Pike-devel

pike-devel@lists.lysator.liu.se

4 participants
3044 discussions

Start a nNew thread

Re: Bzip2 (Pike C Module documentation)
by Niels M�ller () ＠ Pike (-) developers forum 16 Oct '02

16 Oct '02

Is there any spec for the bzip2 format? Analogous to the three RFC:s (DEFLATE, ZLIB, GZIP) that defines gz?

1 0

Re: Bzip2 (Pike C Module documentation)
by Niels M�ller () ＠ Pike (-) developers forum 16 Oct '02

16 Oct '02

IIRC, the algorithms used by bzip works with fairly large blocks. So if you need to sync often (say, two sync points less than a KB or so apart, you probably shouldn't use bzip2). / Niels Möller () Previous text: >2002-10-16 16:15: >Subject: Re: Bzip2 (Pike C Module documentation) >-------------------------------------------------------------------- >*reads docs* > >Reading the libbzip2 docs, it seems that the library is a tad hairy in >this regard. There doesn't seem to be any real support for reading or >writing partially synced bzip2'd streams. There is no sync command, >only end. And the only support I can see about reading multiple ended >compressions from one stream is to check whether there's data left in >the "in" stream after decompression. > >(One should not confuse the high level interface that uses FILE* with >reading/writing partially synced streams.) > > > / Brevbäraren >

1 0

Re: Bzip2 (Pike C Module documentation)
by Johan Sundstr�m (�rkehertig av Lysators rootgrupp) ＠ Pike (-) developers forum 16 Oct '02

16 Oct '02

1 0

Re: Bzip2 (Pike C Module documentation)
by Martin Nilsson (Fake Build Master) ＠ Pike (-) developers forum 16 Oct '02

16 Oct '02

You can use the PExt module for now. We haven't found that many bugs in it... It doesn't support streaming though. / Martin Nilsson (Fake Build Master) Previous text: >2002-10-16 13:54: >Subject: Re: Bzip2 (Pike C Module documentation) >-------------------------------------------------------------------- >I need Bzip2 now. ETA for any of those modules? > > > / Brevbäraren >

2 1

cvs?
by Martin Nilsson (Fake Build Master) ＠ Pike (-) developers forum 16 Oct '02

16 Oct '02

That is the current anon cvs root. The development cvs root at pike.ida.liu.se is only available through named ssh. Make a private messsage to Johan Sundström for development access. / Martin Nilsson (Fake Build Master) Previous text: >2002-10-16 09:38: >Subject: cvs? >-------------------------------------------------------------------- >What is the new, current cvsroot? > >pike.ida.liu.se claims that the most recent version is 7.3, >at :pserver:anon@cvs.roxen.com:/cvs/Pike/7.3. > >(Speaking of which, might I get access to the devel cvs again?) > > > / Brevbäraren >

1 0

Pikefarm doesn't generate new dists properly
by Martin Nilsson (Fake Build Master) ＠ Pike (-) developers forum 15 Oct '02

15 Oct '02

But I can change the values in the database until the real fix propagates.

1 0

Broken tests
by Martin Nilsson (Fake Build Master) ＠ Pike (-) developers forum 15 Oct '02

15 Oct '02

How embarrasing. I forgot a scheduled down time of the netowrk and was cut off in mid-checkin...

5 19

Shifts
by Marcus Comstedt (ACROSS) (Hail Ilpalazzo!) ＠ Pike (-) developers forum 15 Oct '02

15 Oct '02

I'm suspecting the shift operators are not really up to spec: case 1: > 17<<-10; (24) Result: 71303168 > Gmp.mpz(17)<<Gmp.mpz(-10); mpz->lsh on negative number. /pike/home/marcus/Pike/7.3/build/lib/modules/Gmp.so.mpz: 17->`<<(-10) HilfeInput:1: ___HilfeWrapper() > case 2: > (-10)>>30; (39) Result: -1 > (-10)>>31; (40) Result: -1 > (-10)>>32; (41) Result: 0 > Gmp.mpz(-10)>>Gmp.mpz(32); (42) Result: -1 > In the first case, shifting with a negative count should give an error just like the Gmp implementation. In the second case, increasing the shift count should not suddenly change the value once it has stabilized on -1 (again, Gmp does the expected thing). Right?

2 2

Caudium.NBIO
by David Hedbor ＠ Pike developers forum 14 Oct '02

14 Oct '02

Is there any interest in integrating Caudium.NBIO into Pike (post 7.4 split I assume). Caudium.NBIO is a non-blocking io module similar to Pipe.pipe (somewhat less features, but better performance and stability). It's bee used in Caudium for quite some time and might be useful for Pike in general (For example, Roxen would benefit from it - it outperforms Roxen 2.x in data sending speed for large files).

2 4

Newly built Pike segfaults
by David Hedbor ＠ Pike developers forum 14 Oct '02

14 Oct '02

Hmm. This is not good. gcc 3.2, just checked out of CVS, clean tree and I get this: 2050 s->s->str[s->s->len << s->s->size_shift] = 0; (gdb) bt #0 0x080f1875 in string_builder_append (s=0xbffff1b8, from= {ptr = 0x82e6cc4 "/home/neotron/pike/build/7.3-nodebug-asm/lib/modules/Getopt.pmod", shift = -1073745736}, len=1) at /home/neotron/pike/7.3/src/stralloc.c:2050 #1 0x0811783c in append_path_unix (s=0xbffff1b8, path= {ptr = 0x82e6cc4 "/home/neotron/pike/build/7.3-nodebug-asm/lib/modules/Getopt.pmod", shift = 0}, len=64) at /home/neotron/pike/7.3/src/combine_path.h:73 #2 0x081179be in f_combine_path_unix (args=2) at /home/neotron/pike/7.3/src/stralloc.h:89 #3 0x0806ecfb in opcode_F_CALL_BUILTIN_AND_RETURN (arg1=-1073745496) at /home/neotron/pike/7.3/src/interpret_functions.h:2059 #4 0x0831a15a in ?? () #5 0x08071510 in mega_apply (type=3221221800, args=-1073745496, arg1=0xbffff1a8, arg2=0xbffff1a8) at /home/neotron/pike/7.3/src/interpret.c:1415 #6 0x08072139 in apply (o=0x402d7000, fun=0xbffff1a8 "\200m.\b\220\034\017\b\001", args=65536) at /home/neotron/pike/7.3/src/interpret.c:1718 #7 0x080c18fd in main (argc=7, argv=0xbffff3d4) at /home/neotron/pike/7.3/src/main.c:726 #8 0x400b7082 in __libc_start_main () from /lib/i686/libc.so.6 (gdb) p s $1 = (struct string_builder *) 0xbffff1b8 (gdb) p *s $2 = {s = 0xbffff1a8, malloced = 256, known_shift = 0} (gdb) p *s->s $3 = {refs = 137260416, size_shift = 135208080, len = 1, hval = 27, next = 0xbffff1a8, str = ""} (gdb) p from $4 = { ptr = 0x82e6cc4 "/home/neotron/pike/build/7.3-nodebug-asm/lib/modules/Getopt.pmod", shift = -1073745736}

2 2

Jump to page:

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