Quoting Kenton Varda (2017-04-09 18:35:48)
>1) libcapnp and libkj together add up to some 730k of code (text
>segment) these days. Unless emscripten builds are significantly
>smaller, that's probably too big.
Hard to know without trying it, but it may well be the case that wasm
builds
On Sunday, April 9, 2017 at 7:22:26 PM UTC-7, Kenton Varda wrote:
>
> On Sun, Apr 9, 2017 at 7:04 PM, Wink Saville > wrote:
>>
>> 2) The interface between JS and WASM would probably be very slow if every
>>> accessor has to go through it. You'd probably end up wanting to translate
>>> the whol
serial/deserializer
On Sunday, April 9, 2017 at 8:30:51 PM UTC-7, Mark Miller wrote:
>
> "serdes"?
>
>
> On Sun, Apr 9, 2017 at 10:22 PM, Kenton Varda > wrote:
>
>> On Sun, Apr 9, 2017 at 7:04 PM, Wink Saville > > wrote:
>>>
>>> 2) The interface between JS and WASM would probably be very slow if
"serdes"?
On Sun, Apr 9, 2017 at 10:22 PM, Kenton Varda wrote:
> On Sun, Apr 9, 2017 at 7:04 PM, Wink Saville wrote:
>>
>> 2) The interface between JS and WASM would probably be very slow if every
>>> accessor has to go through it. You'd probably end up wanting to translate
>>> the whole capnp
On Sun, Apr 9, 2017 at 7:04 PM, Wink Saville wrote:
>
> 2) The interface between JS and WASM would probably be very slow if every
>> accessor has to go through it. You'd probably end up wanting to translate
>> the whole capnp to a JSON object upfront, which of course defeats a lot of
>> the purpos
On Sunday, April 9, 2017 at 3:36:11 PM UTC-7, Kenton Varda wrote:
>
> I don't think WASM is likely to be a good fit here, for two reasons:
>
> 1) libcapnp and libkj together add up to some 730k of code (text segment)
> these days. Unless emscripten builds are significantly smaller, that's
> pro
I understand, just mak'n sure, thanks for the feed back.
--
You received this message because you are subscribed to the Google Groups
"Cap'n Proto" group.
To unsubscribe from this group and stop receiving emails from it, send an email
to capnproto+unsubscr...@googlegroups.com.
Visit this group
Hi Stephan,
Sorry, I misread your question.
Currently there is no way to generate the headers and c++ in separate
folders. You will have to move one or the other after the fact.
-Kenton
On Sun, Apr 2, 2017 at 12:38 PM, Stephan Opfer
wrote:
> HI Kenton,
>
> this would generate the .c++ and .h
Hi,
You want to use capnp::FlatArrayMessageReader as defined in
capnp/serialize.h .
Note that it wants an array of capnp::word, not bytes. You can use
reinterpret_cast, but note that the buffer needs to be 64-bit aligned (or
32-bit aligned on 32-bit systems). This is usually easy to guarantee but
Hi Wink,
You haven't missed anything.
To do an official release, we need to make sure Cap'n Proto works correctly
in a wide variety of configurations, including:
- Linux/GCC4.7, Linux/GCC5, Linux/GCC6, Linux/Clang3.3, Linux/Clang4
- Windows/MSVC, Windows/MinGW, Windows/Cygwin
- MacOS, FreeBSD, An
I don't think WASM is likely to be a good fit here, for two reasons:
1) libcapnp and libkj together add up to some 730k of code (text segment)
these days. Unless emscripten builds are significantly smaller, that's
probably too big. (If you stick to lite mode, it's still 401k, which is
still probab
On Wed, Apr 5, 2017 at 9:01 PM, wrote:
> Anything ever happen with this?
>
(Moderator's note: This message was actually sent four days ago, before the
recent activity on this thread, but got caught in the moderation queue.)
--
You received this message because you are subscribed to the Google
Anything ever happen with this?
On Tuesday, October 4, 2016 at 10:13:51 AM UTC-5, Julián Díaz wrote:
>
> Generated TypeScript it'll be, then! It'll be pretty trivial to
> subsequently invoke the TypeScript compiler on the TypeScript generated
> code to get "free" ES5 code for everyone else.
>
>
13 matches
Mail list logo
