bc-compile is more conforming now (we validate definition and name, also our
closure compilation is a bit better, still broken though).
Also improve some error messages and add periods at the end.
This keyword argument was added in SBCL in 2010 and is used to pass
ABORT to CL:CLOSE.
The absence of this would obviously cause errors when code expects
this to be available. For example, this happens in usocket's
SOCKET-CONNECT and would cause several errors in their test suite
with obscure error messages like
Wrong number of arguments passed to function #<compiled-function 0000000003562e80>.
This is untested on Windows.
To avoid confusion we create a separate numeric branch from 3.1.7 (ASDF
3.2.0 release is on its way).
This change backports fix for static-library builds. Fixes#177.
NETDB_INTERNAL and NETDB_SUCCESS are not defined by POSIX, so some
libc implementations (such as musl) do not have them. When compiling
sockets contrib, check if these are defined and, if not, define them to
be zero.
This version has fixed make-image and uses run-program.
ECL backend updates possible thanks to help of Francois-Rene Rideau.
Signed-off-by: Daniel Kochmański <daniel@turtleware.eu>
Our ffi isn't uffi 2.0 compatible and most systems wants this, when
they say (require 'uffi). Also if system has asdf dependency on
quicklisp) we have package name error.
We'll depend with UFFI 2.0 compability on cffi-uffi-compat, what is
the saniest approach. Some ECL specific patches will land there soon
to improve performance - no internal FFI api will change. It's just
nick removal.
Signed-off-by: Daniel Kochmański <daniel@turtleware.eu>
Both byte and C compilers didn't bind lexically *readtable*, what lead
to pollution of other compiled packages. Fixes#95 and #59.
Signed-off-by: Daniel Kochmański <daniel@turtleware.eu>
The backslash in '\0' got lost on the way to the generated C file
(build/ext/sockets.c). There may be more of these issues elsewhere
in the code base.
This resolves CIDs 66405 and 66413 (Buffer not null terminated).
