perl6/nqp

bump VERSION to 2012.08.1

Add missing mark in KnowHOWREPR that could lead to segfaults and othe… …

…r corruption.

stop building the p5 regex bits …

rakudo does not use them yet, and that step breaks the build on non-windows
we will figure out a proper fix after the rakudo release.

Revert "unbreak the build on non-windows" …

This reverts commit c7bed44.
It was wrong because building p5regex with stage0
involves the wrong serialization context.

unbreak the build on non-windows

bump VERSION to 2012.08

Tests for \w, \W, \d and \D.

Various meta-chars tests.

Get various backslash sequences in place.

Handle funkiness in Perl 5 character class parsing, and add some test… …

…s to cover it.

Run some basic tests for Perl 5 character classes.

First cut of Perl 5 style character classes.

Remove a few pieces we won't need in P5regex.

Add some quantifier tests.

Get p5quantifier largely in place and working.

Look in the correct places for tests.

Start getting various of the p5metachar bits in place, and eliminate … …

…a few bits that will be unrequired.

Add test harness for p5regex support and copy over rx_basic, which is… …

… simple enough to be valid p5 and p6. Passes.

Copy P6Regex to get a P5Regex. Start making a few of the basic change… …

…s needed, looking to STD_P5 for inspiration.

Update NQP release_guide, create "make release" target in Makefile.

makefile whitespace cleanup …

tabs missing

Add :nohandler option to PAST::Op loop nodes.

Enable parsing of :nohandler in QAST::Op while/etc loops.

Add some missing op variants.

Register type inference fix.

Fix segfault bug when write-barriering CStruct REPRd objects. …

child_objs is now allocated with mem_sys_allocate_zeroed, and a child object
is ignored if child_objs[i] is NULL. This fixes a bug where write-barriering
would trip over a bogus pointer when write-barriering a struct returned from C
and a field was unread before passing it to another C function.

Re-enable re-use of strings on the string heap; it makes a fairly not… …

…able difference to the output size of CORE.setting compilation, and a small but certainly worthwhile difference to CORE.setting compilation time. Causes no issues here; hopefully not for anyone else now either.

Add a way for ops evaluating to native types to attach that informati… …

…on to nodes.

Expose nativecall write-barriering through an nqp::op.

Provide a way to let a HLL specify that returned values should always… …

… be boxed. Outsourcing this to QAST has various benefits, including keeping this out of the QAST tree itself. Then inlining doesn't have to worry over this detail.

Switch handling of attribute bind values to the new Want-aware coerci… …

…on scheme.

Provide a better, more uniform way to handle Want and coercion (nothi… …

…ng is switched to using it yet, though).

Don't lose namedness of arguments when inlining.

Fix over-sharing of children.

Add a QAST::InlinePlaceholder to represent a place in the QAST where … …

…an argument needs to be filled in when inlining. Implement a first cut of producing a tree with the placeholders substituted in.