[Rxtx] rxtx 2010 rewrite

Tue Nov 23 11:49:22 MST 2010

Check out the current repo using these instructions:

http://rxtx.qbang.org/wiki/index.php/Retrieving_Source_Code

and
 you will find the rewrite in the Rewrite2010 folder. There is a pdf 
document in that folder that gives an overview of the rewrite.

-Adrian

--- On Tue, 11/23/10, Brian Schlining <bschlining at gmail.com> wrote:

From: Brian Schlining <bschlining at gmail.com>
Subject: Re: [Rxtx] rxtx 2010 rewrite
To: rxtx at qbang.org
Date: Tuesday, November 23, 2010, 8:17 AM

Where's the source code for the rewrite?

On Tue, Nov 23, 2010 at 06:25, Adrian Crum <adrian.crum at yahoo.com> wrote:

I will be working on unit tests next. I have experience with JUnit and Cobertura. The rewrite contains software loopbacks for testing without hardware loopbacks, and I was thinking hardware loopbacks could be used via settings in the properties file - so CI can test for regressions. My hope is once we have 100% test coverage on the rewrite, the community will have greater confidence in it.

-Adrian

--- On Mon, 11/22/10, Trent Jarvi <tjarvi at qbang.org> wrote:

> From: Trent Jarvi <tjarvi at qbang.org>

> Subject: Re: [Rxtx] rxtx 2010 rewrite

> To: rxtx at qbang.org

> Cc: "Dave" <davestechshop at gmail.com>

> Date: Monday, November 22, 2010, 7:57 PM

>

> I'm not for or against the rewrite effort.  I do

> represent one of those people needing 'stable now' when I go

> to work.  Here at home I'm interested in the future of

> RXTX for all those neat projects I hear about. I'm all for

> some experimenting and growing.

>

> What RXTX lacks is not coders.  We have had many ideas

> bouncing around and some code that deals with things like

> serial ports that vanish (USB).  The C++ rewrite is

> proof of interest there and I'm glad to see it.

>

> So whats missing?

>

> The JCP required test suites.  CommAPI and RXTX lack

> test suites. CommAPI was grandfathered into the JCP and

> shortly later 'broken' for a customer right in the JCP by

> adding Solaris specific hooks.

>

> Testing CommAPI/RXTX is difficult.

>

> At work I have test suites but the coverage is specific to

> those needs - these tests existed before I started working

> there.  Maybe they cover your needs, maybe not.  I

> use those to qualify RXTX releases which is alot more than I

> used to do.  The tests are in MATLAB code going through

> a layer of java and don't lend well to sharing (though its a

> possibility). Those test suites are the best we all have at

> this time as far as I know and that does not bode well for a

> smooth transition to a rewrite for multiple projects.

>

>

> I could run the suites against the rewrite.  Honestly,

> I'm not comfortable that they cover enough for generic

> qualification though.  As an example, the testsuite

> requires a loopback connection.  Fun thing is the other

> side is always configured like your side.  Setting

> hardware flow control failed?  No problem.  Its

> still odd parity?  Good to go!  It failed on the

> other side too.

>

>

> In agreement with Julie, there are many subtle things in

> rxtx that took more than 3 years to flush out.  That

> does not mean we can't figure out how to move forward

> (without the circus wagons).

>

> On Mon, 22 Nov 2010, jfh at greenhousepc.com

> wrote:

>

> > Adrian,

> >

> > As I've said before, some of us are ready, willing and

> able to work on RXTX,

> > but we need something much more stable NOW, and the

> existing code base is

> > the most stable, most time-tested thing going.  It's

> easy to say "it should

> > work reliably", but my experience is that code like

> this typically takes 2

> > to 3 =years= before there is enough experience with

> the code to be called

> > "working reliably".

> >

> > What I'd like to know is who feels they have a

> =reliable=, =stable=,

> > relatively =bug-free= version of the existing code

> that could be a base for

> > something that can tide people over until your work is

> well enough tested

> > that people can start migrating less critical

> applications to it.  My

> > company puts products inside boxes that have to work

> for years on end.  I'm

> > not going to trust my company's existence to a brand

> new re-write, and I

> > suspect there are a number of other business owners or

> consultants who are

> > in the same situation.

> > --

> > Julie Haugh

> > Senior Design Engineer

> > greenHouse Computers, LLC // jfh at greenhousepc.com

> // greenHousePC on

> > Skype

> >

> >

> >       -------- Original

> Message --------

> >       Subject: Re: [Rxtx]

> rxtx 2010 rewrite

> >       From: Adrian Crum

> <adrian.crum at yahoo.com>

> >       Date: Sun, November 21,

> 2010 9:21 pm

> >       To: Dave <davestechshop at gmail.com>,

> rxtx at qbang.org

> >

> >       The rewrite is

> preliminary code - it needs more work. It should

> >       work reliably in a

> Windows environment. The rewrite needs more

> >       community participation

> to move it along.

> >

> >       The best resource for

> help is the RXTX mailing list -

> >       rxtx at qbang.org.

> >

> >

> >

> -----Inline Attachment Follows-----

>

> _______________________________________________

> Rxtx mailing list

> Rxtx at qbang.org

> http://mailman.qbang.org/mailman/listinfo/rxtx

>

_______________________________________________

Rxtx mailing list

Rxtx at qbang.org

http://mailman.qbang.org/mailman/listinfo/rxtx

-- 
~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 
Brian Schlining
bschlining at gmail.com

-----Inline Attachment Follows-----

_______________________________________________
Rxtx mailing list
Rxtx at qbang.org
http://mailman.qbang.org/mailman/listinfo/rxtx

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.qbang.org/pipermail/rxtx/attachments/20101123/b50d588c/attachment-0633.htm>