Pike-devel July 2008

pike-devel@lists.lysator.liu.se

8 participants
41 discussions

GMP 4.1.4 -> 4.2.1, Nettle 1.14 -> 1.15
by Peter Bortas ＠ Pike developers forum 15 Jul '08

15 Jul '08

FYI: The problems with libnettle 1.15 was fixed earlier this week and it's now part of the snapshots. And since I've been using libgmp 4.2.1 for quite some time now I upgraded the 4.1.4 used by the snapshot builder. These are also the versions that will be included in dists unless 4.2.2 has some compelling fixes we need. Speaking of, where has http://www.swox.com/gmp/ gone? There is no GMP page there any more, much less a changelog...

2 4

Protocols.HTTP.http_encode_string
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 15 Jul '08

15 Jul '08

> functionality. An IRI can e.g. contain "ôôÌ" in a unicode context Hmm, that shows that lyskom (or at least my client) isn't a unicode environment. The gibberish there was meant to be a literal euro sign.

2 1

Protocols.HTTP.http_encode_string
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 15 Jul '08

15 Jul '08

Correction: Standards.URI.http_encode does not encode any 8-bit chars, the unused Standards.URI.quote does.

2 3

Process.run
by Peter Bortas (nu med mindre Opera) ＠ Pike (-) developers forum 14 Jul '08

14 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

hilfe completion fixes
by Martin Bähr 14 Jul '08

14 Jul '08

hi, the following hilfe completion fixes will be commited tonight (just got a train to catch before i can do it) [*] fix cases where complete symbols are a prefix for other symbols [*] object member variables should not complete with symbols [*] whitespace after a symbol should offer all infix operators [*] symbol followed by space at beginning of line (or after ;) should complete with operators [*] = should not complete with closing brackets greetings, martin.

1 0

Latest type checks
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 13 Jul '08

13 Jul '08

Errata: It should be "error_type" instead of "error_name" to adhere to the Error.Generic interface.

1 0

7.7/precompile.pike
by H. William Welliver III 11 Jul '08

11 Jul '08

I just tried compiling an external module and noticed that include/ pike/precompile.pike has as its primary line: #include "../lib/modules/Tools.pmod/Standalone.pmod/precompile.pike" which isn't correct when Pike is installed. I temporarily changed the line to add an extra "../" but I know that's not the correct long term solution. Any suggestions? Bill

3 2

pike 7.6.112 bus error while running build
by H. William Welliver III 06 Jul '08

06 Jul '08

I'm in the process of getting pike set up on my new laptop, and I'm having a problem getting things compiled. seems like pike is having problems running during the build. this is osx 10.5.2/intel: Starting program: /Users/hww3/Downloads/Pike-v7.6.112/build/ darwin-9.2.1-i386/pike -DNOT_INSTALLED -DPRECOMPILED_SEARCH_MORE -m/ Users/hww3/Downloads/Pike-v7.6.112/build/darwin-9.2.1-i386/ master.pike /Users/hww3/Downloads/Pike-v7.6.112/src/post_modules/ Shuffler/make_sources.pike /Users/hww3/Downloads/Pike-v7.6.112/src/ post_modules/Shuffler sources.h sources_to_compile Reading symbols for shared libraries ++++....................... done Program received signal EXC_BAD_ACCESS, Could not access memory. Reason: KERN_PROTECTION_FAILURE at address: 0x00000058 0x00012822 in low_mega_apply (type=APPLY_LOW, args=1, arg1=0x818bb4, arg2=0xffffffff) at apply_low.h:115 115 function = new_frame->context.prog->identifiers + ref- >identifier_offset; (gdb) bt #0 0x00012822 in low_mega_apply (type=APPLY_LOW, args=1, arg1=0x818bb4, arg2=0xffffffff) at apply_low.h:115 #1 0x00013900 in mega_apply (type=APPLY_LOW, args=1, arg1=0x818bb4, arg2=0xffffffff) at /Users/hww3/Downloads/Pike-v7.6.112/src/ interpret.c:2004 #2 0x00015bea in call_handle_error () at /Users/hww3/Downloads/Pike- v7.6.112/src/interpret.c:2113 #3 0x000880f2 in main (argc=8, argv=0xbffff59c) at /Users/hww3/ Downloads/Pike-v7.6.112/src/main.c:786 any thoughts? bill

2 2

Secure strings
by Martin Stjernholm, Roxen IS ＠ Pike developers forum 04 Jul '08

04 Jul '08

Regarding the new String.secure feature, what do you think about censoring such strings in sprintf("%O") output? That'd make it very useful to avoid password leakage in backtraces etc. All other operations on it - especially sprintf("%q") - would continue to work as usual.

8 40

comments on "simplified method for external modules"
by Bill Welliver 01 Jul '08

01 Jul '08

I've been following the changes file and just thought I would mention that the subject example includes GNU-makeisms that aren't compatible with standard make. Also, is it appropriate to recommend that users not use the standard module build method? The sample, while simple, seems to require the inclusion of a lot of rules that already exist in dynamic_module_makefile, however without a lot of the edge case handling. I'm all for simplification, but I'm concerned that recommending users write modules in this manner will result in a lot of modules floating around that don't build or install properly. Bill

4 6

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 July 2008