Pike-devel April 2006

pike-devel@lists.lysator.liu.se

7 participants
24 discussions

managing comport settings in linux
by Henrik Grubbstr�m (Lysator) ＠ Pike (-) developers forum 10 Apr '26

10 Apr '26

Not a good idea; the design philosophy of Pike is that Pike-level code should never be able to cause memory corruption at C-level. With ioctls there's no way to know what data a specific ioctl wants without reading the documentation for that specific ioctl for that specific OS. ioctls and fcntls are a way for device handlers to hook in APIs that aren't standard across all filedescriptors. / Henrik Grubbström (Lysator) Previous text: >2004-08-17 13:32: >Subject: managing comport settings in linux >-------------------------------------------------------------------- >What would be nice, is a way to define ioctl methods in >Pike. Perhaps something like > >object i = Stdio.IOCTL(4242 /* request_id */ ); >i->add_int_parameter(); >i->add_string_parameter(); >i->set_return_type(STRING); > >string ret = i(fd, 1, "somestring"); > > >That ought to be possible, or..? > > / Marcus Agehall (PacketFront) >

4 3

Re: Playing music using pike and SDL.
by Martin Nilsson (lvl 60) ＠ Pike (-) developers forum 13 Mar '26

13 Mar '26

Were you not able to get any C code to play through SDL either? Does your SDL library have a mixer compiled in it (I think it is optional)?

5 4

full MySQL 4.x support. or pike roadmap
by thomas.gusenleitner＠com-bet.com ＠ Pike developers forum 01 Mar '26

01 Mar '26

HI to all! mysql 4.x support works only with a mysql CFLAGS=-DUSE_OLD_FUNCTIONS is this correct ? are there plans to make a full 4.x support ? bye, thomas

4 4

Process.run
by Peter Bortas (nu med mindre Opera) ＠ Pike (-) developers forum 15 Jul '08

15 Jul '08

I just checked in a Process.run() in 7.7: http://pike.ida.liu.se/development/cvs/diff.xml?module=Pike&file=7.7/lib/mo… The rational for that is that I was once again reminded by a user this week that getting Process.create_process() or Process.Process right without some corner case blocking is hard. It's not obvious that if you bind pipes to booth stderr and stdout you have to take in to account that OS buffers are not limetless. Process.run() is a thin wrapper around Process.Process that collects stdout, stderr and exitcode and return them in a mapping. This could potentially result in huge strings, but will do what most users want and even already do when they call create_process() with both pipes bound. Something like this should be in 7.8, but I'm open to suggestions for other things that might be usefull to return in the mapping or for changing member names.

5 7

Pike 7.6
by Martin Nilsson (DivX Inc.) ＠ Pike (-) developers forum 23 Aug '06

23 Aug '06

It is time for a new release of Pike 7.6. Is there anyone who can do the release notes? If not, we'll just wait a bit longer...

9 12

rntcl problems
by Niels M�ller (hemma) ＠ Pike (-) developers forum 11 May '06

11 May '06

I just noticed these errors and warnings from the argp build log of kali.roxen.com, http://www.lysator.liu.se/xenofarm/lsh/files/483_81/argpmakelog.txt make[2]: Entering directory `/home/build/xenofarm/client/lsh/kali.roxen.com/buildtmp/dist/lsh-2.0.3/src/argp/testsuite' source='ex1.c' object='ex1.o' libtool=no \ depfile='.deps/ex1.Po' tmpdepfile='.deps/ex1.TPo' \ depmode=none /bin/bash ../depcomp \ rntcl -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I. -g -c ex1.c Command line warning D4025 : overriding '/Z7' with '/Zi' rntcl -g -o ex1.exe ex1.o ../libargp.a 1! Command line warning D4024 : unrecognized source file type '..\libargp.a', object file assumed source='ex3.c' object='ex3.o' libtool=no \ depfile='.deps/ex3.Po' tmpdepfile='.deps/ex3.TPo' \ depmode=none /bin/bash ../depcomp \ rntcl -DHAVE_CONFIG_H -I. -I. -I.. -I.. -I. -g -c ex3.c Command line warning D4025 : overriding '/Z7' with '/Zi' 2! RNTCL: output file not generated (ex3.obj). make[2]: *** [ex3.o] Error 1 The line marked (1!) seems to indicate that rntcl doesn't support explicit library arguments on the command line. Is that right? For the line (2!), I see no clues as to why the compilation fails to produce an object file. I also get the same "RNTCL: output file not generated" error for one of the test programs in nettle's configure, configure:2041: checking for suffix of object files configure:2062: rntcl -c -I/usr/local/include -I/home/build/xenofarm/client/lsh/kali.roxen.com/buildtmp/dist/pfx/include conftest.c >&5 RNTCL: output file not generated (conftest.obj). configure:2065: $? = 1 configure: failed program was: | /* confdefs.h. */ | | #define PACKAGE_NAME "nettle" | #define PACKAGE_TARNAME "nettle" | #define PACKAGE_VERSION "1.15" | #define PACKAGE_STRING "nettle 1.15" | #define PACKAGE_BUGREPORT "nettle-bugs(a)lists.lysator.liu.se" | /* end confdefs.h. */ | | int | main () | { | | ; | return 0; | } configure:2079: error: cannot compute suffix of object files: cannot compile (see http://www.lysator.liu.se/xenofarm/lsh/files/483_81/nettleconfiglog.txt)

2 3

hilfe spewing 00000000
by Adam Montague 29 Apr '06

29 Apr '06

I am trying to get an interactive mode working for my app, to make debugging it easier. I created a "Hilfe" class that inherits Tools.Hilfe.StdinHilfe, and if the app is called with -i, I call add_constant() on this, call Hilfe(), and then return. But when I start my app with -i, I get the welcome to hilfe message, and then it starts spewing 0 at me forever, I am back at my shell, and I have to kill the pike process to get the 0s to stop. Adam

2 4

more on module installation and upgrading
by Bill Welliver 27 Apr '06

27 Apr '06

one thing that hasn't been given a great deal of consideration regarding external modules is the problem of upgrading. If you upgrade a module that's already installed, currently there's no way to know what components make up the existing version of the module. That means the upgraded version may not replace all existing components, leaving a certain amount of potentially problem causing cruft. A good example is when a module gets refactored from one file into multiple files. It becomes an even bigger pain when you start to allow nested modules, such as Public.Parser.XML2 and Public.Parser.XML2.Schematron, for example. I imagine that there'd need to be some sort of registry that described the files comprising the module. In theory, it could be as simple as an array constant in the module that lists all of the constituents, much as we currently include the version and author for use by the Monger system. It would be relatively simple to include this functionality in the 7.8 release of monger, if there wasn't any major objection to this. Also, I haven't heard any objection to the idea of tagging modules in the dist so that they can be downloaded separately. I assume that everyone was at least neutral on the idea? I vaguely remember there being some sort of build environment change recommended by Grubba or Mast... what was that exactly? Any thoughts, comments? Bill

1 0

SQLite
by Marc Dirix 20 Apr '06

20 Apr '06

This might be a configure bug, although it doesn't find SQLite prerequisites it tries to compile anyway. I'm trying to compile pike 7.7.25 from cvs. Output of configure is: SQLite.............. no (dependencies failed) Output of make is: Compilation command was: /usr/src/pike/bin/smartlink gcc -I. -I/usr/src/pike/src/post_modules/ SQLite -I/usr/src/pike/src -I/usr/src/pike/src -I/usr/src/pike/src/ bundles/include -I/usr/local/include -I/usr/X11R6/include - DHAVE_CONFIG_H -g -ggdb3 -mcpu=i686 -I/usr/src/pike/src/bundles/ include -I/usr/local/include -I/usr/X11R6/include -W -Wall -Wno- unused -Wcomment -Wformat -Wimplicit-function-declaration -Wmultichar -Wswitch -Wuninitialized -Wpointer-arith -Wchar-subscripts -Wno-long- long -c sqlite.c -o sqlite.o make[2]: *** [sqlite.o] Error 1 make[2]: Leaving directory `/var/src/pike/src/post_modules/SQLite' make[1]: *** [SQLite] Error 1 Best regards, Marc Dirix

3 10

strcpy overflow in modules/files/file.c
by Adam Montague 20 Apr '06

20 Apr '06

In the function file_connect_unix(), Pike_sp[-args].u.string->len is checked against PATH_MAX, and then its strcpy'd into name.sun_path. The size of a sockaddr_un.sun_path isn't standard, and can be much smaller than PATH_MAX. How should this be fixed, changing the PATH_MAX test to a sizeof(name.sun_path)? Or removing that check altogether, using strncpy and then checking for truncation and giving a "path too long" error instead of the current "Illegal argument. Expected string(8bit)"? Is it worth adding configure tests for strlcpy/cat and using those instead of strncpy/cat? Adam

4 6

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 April 2006