From: Per Bothner <per@bothner.com>
To: list+org.sourceware.kawa@io7m.com
Cc: kawa@sourceware.org
Subject: Re: Are types checked statically?
Date: Thu, 16 Mar 2017 22:57:00 -0000 [thread overview]
Message-ID: <8847c956-30b3-8cf5-f911-8502be64e6b6@bothner.com> (raw)
In-Reply-To: <20170316173830.59958bf8@copperhead.int.arc7.info>
On 03/16/2017 10:38 AM, list+org.sourceware.kawa@io7m.com wrote:
> On 2017-03-16T10:23:24 -0700
> Per Bothner <per@bothner.com> wrote:
>> Kawa currently doesn't have "function types" - the type of all
>> functions is just 'procedure' without any specification of
>> argument or result types. However, the following is on the wish-list:
>> https://www.gnu.org/software/kawa/Ideas-and-tasks.html#Optimized-function-types-and-values-using-MethodHandles
>
> I'm not sure I understand this fully. I'm familiar with JVM internals,
> including method handles, but the above is more an issue of efficiency
> than of correctly checking the types of functions, isn't it? Unless I'm
> mistaken, Kawa lambda expressions have the proper types and the types
> are checked statically, right? They just may not have an efficient
> underlying representation on the JVM.
It's not just performance. A *specific* lambda expression has proper types,
and if you call a *known* procedure it is type-checked. However, there is
no way to say "the value of this variable is an unknown function that takes
(T1 .. Tn) arguments and returns a Tr result". That is a major part of what
the linked proposal does - in addition to enabling better code generation.
It is possible to add procedure types, and only use them for type-checking,
without optimizing the calling convention.
--
--Per Bothner
per@bothner.com http://per.bothner.com/
prev parent reply other threads:[~2017-03-16 22:57 UTC|newest]
Thread overview: 4+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-16 17:02 list+org.sourceware.kawa
2017-03-16 17:23 ` Per Bothner
2017-03-16 17:39 ` list+org.sourceware.kawa
2017-03-16 22:57 ` Per Bothner [this message]
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=8847c956-30b3-8cf5-f911-8502be64e6b6@bothner.com \
--to=per@bothner.com \
--cc=kawa@sourceware.org \
--cc=list+org.sourceware.kawa@io7m.com \
/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;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).