Pike-devel

pike-devel@lists.lysator.liu.se

3044 discussions

unnecessary tpike
by David Hedbor ＠ Pike developers forum 18 Feb '03

18 Feb '03

So, I'm compiling with icc. Problem is that with icc, linking is slow when you have IPO enabled since it performs multi-file optimizations. Pike will ALWAYS create a tpike binary even though there is a working Pike installed (so that step is unnecessary). It adds a significant time to the compilation (several minutes I'd guess). How about, if possible, use the system-installed Pike to run the generation of peep.c etc, skipping the redundant 'tpike' step. I can understand why there's been no optimizations done here since in general it doesn't take a whole lot of time.

5 11

setsid
by Peter Bortas ＠ Pike developers forum 16 Feb '03

16 Feb '03

I have a sudden urge to remove the only expect script on my systems. To do that I have to be able to set passwords with Solaris passwd. I've tried this once before, and failed. Short version of the expect script: ----8<------------------------------------- spawn /usr/bin/passwd -r nis $uname expect { -re assword: { send "$password\n" } } expect { -re assword: { send "$password\n" } } expect { eof { wait } } ----8<------------------------------------- In pike that should be something like this: ----8<------------------------------------- #!/usr/bin/env pike array(Stdio.File) make_ptypair() { Stdio.File pf = Stdio.File(), tf = Stdio.File(); foreach(get_dir("/dev/"), string pty_name) if(pty_name[..2]=="pty" && pf->open("/dev/"+pty_name, "rw")) { if(tf->open("/dev/t"+pty_name[1..], "rw")) return ({ pf, tf }); pf->close(); } return 0; } int main() { string luser = "victim"; array(Stdio.File) f = make_ptypair(); if(!f) { werror("No PTYs available.\n"); return 1; } object p = Process.create_process( ({ "/usr/bin/passwd", luser }), ([ "setsid":f[1] ]) ); f[1]->close(); f[0]->write("bogus\n"); f[0]->write("bogus\n"); return p->wait(); } ----8<------------------------------------- But no, that still gives the password question on the command line: [root] ./setsid.pike New password: Any ideas?

6 20

pike poster design?
by Martin Baehr 16 Feb '03

16 Feb '03

hi, sTeam will be present at the CeBit in germany next month. since it is a pike application i would like to use the opportunity and also advertize pike itself. unfortunately i don't have any design skills. is there any one among this group who might help us out with designing a poster that can be used at various events, such as the CeBit, linux expos/conferences, the pike conference or others? (only a digital rendition would be needed, as we have printing facilities at the university) having such a poster would help the pike project a great deal greetings, martin.

3 3

SDL on NT
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 14 Feb '03

14 Feb '03

post_modules/SDL/configure.in in 7.4 contains this oddity: if test x"$pike_cv_sys_os" = xWindows_NT ; then enable_sdltest=no fi It causes builds to fail completely on NT systems that doesn't have SDL installed (e.g. http://pike.ida.liu.se/development/pikefarm/ result.xml?id=123_15&pike=7_4). If the SDL test doesn't work for some reason, isn't it better to give --disable-sdltest on the specific systems where it's known to be installed?

2 2

Protocols.HTTP.Session
by Mirar ＠ Pike developers forum 14 Feb '03

14 Feb '03

I started to write a Protocols.HTTP.Session class. The goal is an interface to keep cookies throughout a session, and in the future to reuse keep-alive connections, as well as adding better methods for async operations. It can also follow redirects automatically. Example (the Elfwood initial page redirects to the main page if a cookie is set, and it is the first time): | > object s=Protocols.HTTP.Session(); s->get_url("http://elfwood.lysator.liu.se/"); s->get_url("http://elfwood.lysator.liu.se/"); | (1) Result: Request("http://elfwood.lysator.liu.se/" - connected - 361 bytes recieved) | (2) Result: Request("http://elfwood.lysator.liu.se/elfwood.html" - connected - 343 bytes recieved) (Sidenote: What's the autodoc keyword for known bugs? Am I right in guessing it was @known_bugs?)

2 2

Re: Weird problem with Image.SVG in Pike 7.4/7.5
by David Hedbor ＠ Pike developers forum 13 Feb '03

13 Feb '03

Again, gdb/strace output would be good. That way there would be a way to see where it actually dies. By the way, an unrelated suggestion - how about adding support for the 'backtrace()' system call on platforms where it's available? It can be used to, on crashes (or demand) craate a stack trace such as this: Stack dump: { /lib/i686/libpthread.so.0 [0x4003710c] /lib/i686/libc.so.6 [0x401684c8] majesty(Read__17CYLinuxFileSystemiPvUl+0x1d) [0x8268bc1] majesty(Unserialize__11CYLibHeaderi+0x2b) [0x8200b8b] majesty(__11CYLibHeaderi+0x2e) [0x82017d6] majesty(AddLibrary__12CYLibManagerRC8CYString+0x14e) [0x820469e] majesty(AddStaticCAMs__Fv+0x6b) [0x830f0e7] majesty(main+0x5bf) [0x81dcbfb] /lib/i686/libc.so.6(__libc_start_main+0xa2) [0x401573e2] majesty(XMapRaised+0x31) [0x81368b1] } It's very specific to Linux (afaik) but can prove very useful when trying to debug someone elses problem (i.e it's more information than you would have received otherwise). / David Hedbor Previous text: >2003-02-12 22:31: >Subject: Re: Weird problem with Image.SVG in Pike 7.4/7.5 >-------------------------------------------------------------------- >I have just confirmed the suspicion - code compiled with gcc 2.95 and full >optimization works fine. So, now to see what gets overoptimized with gcc 3.2 >-O2 > >marek > > > / Brevbäraren >

4 7

post_modules/_ADT
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 13 Feb '03

13 Feb '03

post_modules/_ADT doesn't build. Forgot to check in array_adapter.h, perhaps? A fix so that the xenofarm clients gets through the build process would be appreciated. make[4]: Entering directory `/home/mast/Pike/7.5/build/linux-2.4.20-i686/post_modules/_ADT' /home/mast/Pike/devel/build/linux-2.4.20-i686/precompile.sh --cache precompile.pike >"/home/mast/Pike/devel/src/post_modules/_ADT/adt.c" "/home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod" || { rm "/home/mast/Pike/devel/src/post_modules/_ADT/adt.c"; exit 1; } precompile: /home/mast/Pike/devel/build/linux-2.4.20-i686/tpike -DNOT_INSTALLED -DPRECOMPILED_SEARCH_MORE -m/home/mast/Pike/devel/build/linux-2.4.20-i686/master.pike /home/mast/Pike/devel/bin/precompile.pike /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod (method=QQ) /home/mast/Pike/devel/lib/modules/ADT.pmod/module.pmod:1:Illegal program pointer. Compiling /home/mast/Pike/devel/src/post_modules/_ADT/adt.c /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod:41:27: array_adapter.h: No such file or directory /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod: In function `pike_module_init': /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod:48: warning: implicit declaration of function `pike_init_ArrayAdapter_module' /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod: In function `pike_module_exit': /home/mast/Pike/devel/src/post_modules/_ADT/adt.cmod:54: warning: implicit declaration of function `pike_exit_ArrayAdapter_module' Compilation command was: /home/mast/Pike/devel/build/linux-2.4.20-i686/smartlink gcc -I. -I/home/mast/Pike/devel/src/post_modules/_ADT -I/home/mast/Pike/devel/build/linux-2.4.20-i686 -I/home/mast/Pike/devel/src -I/usr/local/include -I/usr/X11R6/include -g -mcpu=i686 -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 /home/mast/Pike/devel/src/post_modules/_ADT/adt.c -o adt.o

2 1

Search bug ?
by Mirar ＠ Pike developers forum 13 Feb '03

13 Feb '03

And why isn't this a compile error? | > "hej"=="hej"; | (2) Result: 1 It's not meaningful either.

6 17

Weird problem with Image.SVG in Pike 7.4/7.5
by Marek Habersack 12 Feb '03

12 Feb '03

Hey all, I'm having a problem with Image.SVG in that when trying to do indices(Image.SVG) then Pike hangs quite badly - the only way to stop it is to kill it. It happens only in the Debian packaged version for both Pike versions, the version compiled from the tarballs works fine. Until recently the Debian package was using the --traditional style of installation, but now it uses a modified --new-style one which in 90% resembles the default style with minor differences: - all the pike code is installed in /usr/share/pike/X.Y.Z/ - all the binary files (.o, pike, .so) are in /usr/lib/pike/X.Y.Z/ - the files in /usr/share/pike/X.Y.Z/ are symlinked to from the /usr/lib/pike/X.Y.Z/ so that the layout looks like that: $ pwd /usr/lib/pike/7.5.3/lib/modules ... lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_DWG.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_DWG.pmod -rw-r--r-- 1 root root 6062 2003-02-06 19:40 _Image_DWG.pmod.o -rw-r--r-- 1 root root 10908 2003-02-06 19:42 _Image_FreeType.so -rw-r--r-- 1 root root 39188 2003-02-06 19:42 _Image_GIF.so -rw-r--r-- 1 root root 32844 2003-02-06 19:42 _Image_JPEG.so drwxr-xr-x 2 root root 94 2003-02-06 23:40 _Image.pmod lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_PSD.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_PSD.pmod -rw-r--r-- 1 root root 20777 2003-02-06 19:40 _Image_PSD.pmod.o lrwxrwxrwx 1 root root 58 2003-02-06 23:40 _Image_PS.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_PS.pmod -rw-r--r-- 1 root root 8087 2003-02-06 19:40 _Image_PS.pmod.o -rw-r--r-- 1 root root 768976 2003-02-06 19:42 Image.so -rw-r--r-- 1 root root 8156 2003-02-06 19:42 _Image_SVG.so -rw-r--r-- 1 root root 2964 2003-02-06 19:42 _Image_TIFF.so -rw-r--r-- 1 root root 22540 2003-02-06 19:42 _Image_TTF.so lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_XCF.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_XCF.pmod -rw-r--r-- 1 root root 45024 2003-02-06 19:40 _Image_XCF.pmod.o -rw-r--r-- 1 root root 10756 2003-02-06 19:42 _Image_XFace.so lrwxrwxrwx 1 root root 59 2003-02-06 23:40 _Image_XPM.pmod -> ../../../../../share/pike/7.5.3/lib/modules/_Image_XPM.pmod -rw-r--r-- 1 root root 11477 2003-02-06 19:40 _Image_XPM.pmod.o lrwxrwxrwx 1 root root 53 2003-02-06 23:40 Java.pmod -> ../../../../../share/pike/7.5.3/lib/modules/Java.pmod -rw-r--r-- 1 root root 44761 2003-02-06 19:40 Java.pmod.o so, as far as pike is concerned, the entire tree is in /usr/lib/pike/X.Y.Z/, the module/include paths are: $ pike7.5 --show-paths master.pike...: /usr/lib/pike/7.5.3/lib/master.pike Module path...: /usr/local/share/pike7.5/site_pike /usr/local/share/pike/site_pike /usr/local/lib/pike7.5/site_pike /usr/local/lib/pike/site_pike /usr/lib/pike/7.5.3/lib/modules Include path..: /usr/local/include/pike7.5 /usr/local/include/pike /usr/lib/pike/7.5.3/lib/include Program path..: $ pike7.4 --show-paths master.pike...: /usr/lib/pike/7.4.13/lib/master.pike Module path...: /usr/local/share/pike7.4/site_pike /usr/local/share/pike/site_pike /usr/local/lib/pike7.4/site_pike /usr/local/lib/pike/site_pike /usr/lib/pike/7.4.13/lib/modules Include path..: /usr/local/include/pike7.4 /usr/local/include/pike /usr/lib/pike/7.4.13/lib/include Program path..: Running pike with -t gives the following result (output comes from pike 7.5): - master.pike: 688: 8298f90->low_findprog("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG",".so",0,1) - master.pike: 516: 8298e58->lock(2) - Return: module.c:67()->Builtin()->mutex_key() - master.pike: 555: 8298f90->master_file_stat("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - master.pike: 295: 8298f90->combine_path_with_cwd("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - Return: "/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so" - master.pike: 297: 8298f90->dirname("/usr/lib/pike/7.5.3/lib/modules/_Image_SVG.so") - Return: "/usr/lib/pike/7.5.3/lib/modules" - master.pike: 325: 82988b8->create() - Return: Stat(-rw-r--r-- 8156b) - master.pike: 555: 82988b8->[](1) - Return: 8156 - master.pike: 560: 82988b8->[](3) - Return: 1044556949 the above lines are the very last ones in the output - after that pike is hung. Trying to call Image.SVG() renders the same effect - pike is hung. My next step will be to compile the .deb with full debugging info, but I thought that somebody might have a clue as to what's wrong. TIA, marek

4 13

Weird problem with Image.SVG in Pike 7.4/7.5
by David Hedbor ＠ Pike developers forum 12 Feb '03

12 Feb '03

Naturally, make that s/gcc/gdb/

1 0

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