) /usr/local/pike/7.2.504/lib/modules/Calendar.pmod/TimeRanges.pmod:81: TimeRange()->create("unix",1080599245) ISO: ISO->TimeRange() -:1: ___Foo4711() > Calendar.TimeRange("unix", time()); Calling undefined function. ISO.TimeRange: TimeRange()->create_unixtime_default(1080599068) /usr/local/pike/7.5.23/lib/modules/Calendar.pmod/TimeRanges.pmod:78: TimeRange()->create("unix",1080599068) /usr/local/pike/7.5.23/lib/modules/Calendar.pmod/TimeRanges.pmod:11: ISO->TimeRange() HilfeInput:1: HilfeInput()->___HilfeWrapper() There is the same problem with Calendar.Time.TimeofDay. When looking at the code it seems indeed that the create_unixtime_default function is not defined in either TimeRange nor TimeofDay classes but defined in clases that inherited these one. So I wonder if the doc lie or if the code is wrong ? I also changed the doc for http://pike.ida.liu.se/generated/manual/modref/ex/predef_3A_3A/Calendar/Tim…, replacing "unixtime" by "unix". / David

3 3

Changelog and newsletter
by Bill Welliver 30 Mar '04

30 Mar '04

Okay folks, here's the deal: I'll volunteer to periodically (4-6 times a year should developments dictate) put together a Pike Update. As I see it, things that would be worthwhile in such a production might include: Pike development updates (what's happening, what the plan is) Pike Tips Usage Articles Conference Information Pike-based application updates and information etc. In order to do this, I need some help. I know that there is at least a little resistance to the idea, but having a changelog updated as changes get made would really help me (and other potentially interested parties) figure out what is going on. The reasons I think that a changelog is necessary and beneficial include: 1. A new "feature" or change may involve 5 - 10 or more cvs checkins, the log entries of which may not hint at the actual change being made. 2. Sometimes, even those of us who might follow the changes religously might have trouble figuring out what's going on, or the significance of that change 3. Waiting until a release is made to find out what's happening, while dramatic, is really a bit of a problem from a planning perspective: people _using_ the software should really know what's changing as it happens. 4. Having an awareness of a change lets people plan for it when it's released 5. Allows interested parties to test and (heaven forbid) document the change, possibly helping you, the developer out. I know that there are some people who feel that this is just extra work. To those, I say that someone is making the entry at the end, why not take a few seconds and describe the change (and take credit for it) when it's added. Not every checkin will prompt an entry, only those that complete a change or addition. If there's any doubt about whether a change merits inclusion, err on the side of over documenting and include it. An example of a few changelog entries I have made that represent a few hours of work on my part: - SSL3 client can now present certificates to a server; server can request and validate a client certificate chain presented. (Bill Welliver) - SSL3 module now checks presented certificate chains for validity by examining effectivity, issuers, and completeness of certification chain. (Bill Welliver) - Added method Tools.X509.verify_certificate_chain() that can verify the validity of a chain of X509 certificates. (Bill Welliver) This took me a total of 2 minutes to do, and might have taken less if the work were fresh in my mind. The third entry was one I included that would probably be cut from a release change log, but is certainly not wasted space. I'd like to see a policy added for 7.7 that says if you check in a completed new or changed feature (module, major work, language behavior, etc), you add a brief entry describing what you've done. Any thoughts? Who can make this happen? If anyone has any comments or would like examples of what I envision, please feel free to contact me. Thanks, Bill

1 0

Looking for translators
by David Gourdelier 30 Mar '04

30 Mar '04

Hi, The Camas team is looking for translators willing to translate the next Camas release in the language you know. For now the languages for which we have an up to date translations are: . English . French . Spanish . Polish And thus the Camas team is looking for anyone who knows a language except these. Mainly we are looking for: . German . Italian * . Swedish * . Finish * . Czech * . Dutch * . Estonian * . Hungarian * . Portuguese * . Russian The languages with a * are languages for which the translation is not up to date for this release. The translation is very easy and it'll require you only a text editor and a few minutes. If you are willing to translate, just send an email to me and I'll send you a file that you'll just need to translate and send me back. Thank you for your help. Best regards. -- David Gourdelier

2 1

Re: ADD_FUNCTION() macro on pike 7.5 ?
by Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum 29 Mar '04

29 Mar '04

>The recommended way about using pike -x module need some special >version of autoconf... that is *not* present on all target machines. You don't need to run autoconf on the target machine. Just run "pike -x module --autoconf" on a machine which has the appropriate autoconf version, and include the generated configure script in the distribution. pike -x module on the target machine will see that the configure script already exists and has a timestamp that is newer than configure.in, and it will not try to rerun autoconf. / Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) Previous text: >2004-03-29 19:11: >Subject: Re: ADD_FUNCTION() macro on pike 7.5 ? >-------------------------------------------------------------------- > >Le 29 mars 04, à 13:35, Martin Nilsson (provokatör) @ Pike (-) >developers forum a écrit : > >>> >>> but outside that, eg custom pike C modules made (hardly) by hand, the >>> .pmod is not automatically generated. >> >> Not if you use the correct build procedure (pike -x module) for >> external >> modules. > >The recommended way about using pike -x module need some special >version of autoconf... that is *not* present on all target machines. > >So wa have to make a stable way to make it. > >>> >>> This one is good.... but external (home made) modules are allways >>> dynamic modules (eg .so)... >> >> No. If the Pike can not handle dynamic modules (because it is not >> possible on that platform) you would compile pike with the external >> modules as static modules. > >So you have to recompile pike isn't it ? > >/Xavier >-- >Xavier Beaudouin - Unix System Administrator & Projects Leader. >President of Kazar Organization : http://www.kazar.net/ >Please visit http://caudium.net/, home of Caudium & Camas projects > > > > / Brevbäraren >

1 0

pike -t
by David Gourdelier 28 Mar '04

28 Mar '04

Hi, I have a strange problem with pike -t in 7.5 and Caudium. When I start Caudium with -t, I have the following error at compile time: /usr/local/pike/7.5.23/lib/modules/Protocols.pmod/HTTP.pmod/Session.pike:1005 Error: Cannot inherit program which is not fully compiled yet. The strange thing is that it doesn't happen if I do pike -t -e 'Protocols.HTTP.Session;' for example. There is a backtrace attached to this mail and as you can see the error is generated from Pike itself (compile_error). Do you have any ideas ? Thank you for your help. / David Gourdelier

1 0

7.5/7.7
by Martin Nilsson (saturator) ＠ Pike (-) developers forum 25 Mar '04

25 Mar '04

There is a lot going on in the 7.5 CVS right now. I suggest that we make some sort of branching, either with CVS branches or simply make a copy of 7.5 to 7.7 so that we can get something stable enough for 7.6 as soon as possible. Do feedback.

13 27

Gmp.mpq
by Martin Nilsson (provokat�r) ＠ Pike (-) developers forum 23 Mar '04

23 Mar '04

Gmp.mpq calls get_mpz_from_digits... > Gmp.mpq("1/3"); Invalid digits, cannot convert to Gmp.mpz. Unknown program: Gmp.mpq(0/1)->create("1/3") HilfeInput:1: HilfeInput()->___HilfeWrapper()

1 0

7.6
by Martin Nilsson (provokat�r) ＠ Pike (-) developers forum 23 Mar '04

23 Mar '04

Great work grubba! I see the following blocker bugs. Please add your own. [ ] Fix dmalloc leak. [ ] Fix Windows java problem. [ ] Fix Windows dynamic load of the Image module (?) [ ] Verify the whole path from (make doc;make export) to make install. [ ] Get Windows packaging to work for 7.5. [ ] Write change log/release notes Now is really the time for last requests from Roxen, Caudium and Aido developers.

8 23

7.6
by Bill Welliver 22 Mar '04

22 Mar '04

Are we talking days, weeks or months before this 7.5/7.7 split? I've got a few non-core items that I think I (or others) need to address before 7.6: 1. Autodocs: there are a number of modules and or functions that aren't documented well or at all (the readline and termcap stuff comes to mind). Also, one of the biggest complaints about the modref is that there are so few examples. If everyone could just take an hour or two to document or expand the documentation for a few functions, it would be a great help. 2. Modules: is there a general feeling about how big a deal external modules are going to be for 7.6? Do we need to test the procedures out more, and have documentation about how developers/end users go about creating and installing them? Should I try to finish my "module grab and install" code? The "server side" of this is almost finished, but the client end is barely started. Anyone want to help out with this? Bill

3 5

cygwin days
by Martin Nilsson (provokat�r) ＠ Pike (-) developers forum 21 Mar '04

21 Mar '04

I have successfully built Pike on cygwin. There are currently a few issues you have to get around. Pike doesn't support cygwin paths. If /cygdrive/x/foo gets handed to a file function it should instead use x:/foo (well, x:\foo). Could this be fixed in fdlib? Stdio.Terminfo can not decode the cygwin terminfo file. sizeof(strbuf)==sstr-2. mt_lock( &aap_timeout_mutex ); in HTTPLoop/accept_and_parse.c triggers some sort of thread bug during startup. Just build without HTTPLoop.

5 11

Jump to page:

2026

2025

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