boost FSM and another idea
hey boosters i was searching for finite state machine ideas and i came across some old posts from Bryan Ross and Andreas Huber regarding a boost fsm. i love boost, even tho i havent really used it much. i intend to get to grips w/ boost::graph soon. i dont kno if there is now a boost::fsm. the above messages were in 2002! anyway, i have been using an object oriented fsm pattern for years now which uses virtual function dispatching in place of the old function pointer table thingy. i find it very robust & easy to understand. i have made a few preliminary ventures into boosting it into a generic template, and i wonder if theres still any interest in this idea. i have a tendency to see everything in computer science as either a graph or a proposition, so i find fsm's almost universally applicable... heheh another offspring of my fevered brain is a little lisp (or maybe its a scheme..?), & i wondered about adding something like this to boost. it is already amenable to templated type extensions and *i* like it. i call it terp, so maybe it could be called boost::terp. any comments? cheers. jp the a/p
jp wrote:
hey boosters
i was searching for finite state machine ideas and i came across some old posts from Bryan Ross and Andreas Huber regarding a boost fsm.
Boost fsm is still up and running in development. Look in the sand-box CSV for the code. Not sure how updated the code is, but Andreas is certainly still working on it. I like the look of FSM very much, but had troubles compiling it with bcc32 5.6.4 last time I tried so have stuck with out current method of implementing statemachines. HTH Russell
"Russell Hind" <rhind@mac.com> wrote
Boost fsm is still up and running in development. Look in the sand-box CSV for the code. Not sure how updated the code is, but Andreas is certainly still working on it.
I like the look of FSM very much, but had troubles compiling it with bcc32 5.6.4 last time I tried so have stuck with out current method of implementing statemachines.
Just days ago new version of fsm was released. I tried BCB port already but it needs some more work. If you can help, it'll be great. /Pavel
Pavel Vozenilek wrote:
Just days ago new version of fsm was released.
I tried BCB port already but it needs some more work. If you can help, it'll be great.
I would like to. I'll just have to see how the next couple of weeks go work-wise. I am very interested in fsm, though, so i'll do my bets. Cheers Russell
Russell Hind <rhind <at> mac.com> writes:
I would like to. I'll just have to see how the next couple of weeks go work-wise. I am very interested in fsm, though, so i'll do my bets.
Since I have never worked with bcc, Pavel and I would very much appreciate any help, be it advice on how to work around bcc bugs or actual porting... Thanks & Regards, Andreas
Pavel Vozenilek <pavel_vozenilek <at> hotmail.com> writes:
"Russell Hind" <rhind <at> mac.com> wrote
Boost fsm is still up and running in development. Look in the sand-box CSV for the code. Not sure how updated the code is, but Andreas is certainly still working on it.
I like the look of FSM very much, but had troubles compiling it with bcc32 5.6.4 last time I tried so have stuck with out current method of implementing statemachines.
Just days ago new version of fsm was released.
I tried BCB port already but it needs some more work. If you can help, it'll be great.
I know, Pavel, I promised to help you a week ago. Sorry for not getting back to you. Unfortunately, there were some unforeseen design-issues and, as I heard today, one remaining minor compilation problem with gcc on Linux. I hope to have everything resolved by the end of the week. I'll be able to help you after that. Regards, Andreas
jp <jp <at> comrad.co.nz> writes:
i dont kno if there is now a boost::fsm. the above messages were in 2002!
You'll find the most recent version of boost::fsm here: http://boost-sandbox.sf.net/fsm.zip I think code and documentation are pretty mature now and the library is already being used in two independent real-world projects. I'll make an announcement in all relevant newsgroups in a few days. If feedback is as I hope I'll ask for a formal review here at boost.
anyway, i have been using an object oriented fsm pattern for years now which uses virtual function dispatching in place of the old function pointer table thingy. i find it very robust & easy to understand. i have made a few preliminary ventures into boosting it into a generic template, and i wonder if theres still any interest in this idea.
Although it's a little late now, I'm certainly interested in how your library works. Do you have a tutorial or some example code demonstrating it's use? Regards, Andreas
Andreas Huber <ah2003@gmx.net> writes:
jp <jp <at> comrad.co.nz> writes:
i dont kno if there is now a boost::fsm. the above messages were in 2002!
You'll find the most recent version of boost::fsm here:
Not meant to compete with Andreas' submission, but I've enclosed the latest FSM examples from the upcoming MPL book. There are two implementations of the framework that use essentially the same interface (an identical DSEL) but entirely different dispatching mechanisms. The first one dispatches with a "compile-time case statement" keyed on the current state. The second one dispatches with an O(1) lookup in a static const table generated at initialization time. Other schemes and optimizations are clearly possible. The approach shows the power of declarative languages for expressing domain-specific programs. Cheers, -- Dave Abrahams Boost Consulting www.boost-consulting.com
"David Abrahams" <dave@boost-consulting.com> wrote in message news:uvfkwcl5q.fsf@boost-consulting.com...
[...] Not meant to compete with Andreas' submission, but I've enclosed the latest FSM examples from the upcoming MPL book. [...]
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here). Dave --- Outgoing mail is certified Virus Free. Checked by AVG anti-virus system (http://www.grisoft.com). Version: 6.0.581 / Virus Database: 368 - Release Date: 2/9/2004
"David B. Held" <dheld@codelogicconsulting.com> writes:
"David Abrahams" <dave@boost-consulting.com> wrote in message news:uvfkwcl5q.fsf@boost-consulting.com...
[...] Not meant to compete with Andreas' submission, but I've enclosed the latest FSM examples from the upcoming MPL book. [...]
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm. What about this? ====================== -- Dave Abrahams Boost Consulting www.boost-consulting.com
Not yet. (Outlook Express) "David Abrahams" <dave@boost-consulting.com> wrote in message news:u7jxb9482.fsf@boost-consulting.com...
"David B. Held" <dheld@codelogicconsulting.com> writes:
"David Abrahams" <dave@boost-consulting.com> wrote in message news:uvfkwcl5q.fsf@boost-consulting.com...
[...] Not meant to compete with Andreas' submission, but I've enclosed the latest FSM examples from the upcoming MPL book. [...]
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm.
What about this?
---------------------------------------------------------------------------- ----
======================
---------------------------------------------------------------------------- ----
-- Dave Abrahams Boost Consulting www.boost-consulting.com
---------------------------------------------------------------------------- ----
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
"David Abrahams" <dave@boost-consulting.com> wrote in message news:u7jxb9482.fsf@boost-consulting.com...
"David B. Held" <dheld@codelogicconsulting.com> writes:
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm.
What about this?
It's not just yours Dave. Attachments since 3/21 show(Outlook Express) the attachment paperclip but no attachments are available. These include attachment from Vladimir Prus, Ulrich Eckhardt, and Larry Evans. Perhaps something odd with GMane? Jeff F
Jeff Flinn wrote:
"David Abrahams" <dave@boost-consulting.com> wrote in message news:u7jxb9482.fsf@boost-consulting.com...
"David B. Held" <dheld@codelogicconsulting.com> writes:
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm.
What about this?
It's not just yours Dave. Attachments since 3/21 show(Outlook Express) the attachment paperclip but no attachments are available. These include attachment from Vladimir Prus, Ulrich Eckhardt, and Larry Evans. Perhaps something odd with GMane?
Hardly. I don't see anything via regular email reader. Dave's message, in particular, includes three attachments, with different bits of signature, but nothing that was meant to be attached. Looks more like mailman issue. - Volodya
Vladimir Prus <ghost@cs.msu.su> writes:
Jeff Flinn wrote:
"David Abrahams" <dave@boost-consulting.com> wrote in message news:u7jxb9482.fsf@boost-consulting.com...
"David B. Held" <dheld@codelogicconsulting.com> writes:
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm.
What about this?
It's not just yours Dave. Attachments since 3/21 show(Outlook Express) the attachment paperclip but no attachments are available. These include attachment from Vladimir Prus, Ulrich Eckhardt, and Larry Evans. Perhaps something odd with GMane?
Hardly. I don't see anything via regular email reader. Dave's message, in particular, includes three attachments, with different bits of signature, but nothing that was meant to be attached. Looks more like mailman issue.
Indeed. I don't know if it's a recent change, but the list of allowable attachment types has been set to: multipart/mixed multipart/alternative text/plain I just added: text/x-cplusplus I'm not sure if there should be any restriction, or how the restriction got there in the first place. Here are the attachments again. -- Dave Abrahams Boost Consulting www.boost-consulting.com
David Abrahams <dave@boost-consulting.com> writes:
[snip]
Indeed. I don't know if it's a recent change, but the list of allowable attachment types has been set to:
multipart/mixed multipart/alternative text/plain
I just added:
text/x-cplusplus
I can think of a large number of possible additional mime types that should likely be allowed, such as the following alternative mime types for C++ and C code: text/x-c++src text/x-c++hdr text/x-csrc text/x-chdr In addition, text/x-patch is sometimes used for patch files, which I've attached to messages to this list on a few occasions in the past. I can also imagine some cases in which it would be useful to be able to attach HTML, XML, and (other) SGML files, which cover a large number of mime types. What is the reason for restricting the types of attachments anyway? There doesn't seem to be much or any spam which gets through to the list. -- Jeremy Maitin-Shepard
David Abrahams <dave@boost-consulting.com> writes:
Jeremy Maitin-Shepard <jbms@attbi.com> writes:
What is the reason for restricting the types of attachments anyway?
If you re-read my posting, you'll see that I don't know. I'm hoping someone who did it will speak up.
Sorry, I should have read more carefully. -- Jeremy Maitin-Shepard
From: David Abrahams <dave@boost-consulting.com>
--=-=-=
"David B. Held" <dheld@codelogicconsulting.com> writes:
"David Abrahams" <dave@boost-consulting.com> wrote in message news:uvfkwcl5q.fsf@boost-consulting.com...
[...] Not meant to compete with Andreas' submission, but I've enclosed the latest FSM examples from the upcoming MPL book. [...]
Maybe it's just my OE, but I can't see your attachment (or the last three that have been posted here).
Hmm.
What about this?
Nope. I'm using rmail in emacs. I didn't reply with all of the headers, but there's no attachment.
--=-=-=
======================
--=-=-=
-- Dave Abrahams Boost Consulting www.boost-consulting.com
--=-=-= Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline
_______________________________________________ Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost --=-=-=--
-- Rob Stewart stewart@sig.com Software Engineer http://www.sig.com Susquehanna International Group, LLP using std::disclaimer;
participants (4)
-
Andreas Huber -
David Abrahams -
David B. Held -
David Bergman -
Jeff Flinn -
Jeremy Maitin-Shepard -
jp -
Pavel Vozenilek -
Rob Stewart -
Russell Hind -
Stefano Delli Ponti -
Vladimir Prus