From: Xavier Leroy <xleroy@pauillac.inria.fr>
To: caml-list@pauillac.inria.fr
Subject: Objective Caml 1.03
Date: Tue, 29 Oct 1996 16:18:10 +0100 (MET) [thread overview]
Message-ID: <199610291518.QAA05680@pauillac.inria.fr> (raw)
Release 1.03 of Objective Caml is now available.
This is mostly a bug-fix release that corrects the main problems
reported with 1.02. See the end of this message for a detailed list
of changes.
The complete sources, binaries for Windows, and diffs from 1.00 are
available by anonymous FTP at ftp://ftp.inria.fr/lang/caml-light.
See http://pauillac.inria.fr/ocaml/ for documentation and general info
about Objective Caml
- Xavier Leroy
Objective Caml 1.03:
--------------------
* Typing:
- bug with type names escaping their scope via unification with
non-generalized type variables '_a completely fixed;
- fixed bug in occur check : it was too restrictive;
- fixed bug of coercion operators;
- check that no two types of the same name are generated in a module
(there was no check for classes);
- "#install_printer" works again;
- fixed bug in printing of subtyping errors;
- in class interfaces, construct "method m" (without type) change
the status of method m from abstract to concrete;
- in a recursive definition of class interfaces, a class can now
inherit from a previous class;
- typing of a method make use of an eventual previously given type
of this method, yielding clearer type errors.
* Compilation (ocamlc and ocamlopt):
- fixed bug in compilation of classes.
* Native-code compilation:
- optimization of functions taking tuples of arguments;
- code emitter for the Motorola 680x0 processors (retrocomputing week);
- Alpha/OSF1: generate frame descriptors, avoids crashes when e.g.
exp() or log() cause a domain error; fixed bug with
String.length "literal";
- Sparc, Mips, HPPA: removed marking of scanned stack frames
(benefits do not outweight cost).
* Standard library:
- Arg.parse now prints documentation for command-line options;
- I/O buffers (types in_channel and out_channel) now heap-allocated,
avoids crashing when closing a channel several times;
- Overflow bug in compare fixed;
- GC bug in raising Sys_error from I/O functions fixed;
- Parsing.symbol_start works even for epsilon productions.
* Foreign interface: main() in C now working, fixed bug in library
order at link time.
* Thread library: guard against calling thread functions before Thread.create.
* Unix library: fixed getsockopt, setsockopt, open_process_{in,out}.
* Perl-free, cpp-free, cholesterol-free installation procedure.
reply other threads:[~1996-10-29 19:52 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=199610291518.QAA05680@pauillac.inria.fr \
--to=xleroy@pauillac.inria.fr \
--cc=caml-list@pauillac.inria.fr \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox