Nabble - io

Re: A wiki written in Io?

2008-09-06T17:44:29Z

--- In iolanguage@..., "Danya Alexeyevsky" <me.dendik@...>
wrote:
>
> Looks like http://xeny.net/files/iowiki/ is a mirror of
> ftp://xeny.net/iowiki/ and is working. (Instructions about the wrapper
> are on the page CurrentIoWrapper on the wiki).
>
> Cheers.

IoWiki is in daily use, but it will only run on an old version of Io
-- version 20060502, to be specific. You can see it in action at
http://xeny.net/IoWiki. You're welcome to play with it, but I've had
spam problems so you'll have to email me for a password for editing:
jason@....

Danya has written a wrapper which will let it run on a more recent
version of Io. See http://xeny.net/InstallingIoWiki. Danya is my
hero. I don't know whether it runs on the very latest version of Io,
but even if it doesn't, it could be updated easily enough.

A more interesting project would be to update the IoWiki code itself
(rather than the wrapper) to run on the latest version of Io. I'd
love it if someone would do that. You could get most (or all?) of the
hints you'd need from Danya's wrapper. The latest IoWiki code is at
http://xeny.net/files/iowiki/iowiki-0.1.05.tgz. I'm not going to be
updating it any time soon because Io has changed a fair bit since last
time I programmed in it ... not the language philosophy, of course,
but enough of the methods that it would take me a while to get back up
to speed.

BTW, my code is a mess, but only in the sense of being a bit verbose
and inelegant. It should be pretty easy to read.

Jason

can io work on multiple nodes and coordinate through actors?

2008-09-06T01:30:55Z

curious

Re: Io on .NET?

2008-08-31T06:28:37Z

Here is some report about IoNET. After I finished all the Io skeleton
(now only left adding IoVM Classes, like IoList and finished all the
slots handlers to meets requirements to be compatible with original
Io), I decided to make the bridge into the whole CLR .NET with Reflection.

Here is sample session of proposed way to interact with all CLR classes.
http://community.livejournal.com/ru_iolanguage/3438.html

An here is the screenshot of calling native .NET methods:
http://synrc.com/io/CreateNETObjects.jpg

Regards,
Maxim

Re: Question about memory allocation

2008-08-29T23:20:09Z

Steve Dekorte <steve@...> writes:

> On 2008-08-29, at 8:15 AM, Friedrich wrote:
>> So what about something along this lines:
>
>
> Sounds good - I'd accept a patch to deal with this. Io should use
> io_calloc() and io_malloc() for all allocs (if not, we can fix those),
> so we should be able to add the hook there.
Well the call sequence is:
io_calloc
io_real_calloc
io_real_malloc
and mentioned function

so this is the lowest level.

So that seems to be the proper place for erro handling. I guess in the
end one has to deal with two places with failing memory requestion in
the
allocation routines and in the reallocation routines. Both could have
a Hander while the allocation/reallocation fails.

This could be a simple wrapper. And or one has to check what other
parts of the existing Error handling could be used.

One suggestion would be to have two tiny layers above calloc (it seems
malloc is not used at all)

void * standardAllocateFunction(size_t size, char* file, int line) {
void *result = calloc(....);
if (NULL == self) {
fprintf(stderr, "Request for %d bytes of memory failed at %s,
line %d\n", size, file, line);
abort(); /* Or something customizable */
}

And reroute all stuff to this maybe Object would be the proper place
for it. One than could replace it with something loke
setAllocationFunction = gc_malloc or whatever.

One strategy for handling errors might be calling into the GC and see
whether it can squeeze out some bytes ;-)

It seems there is some other code in Common.c to handle memory
stuff. So I assume that this has not fully settled down. But one
definitly will get a crash if the memory allocation fails and one does
not get an intermediate backtrace or the like.

I've not checked all but IIRC ther is some function for printing a
backtrace maybe calling it while the memory allocation fails is a good
idea

Regards
Friedrich

Re: Question about memory allocation

2008-08-29T15:45:40Z

On 2008-08-29, at 8:15 AM, Friedrich wrote:
> So what about something along this lines:

Sounds good - I'd accept a patch to deal with this. Io should use
io_calloc() and io_malloc() for all allocs (if not, we can fix those),
so we should be able to add the hook there.

- Steve

RE: Re: Io on .NET?

2008-08-29T08:56:50Z

Nice, keep on truckin' !

-edwin

To: iolanguage@...: maxim@...: Thu, 28 Aug 2008 17:29:19 +0000Subject: [Io] Re: Io on .NET?

I will run all the bechmarks and unit tests when finish all the coreprimitives. BTW, please change at http://iolanguage.com/blog/blog.cgi"A project at Syncrc to port Io to .Net."to"A project at Synrc to port Io to .NET.";)--- In iolanguage@..., Steve Dekorte <steve@...> wrote:>> > On 2008-08-27, at 1:56 PM, Maxim Sokhatsky wrote:> > here is example of working IoNET session> >> > http://synrc.com/io/io-session.txt> >> > Now I'm just adding slots handlers to objects to be identical to> > original IoVM. No coroutines up for now and error handling and other> > core objects such as Date, Directory, File.... Example session is just> > a proof the concept. I started only 23-th of August.> > Nice work. :-) I've only had a chance to scan the implementation but > it looks pretty good. It's nice that .Net (appears) to handle all the > garbage collection for you. Have you done any performance tests?> > - Steve>

_________________________________________________________________
Be the filmmaker you always wanted to be—learn how to burn a DVD with Windows®.
http://clk.atdmt.com/MRT/go/108588797/direct/01/

Question about memory allocation

2008-08-29T08:15:53Z

I browsed a bit through the sources and found this:
MemoryBlock *MemoryBlock_newWithSize_file_line_(size_t size, char *file, int line)
{
MemoryBlock *self = calloc(1, sizeof(MemoryBlock) + size);
self->size = size;
self->allocNum = allocs;
self->file = file;
self->line = line;
MemoryBlock_insertAfter_(self, baseblock());

allocs ++;
allocatedBytes += size;
if (allocatedBytes > maxAllocatedBytes) maxAllocatedBytes = allocatedBytes;
return self;
}

It seems that calloc never can fail. Which seems to be dangerous
assumption dangerous assumption. I'm quit aware that a memory reques
hardly failed these day but what if?
So what about something along this lines:
MemoryBlock *MemoryBlock_newWithSize_file_line_(size_t size, char *file, int line)
{
MemoryBlock *self = calloc(1, sizeof(MemoryBlock) + size);
if (NULL == self) {
fprintf(stderr, "Request for %d bytes of memory failed at %s,
line %d\n", size, file, line);
abort(); /* Or something customizable */
}
self->size = size;
self->allocNum = allocs;
self->file = file;
self->line = line;
MemoryBlock_insertAfter_(self, baseblock());

allocs ++;
allocatedBytes += size;
if (allocatedBytes > maxAllocatedBytes) maxAllocatedBytes = allocatedBytes;
return self;
}

Regards
Friedrich

Re: Io on .NET?

2008-08-28T10:29:19Z

I will run all the bechmarks and unit tests when finish all the core
primitives. BTW, please change at http://iolanguage.com/blog/blog.cgi

"A project at Syncrc to port Io to .Net."

to

"A project at Synrc to port Io to .NET."

;)

--- In iolanguage@..., Steve Dekorte <steve@...> wrote:

>
>
> On 2008-08-27, at 1:56 PM, Maxim Sokhatsky wrote:
> > here is example of working IoNET session
> >
> > http://synrc.com/io/io-session.txt
> >
> > Now I'm just adding slots handlers to objects to be identical to
> > original IoVM. No coroutines up for now and error handling and other
> > core objects such as Date, Directory, File.... Example session is just
> > a proof the concept. I started only 23-th of August.
>
> Nice work. :-) I've only had a chance to scan the implementation but
> it looks pretty good. It's nice that .Net (appears) to handle all the
> garbage collection for you. Have you done any performance tests?
>
> - Steve
>

Re: embedding Io

2008-08-28T01:07:15Z

Here's a small helper I hacked together. As many may know there is
this pkg-config stuff. I guess it'll be quite handy to have something
for io also. Here's the code for that:

prefix=/home/frido/src/io/_build
exec_prefix=${prefix}/binaries
libdir=${exec_prefix}/lib
includedir=${prefix}/headers/

Name: Io
Description: Flags for embedding Io into C programs
Author: Friedrich Dominicus (frido at q-software-solutions dot de)
Version: 0.0.1
Libs: -L${libdir} -liovmall -ldl -lm
Cflags: -std=c99 -g -O2 -I${includedir}

Well you have to adopt the path for your needs. It seems that during
install the headers for Io, don't get installed in some well know
places, I guess that's due to the fact the AddOn stuff works, however
it might be not the worst to have some place for that stuff maybe
somehting like
/usr/local/lib/io or the like?

The stuff was just tried here on my machine. I guess not much can go
wrong with it but who knows?

Happy Ioing
Friedrich

Re: embedding Io

2008-08-28T00:47:36Z

Steve Dekorte <steve@...> writes:
>
> This looks like a possible compiler issue - which compiler and version
> are you using and on which platform?
It dosn't have occurred to me that this may be a compiler problem. I
tried gcc-4.1 and got this:
gcc-4.1 -g -std=c99 -I/home/frido/src/io/_build/headers -L/home/frido/src/io/_build/lib test1.c -liovmall -ldl -lm
In file included from /home/frido/src/io/_build/headers/IoSeq.h:109,
from /home/frido/src/io/_build/headers/IoState.h:21,
from test1.c:1:
/home/frido/src/io/_build/headers/IoSeq_immutable.h:103:80: warning: no newline at end of file

Ah well, so yes it seems to be a problem with gcc-4.3.

Regards
Friedrich

--
Q-Software Solutions GmbH; Sitz: Bruchsal; Registergericht: Mannheim
Registriernummer: HRB232138; Geschaeftsfuehrer: Friedrich Dominicus

Re: embedding Io

2008-08-28T00:29:37Z

Steve Dekorte <steve@...> writes:

> On 2008-08-25, at 11:10 PM, Friedrich wrote:
>
>> I've tried:
>> gcc -g -std=c99 -I/home/frido/src/io/_build/headers -L/home/frido/
>> src/io/_build/lib test1.c -liovmall -ldl -lm
>> /home/frido/src/io/_build/lib/libiovmall.a(List.o): In function
>> `List_size':
>> /home/frido/src/io/libs/basekit/source/List_inline.h:75: multiple
>> definition of `List_size'
>> /tmp/ccYBDJRy.o:/home/frido/src/io/_build/headers/List_inline.h:75:
>> first defined here
>> /home/frido/src/io/_build/lib/libiovmall.a(List.o): In function
>> `List_rawAt_':
>>
>> then a few flags like -DIO_IN_C_FILE and the like but have not got the
>> "right" combination.
>>
>> So what are the right flags?
>
> This looks like a possible compiler issue - which compiler and version
> are you using and on which platform?

gcc 4.3.1 (according to gcc -v)

>gcc -v
Using built-in specs.
Target: x86_64-linux-gnu
Configured with: ../src/configure -v --with-pkgversion='Debian 4.3.1-6' --with-bugurl=file:///usr/share/doc/gcc-4.3/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --enable-nls --with-gxx-include-dir=/usr/include/c++/4.3 --program-suffix=-4.3 --enable-clocale=gnu --enable-libstdcxx-debug --enable-objc-gc --enable-mpfr --enable-cld --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu
Thread model: posix
gcc version 4.3.1 (Debian 4.3.1-6)

As you see it's a 64-bit System.

Regards
Friedrich

Re: Re: Io on .NET?

2008-08-27T15:55:50Z

On 2008-08-27, at 1:56 PM, Maxim Sokhatsky wrote:
> here is example of working IoNET session
>
> http://synrc.com/io/io-session.txt
>
> Now I'm just adding slots handlers to objects to be identical to
> original IoVM. No coroutines up for now and error handling and other
> core objects such as Date, Directory, File.... Example session is just
> a proof the concept. I started only 23-th of August.

Nice work. :-) I've only had a chance to scan the implementation but
it looks pretty good. It's nice that .Net (appears) to handle all the
garbage collection for you. Have you done any performance tests?

- Steve

Re: Io on .NET?

2008-08-27T13:56:56Z

here is example of working IoNET session

http://synrc.com/io/io-session.txt

Now I'm just adding slots handlers to objects to be identical to
original IoVM. No coroutines up for now and error handling and other
core objects such as Date, Directory, File.... Example session is just
a proof the concept. I started only 23-th of August.

--- In iolanguage@..., Steve Dekorte <steve@...> wrote:

>
>
> On 2008-08-25, at 7:28 PM, Maxim Sokhatsky wrote:
> > We started Io Language for .NET CLR implementation.
> > You may check it at our repository:
> >
> > https://synrc.com/repository/io
>
> Hi Maxim,
>
> Neat! How far along is it?
>
> Cheers,
> - Steve
>

Re: Re: Io on .NET?

2008-08-27T13:46:31Z

On 2008-08-25, at 7:28 PM, Maxim Sokhatsky wrote:
> We started Io Language for .NET CLR implementation.
> You may check it at our repository:
>
> https://synrc.com/repository/io

Hi Maxim,

Neat! How far along is it?

Cheers,
- Steve

Re: embedding Io

2008-08-27T13:43:22Z

On 2008-08-25, at 11:10 PM, Friedrich wrote:

> I've tried:
> gcc -g -std=c99 -I/home/frido/src/io/_build/headers -L/home/frido/
> src/io/_build/lib test1.c -liovmall -ldl -lm
> /home/frido/src/io/_build/lib/libiovmall.a(List.o): In function
> `List_size':
> /home/frido/src/io/libs/basekit/source/List_inline.h:75: multiple
> definition of `List_size'
> /tmp/ccYBDJRy.o:/home/frido/src/io/_build/headers/List_inline.h:75:
> first defined here
> /home/frido/src/io/_build/lib/libiovmall.a(List.o): In function
> `List_rawAt_':
>
> then a few flags like -DIO_IN_C_FILE and the like but have not got the
> "right" combination.
>
> So what are the right flags?

This looks like a possible compiler issue - which compiler and version
are you using and on which platform?

Re: Io on .NET?

2008-08-26T03:44:16Z

Just compiled DLR. Total 1.2MB. IoNET now about 50K and will not be
more 100K. Next, the DLR lib is only needed when we start to implement
JIT for IoNET. Anyway this functionlity can be done with from 10 up to
20 lines of code. So I decided to forget about DLR framework.

DLR is not patch for CLR it ss only the helper library to produce
dynamic delegates wich is used in dynamic languages. Also DLR contains
useless for Io AST tree parser and other things useful in Iron* languages.

--- In iolanguage@..., "Daniele Alessandri"
<suppakilla@...> wrote:

>
> On Tue, Aug 26, 2008 at 04:28, Maxim Sokhatsky <maxim@...> wrote:
>
> > We started Io Language for .NET CLR implementation.
> > You may check it at our repository:
> >
> > https://synrc.com/repository/io
>
> Cool.
> Out of curiosity, why don't you build your implementation of Io for
> the .NET runtime on top of the Dynamic Language Runtime (DLR)?
>
> --
> Daniele Alessandri
> http://www.clorophilla.net/blog/
>

Re: Re: Io on .NET?

2008-08-25T23:35:14Z

On Tue, Aug 26, 2008 at 04:28, Maxim Sokhatsky <maxim@...> wrote:

> We started Io Language for .NET CLR implementation.
> You may check it at our repository:
>
> https://synrc.com/repository/io

Cool.
Out of curiosity, why don't you build your implementation of Io for
the .NET runtime on top of the Dynamic Language Runtime (DLR)?

--
Daniele Alessandri
http://www.clorophilla.net/blog/

embedding Io

2008-08-25T23:10:26Z

The documentation shows an example, but does not tell how to compile
that program. What flags are needed and against which library does one
has to link.

I've tried:
gcc -g -std=c99 -I/home/frido/src/io/_build/headers -L/home/frido/src/io/_build/lib test1.c -liovmall -ldl -lm
/home/frido/src/io/_build/lib/libiovmall.a(List.o): In function `List_size':
/home/frido/src/io/libs/basekit/source/List_inline.h:75: multiple definition of `List_size'
/tmp/ccYBDJRy.o:/home/frido/src/io/_build/headers/List_inline.h:75: first defined here
/home/frido/src/io/_build/lib/libiovmall.a(List.o): In function
`List_rawAt_':

then a few flags like -DIO_IN_C_FILE and the like but have not got the
"right" combination.

So what are the right flags?

Regards
Friedrich

Re: Re: Io on .NET?

2008-08-25T21:27:22Z

btw, may be move repo to github?

2008/8/26 Maxim Sokhatsky <maxim@...>:

> Hello, all!
>
> We started Io Language for .NET CLR implementation.
> You may check it at our repository:
>
> https://synrc.com/repository/io
>
> Regards,
>
> Maxim Sokhatsky - maxim@...
> Synrc Research Center - http://synrc.com
>
>

--
берегите мозг
Сёмка

Re: Re: Io on .NET?

2008-08-25T21:24:00Z

That's perfect, may be i'm join.

2008/8/26 Maxim Sokhatsky <maxim@...>:

--
берегите мозг
Сёмка

Re: Io on .NET?

2008-08-25T19:28:33Z

Hello, all!

We started Io Language for .NET CLR implementation.
You may check it at our repository:

https://synrc.com/repository/io

Regards,

Maxim Sokhatsky - maxim@...
Synrc Research Center - http://synrc.com

Re: Some issues

2008-08-18T12:08:01Z

2008/8/10 Steve Dekorte <steve@...>:
>> 3. I use 'nil' alot. That is why I need these:
>
> I get the feeling you're after an Objective-C style handling of nil
> where it's a black hole for messages sent to it. To get this effect
> for messages it doesn't implement, you could do:
>
> nil forward := method(nil)

I'd argue that's probably not what you want to do all the time, and
could lead to bugs coming purely from forgetting to handle a nil case
in some dark corner somewhere. These are probably all the Java in my
head talking, but a colleague of mine took the time to jot down some
thoughts on this you might find useful:

http://www.markhneedham.com/blog/2008/08/16/null-handling-strategies/

Cheers!

- Carlos

Re: Some issues

2008-08-09T17:15:29Z

> 3. I use 'nil' alot. That is why I need these:

I get the feeling you're after an Objective-C style handling of nil
where it's a black hole for messages sent to it. To get this effect
for messages it doesn't implement, you could do:

nil forward := method(nil)

Re: Some issues

2008-08-09T05:38:24Z

With respect to your methods on nil... Anything that returns nil itself, just prepend in your code a ? So for instance:

foo := File standardInput readLine split
foo append(list(1, 2))
newfoo := foo ?flatten

Just as an example. What ? Does is it if it cannot find its argument (in this case, flatten) in the lookup tree of the receiver (in this case, foo), then it will simply return nil.

Your nil isKindOf is VERY bad. It should not be implemented period. What happens if i do nil isKindOf(nil) or worse yet, someValueThatCanBeNil isKindOf(nil) (or someValueThatCanBeNil isNil (since isNil uses isKindOf under the hood))... What happens?

As well, nil atPut is poorly thought out. Consider: someMap atIfAbsentPut("foo", 42) ... Since atIfAbsentPut uses atPut (or at least, should) after the error checking, what this is going to do is when you do someMap at("foo") you will get a Map back when I expected the value, since "foo" did not exist in someMap prior to my calling. It's a bad idea.

As for the rest, perhaps. But some of those seem more like wanting the language to save you from error checking.

jer

On 8/9/08 3:44 AM, "Bart Botma" <bart_botma@...> wrote:

Hi,

1. The issues button on the Io website does not work anymore.

Exception: compile error: expected a message but instead found a open
group character

Coroutine_0x15ce830 stack trace
------------------------------
clone [lexer] line 1
DB open common.io line 55

2. With Io-2008-03-30 I noticed yesterday at work on my dual-core dell
laptop that:
list() sortBy(block(x,y,(x > y)))
gave me a segmentation fault, but today on my dual PIII with the same
release that does not happen. Strange. I'll look further into this
next week behind my laptop.

3. I use 'nil' alot. That is why I need these:

nil isEmpty := method(true)
nil size := method(0)
nil flatten := method(nil)
nil foreach := method(nil)
nil isKindOf := method(x,false)
nil append := method(x,List clone append(x))
nil contains := method(x,false)
nil sortBy := method(x,nil)
nil keys := method(nil)
nil values := method(nil)
nil at := method(s,nil)
nil atPut := method(s,x,Map clone atPut(s,x))

Could such definitions become part of a 'nil' prelude one day? Would
this help the average end-user?

Bart

------------------------------------

Yahoo! Groups Links

Some issues

2008-08-09T00:44:46Z

Hi,

1. The issues button on the Io website does not work anymore.

Exception: compile error: expected a message but instead found a open
group character

Coroutine_0x15ce830 stack trace
------------------------------
clone [lexer] line 1
DB open common.io line 55

2. With Io-2008-03-30 I noticed yesterday at work on my dual-core dell
laptop that:
list() sortBy(block(x,y,(x > y)))
gave me a segmentation fault, but today on my dual PIII with the same
release that does not happen. Strange. I'll look further into this
next week behind my laptop.

3. I use 'nil' alot. That is why I need these:

nil isEmpty := method(true)
nil size := method(0)
nil flatten := method(nil)
nil foreach := method(nil)
nil isKindOf := method(x,false)
nil append := method(x,List clone append(x))
nil contains := method(x,false)
nil sortBy := method(x,nil)
nil keys := method(nil)
nil values := method(nil)
nil at := method(s,nil)
nil atPut := method(s,x,Map clone atPut(s,x))

Could such definitions become part of a 'nil' prelude one day? Would
this help the average end-user?

Bart

Re: URL / Protocol 'https' unsupported ?

2008-08-08T11:56:15Z

Sorry, obsessive compulsive disorder at work here: I should have said
"OpenSSL binding".

On Fri, Aug 8, 2008 at 11:54 AM, Phoenix Sol <phoenix@...>wrote:

> There's an SSL addon available ('SecureSocket' I think), and I have a need
> for it as well, but have yet to succeed at getting it to work due to time
> constraints.
>
> There was some discussion about it a month or two ago; please check the
> archive ;-)
>
> At kind of a glance, it seems like it just needs to be integrated into the
> Socket (or maybe just URL) addon (but like I said, I don't really know if
> it's even working right now).
>
> I'm hoping to have some time to play with it soon!!
>
>
> If anybody gets it to work (or has it working), please post some info =)
>
>
>
>
> On Fri, Aug 8, 2008 at 11:34 AM, Rich Collins <richcollins@...>wrote:
>
>> I think it is still a TODO. Perhaps someone else can verify that there
>> is no SSL support.
>>
>> On Aug 8, 2008, at 11:23 AM, bpsi wrote:
>>
>> Have been lurking for awhile, and have started firing up IO in a real
>> project at work -- I need to pull XML objects out of a web-based
>> ticketing system that is served by HTTPS
>>
>> I'm getting a "Protocol 'https' unsupported" by the URL class -- is
>> this a known behavior? Any straightforward way to enable the support?
>>
>> Or is it on a perpetual 'to do' list somewhere
>>
>> Thanks for the time and consideration!
>>
>> -David
>>
>>
>>
>
>
>
>
> --
> Phoenix Sol
> 541-646-8612
> 130 'A' Street,
> Suite 6
> Ashland, Oregon
> 97520
>

--
Phoenix Sol
541-646-8612
130 'A' Street,
Suite 6
Ashland, Oregon
97520

Re: URL / Protocol 'https' unsupported ?

2008-08-08T11:54:02Z

There's an SSL addon available ('SecureSocket' I think), and I have a need
for it as well, but have yet to succeed at getting it to work due to time
constraints.

There was some discussion about it a month or two ago; please check the
archive ;-)

At kind of a glance, it seems like it just needs to be integrated into the
Socket (or maybe just URL) addon (but like I said, I don't really know if
it's even working right now).

I'm hoping to have some time to play with it soon!!

If anybody gets it to work (or has it working), please post some info =)

On Fri, Aug 8, 2008 at 11:34 AM, Rich Collins <richcollins@...> wrote:

> I think it is still a TODO. Perhaps someone else can verify that there
> is no SSL support.
>
> On Aug 8, 2008, at 11:23 AM, bpsi wrote:
>
> Have been lurking for awhile, and have started firing up IO in a real
> project at work -- I need to pull XML objects out of a web-based
> ticketing system that is served by HTTPS
>
> I'm getting a "Protocol 'https' unsupported" by the URL class -- is
> this a known behavior? Any straightforward way to enable the support?
>
> Or is it on a perpetual 'to do' list somewhere
>
> Thanks for the time and consideration!
>
> -David
>
>
>

--
Phoenix Sol
541-646-8612
130 'A' Street,
Suite 6
Ashland, Oregon
97520

Re: URL / Protocol 'https' unsupported ?

2008-08-08T11:34:18Z

I think it is still a TODO. Perhaps someone else can verify that
there is no SSL support.

On Aug 8, 2008, at 11:23 AM, bpsi wrote:

> Have been lurking for awhile, and have started firing up IO in a real
> project at work -- I need to pull XML objects out of a web-based
> ticketing system that is served by HTTPS
>
> I'm getting a "Protocol 'https' unsupported" by the URL class -- is
> this a known behavior? Any straightforward way to enable the support?
>
> Or is it on a perpetual 'to do' list somewhere
>
> Thanks for the time and consideration!
>
> -David
>
>
>

URL / Protocol 'https' unsupported ?

2008-08-08T11:23:13Z

Have been lurking for awhile, and have started firing up IO in a real
project at work -- I need to pull XML objects out of a web-based
ticketing system that is served by HTTPS

I'm getting a "Protocol 'https' unsupported" by the URL class -- is
this a known behavior? Any straightforward way to enable the support?

Or is it on a perpetual 'to do' list somewhere

Thanks for the time and consideration!

-David

Re: Protobuffers

2008-08-05T17:00:57Z

I have placed a copy up at http://waspvm.googlepages.com; look for
libgpb-0.0.tar.gz. It's pretty minimal -- it can encode / decode
messages, seek new ones, and is fairly efficient about it. It is far,
far less complicated than Google's C++ version. :)

As for MPI.. Maybe another day. :)

--Scott.

On Tue, Aug 5, 2008 at 5:20 PM, Steve Dekorte <steve@...> wrote:

>
> On 2008-08-05, at 4:57 AM, Scott Dunlop wrote:
>> I have done some preliminary work on an ANSI C implementation of GPB;
>> it encodes and decodes messages just fine. If you like, I'd be happy
>> to toss it in the public domain and let you have at it. (I was
>> originally writing it for a plugin to Wasp Lisp, but I got distracted
>> by work again..)
>
> Sure - I'll take a look. Thanks.
>
> Btw, if anyone is interested in working on an MPI binding, I'd be glad
> to help.
>
> http://en.wikipedia.org/wiki/Message_Passing_Interface
>
> - Steve
>

another MPI link

2008-08-05T15:22:11Z

http://www.open-mpi.org/

Re: Protobuffers

2008-08-05T15:20:56Z

On 2008-08-05, at 4:57 AM, Scott Dunlop wrote:
> I have done some preliminary work on an ANSI C implementation of GPB;
> it encodes and decodes messages just fine. If you like, I'd be happy
> to toss it in the public domain and let you have at it. (I was
> originally writing it for a plugin to Wasp Lisp, but I got distracted
> by work again..)

Sure - I'll take a look. Thanks.

Btw, if anyone is interested in working on an MPI binding, I'd be glad
to help.

http://en.wikipedia.org/wiki/Message_Passing_Interface

- Steve

Re: Protobuffers

2008-08-05T04:57:05Z

I have done some preliminary work on an ANSI C implementation of GPB;
it encodes and decodes messages just fine. If you like, I'd be happy
to toss it in the public domain and let you have at it. (I was
originally writing it for a plugin to Wasp Lisp, but I got distracted
by work again..)

--Scott.

On Thu, Jul 10, 2008 at 2:23 AM, Friedrich Dominicus
<frido@...> wrote:

> Steve Dekorte <steve@...> writes:
>
>> Would anyone like to implement a binding for Io?
>> http://code.google.com/p/protobuf/
> Well I've looked through the code for three things
> 1) plain C
> 2) Ruby
> 3) Io on my back.
>
> I read the source and guess I have an idea on how it works. But I'm
> ashamed to tell. I can't stand C++, it's ridicolous I know, however it's
> such
> an ugly language, it hurts reading it and I guess it's not much better
> writing it. But for this code I have to tell it can be read, and it would be
> stupdid not to use what they all have implemente, that would mean for
> the code generator one has to write some other C++ code. And this is
> something I happily leave to the C++ lovers. They must exist
> ... somewhere.
>
> Regards
> Friedrich
>
>

Re: maybe I hit a trouble spot

2008-07-31T02:36:39Z

On 2008-07-30, at 7:35 AM, Friedrich wrote:

> Steve Dekorte <steve@...> writes:
> It seems the best solution for this sort of task would be to hand the
> Regex a file pointer and have call back as needed instead of hand
> feeding it each line.
>
>> On 2008-07-29, at 10:47 PM, Friedrich wrote:
>>> You mean slurp in the whole file and then give it to
>>> the Regex?
>> No, just handing Regex a file stream to read.
> How would that look? I can not see any such thing in the Regex AddOn

There isn't anything in the current Regex object to do it and I don't
know if PCRE can accept a file pointer.

Re: maybe I hit a trouble spot

2008-07-30T07:35:21Z

Steve Dekorte <steve@...> writes:

> On 2008-07-29, at 10:47 PM, Friedrich wrote:
>> You mean slurp in the whole file and then give it to
>> the Regex?
> No, just handing Regex a file stream to read.
How would that look? I can not see any such thing in the Regex AddOn

Regards
Friedrich