The problem with kj::range() is that if the inputs are not recognized as the same type, the typechecker can get annoyed.

omit needless memset

remove unneeded kj-async dependency in Makefile.am and capnp-json.pc.in

fix various bugs in canonicalization

fix typo in canonicalization docs

fix bug where PointerBuilder::getPointerType() can corrupt PointerBuilder::segment

update mention of non-existent 'stepBits' field

Make sure to clear tag word when upgrading a struct list.

Edit error message for the case where copyPointer() hits a FAR pointer.

The existing error message suggests this case can only arise if there is a
a bug in the library. However, malformed input can trigger the error too.
In particular, the error gets thrown when a non-double far pointer resolves
to another far pointer.

fix bug in parsing of non-ascii constants

Fix MSVC build errors

Closes #408. Perhaps the same bug as in commit cff9ba05?

Add HTTP client and server implementation.

The old semantics were intended to allow more implementation freedom, but I find in my actual code it's very convenient to rely on the actual implementation details. Other implementations will just have to deal with it.