Pike-devel September 2009

pike-devel@lists.lysator.liu.se

3 participants
5 discussions

Pike 7.8.352 released
by Peter Bortas ＠ Pike developers forum 24 Sep '09

24 Sep '09

The Pike developers are proud to present Pike 7.8.316, a new stable release. Download from http://pike.ida.liu.se/download/pub/pike/latest-stable/ Changes since Pike 7.8.316 (second 7.8 release): ---------------------------------------------------------------------- General ------- o ADT.Heap The compat functions top() and size() are now marked as deprecated. Extensions and New Functions ---------------------------- o Image.Image Added support for sending Color objects to the color() function. o Sql.Sql Added support for fetching result sets as JSON-encoded strings. o Calendar Updated timezone data to 2009j. o Parser.XML.Tree Output from render_xml() is now on canonical form. o Added "pike -x httpserver" that implements a minimal HTTP-server exporting the current directory. Optimizations ------------- o Lowered startup time by extending encode_value() and MasterCodec so that master.pike can be precompiled. o Lowered startup time by reducing the number of MEMSET() calls done on startup. o Allow threads while calling SDL_GL_SwapBuffers. o Added optimization of has_value() and search() for single-character strings. o Added optimization when combining two arrays where the second array contains a single item. Bug fixes --------- o Fixed casting of floats to strings to again work more like in Pike 7.6 and earlier: The string will always contain either a "." or an "e" to signify that it is a float, and it will no longer display so many digits that the binary/decimal conversion errors become visible. o Protocols.HTTP.Server.HeaderParser will now discard parts of a header value if the header value contains a newline (like "Host: google.com\n.hax.net\r\n") and headers without colon. o file_open_socket() now attempts to set SO_REUSEPORT on the socket. Potentially fixes issues on FreeBSD 7.x where ports aren't reused. o Calendar.TimeRanges now knows that `+() and `-() can get any number of arguments. o The experimental extended ZFS support is disabled since it pulled in unwanted dependencies. o Fixed propagation of changed module directories for joinnodes by zapping the joinnode cache. o Fixed memory leak in Image.WBF. o Fixed potential NULL-deref in Image.XWD. o Fixed potential NULL-deref in Image.PNG. Building and installing ----------------------- o Nettle 2.0 is now supported.

2 2

Bundles
by Magnus Holmgren, Millnet/Lysator/Debian/Mensa ＠ Pike developers forum 23 Sep '09

23 Sep '09

To avoid freeness issues etc., I would be very happy if there were a variant of the Pike source tarball without bundled libraries. From debian/copyright: > The upstream tarball has been repacked without the 'bundles' > subdirectory, which contains redundant copies of some libraries, to > avoid shipping some non-DFSG-compliant bits (see the gmp package) > and having to include the copyright information and licenses of > those libraries. In any case, why is there both a libffi-3.0.4.tar.gz and its unpacked contents?

4 5

Pike 7.8.346 beta
by Peter Bortas ＠ Pike developers forum 22 Sep '09

22 Sep '09

New build that I hope to release as a stable towards the end of the week: http://pike.ida.liu.se/download/pub/pike/beta/7.8.346/ Windows build avilable. Changes since Pike 7.8.316 (second stable 7.8 release): ------------------------------------------------------- General ------- o ADT.Heap The compat functions top() and size() are now marked as deprecated. Extensions and New Functions ---------------------------- o Image.Image Added support for sending Color objects to the color() function. o Sql.Sql Added support for fetching result sets as JSON-encoded strings. o Calendar Updated timezone data to 2009j. o Added "pike -x httpserver" that implements a minimal HTTP-server exporting the current directory. Optimizations ------------- o Lowered startup time by extending encode_value() and MasterCodec so that master.pike can be precompiled. o Lowered startup time by reducing the number of MEMSET() calls done on startup. o Allow threads while calling SWL_GL_SwapBuffers. o Added optimization of has_value() and search() for single-character strings. o Added optimization when combining two arrays where the second array contains a single item. Bug fixes --------- o Fixed casting of floats to strings to again work more like in Pike 7.6 and earlier: The string will always contain either a "." or an "e" to signify that it is a float, and it will no longer display so many digits that the binary/decimal conversion errors become visible. o Protocols.HTTP.Server.HeaderParser will now discard parts of a header value if the header value contains a newline (like "Host: google.com\n.hax.net\r\n") and headers without colon. o file_open_socket() now attepts to set SO_REUSEPORT on the socket. Potentially fixes issues on FreeBSD 7.x where ports aren't reused. o Calendar.TimeRanges now knows that `+() and `-() can get any number of arguments. o The experimental extended ZFS support is disabled since it pulled in unwanted dependencies. o Fixed propagation of changed module directories for joinnodes by zapping the joinnode cache. o Fixed memory leak in Image.WBF. o Fixed potential NULL-deref in Image.XWD. o Fixed potential NULL-deref in Image.PNG. Building and installing ----------------------- o Nettle 2.0 is now supported.

6 13

multi-m_delete
by Arjan van Staalduijnen 09 Sep '09

09 Sep '09

In various parts of my code I wish to remove a set of indices from a mapping. According to the documentation the index to be removed can be a mixed type, so it seems I can only remove indices from a mapping by performing m_delete's one-by-one. Code like: array removable_indices = ({ "a", "b", "c" }); mapping data = ([ "a" : "A", "b" : "B", "d" : "D" ]); m_delete(data, removable_indices); will not have the same result as removing indices "a", "b" and "c" through three separate m_delete statements. Is there a way to tell pike I wish to delete a list of indices, allowing pike to maybe perform such an operation more efficient than through separate statements? Or any chance such a statement or syntax might appear in the future? (Hmmmm, I haven't checked m_delete(data, @removable_indices); ) Regards, Arjan

4 8

arrays, multisets, mappings and tuples (was: avoiding unnecessary memory (re)assigns)
by Martin Bähr 08 Sep '09

08 Sep '09

since this is moving into speculation now, i am moving this topic to the devel forum... On Tue, Sep 08, 2009 at 06:44:16PM +0200, Martin Stjernholm wrote: > > what is the reason for this difference? > Historical, I think. Once upon a time multisets were implemented as > sorted arrays. hmm? i don't see how that explains why multisets are binary trees. the explanation you give below makes much more sense. > But there would be little use to have hash tables in two forms. If you > want a hash table, use a mapping. i would like an optimized multiset. at this point the only difference between a mapping and a multiset is the fact that i can write it without values: (< "key", "key" >) vs: ([ "key":value ]), and that iteration over a multiset includes keys not just once. i always saw multisets as syntactic sugar for a special form of mappings. were it not for the future plans... > Rather, the intention is at some point to extend multisets to be able > to take values, just like mappings, and to take a custom sort > function, so that their binary tree structure is fully usable from the > pike level. Most low-level support for that is already in place. sounds great. binary trees are a useful thing to have, although i wonder how they would relate to current use of multisets. aren't they a whole different thing alltogether? and what about the necessary syntax change? we would get (< key, ...>) and (< key:value, ...>), at which point we might as well allow: ([ key,... ]) (ie: a hashtable based set implementation, aka syntax sugar for mapping(mixed:int(0..1)) and, hmm here is some wood for the fire of the discussion for tuple syntax. how about: tuple(mixed:mixed:mixed) foo = a:b:c; then we can write arrays of tuples like this: ({ a:b:c, ... }) and the key:value pairs of mappings could be treated as tuples as well, same with multisets which would be generalized to: multiset(tuple(mixed:mixed)) leading naturally to multisets with values and mappings without values. at least on the syntax level. in the implementation the problem would be that the first element of a tuple would have to be the key, and not the whole tuple. and nested tuples would not be possible with this syntax either. greetings, martin.

2 1

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 September 2009