[ts-gen] Re: quick note re: build environment

[R P Herrold]

On Wed, 1 Nov 2006, an anonymous inquirant wrote:

> "libmysqlclient15-dev is probably not strictly required -- I
> did that list above from memory, and after testing, it turns
> out that it is not needed to build the shim."
>
> Actually, after my latest fiddling session I can say that it 
> most definitely IS required to build the shim.  Without it, 
> I get a ton of errors.  This is probably at least partly due 
> to the fact that the build requires mysql.h, which isn't 
> present *anywhere* (ls -R|grep "mysql.h" confirms) until I 
> install libmysqlclient15-dev.
>
> Just an FYI--thought it might be worth including a note 
> somewhere in INSTALL for ubuntu/debian users like me.

Consider it done, at least as a start -- Google will pick this 
up from the archive; I moved the offline Quantian box back 
into one of the racks at work, and will probably power it up, 
and see if I cannot get it building and running with the shim 
in the next few days.

Stabilizing the code to also build cleanly with g++-gcc4 will 
be a little longer in coming, as there will be more rework 
needed.

> Still haven't managed to get the shim to be version-happy 
> yet, but still a few more things I'm going to try before I 
> start whining again.

I know that Bill pushed a nightly slipstream externally about 
1800 local today -- I see that it has been promoted into lead 
at:
 	ftp://ftp.trading-shim.org/pub/local/trading-shim/
by our cronned release management scripts.

It has been a very busy and productive week;  NEWS has the 
following, just in part, which relate to this week's efforts:

date: Thu Nov  2 2006
date: Wed Nov  1 2006

     * implement Cancel form of wire command, to be used with
 		Submit form
     * fall back to 862.9, due to intermittent syntax errors
 		during 864.4 market data message processing
     * Port shim to tws 862.9, breaking, for now, backward
 		compatability
     * integrate regression tests to one large functional test
 		against a common process instance of the shim,
 		exposing some problems due to the allocation
 		and numbering of tick ids, again  a known problem
     * add tick and past commands as market data and history
 		query alternative to loading subscriptions
 		from the database; as a known bug, they
 		interact badly with one another for now
     * Port shim to tws 864.4
     * Port shim to tws 862.9, breaking, for now, backward
 		compatability with the previous development
 		platform, 855.5
     * Improve the quality of the regression test scripts, for
 		the increased convenience of those who wish to
 		adapt them for  external use.
     * Fix intermittent but fatal handshake problem; please
 		consider upgrade

I posted about the screenshot demonstrating limited Order 
function, and we remain on the trail of nailing our needs in 
this space down; As I noted in the screenshoot, the order 
completed its round trip, from transmission of a well-formed 
request to fill confirmation, in 0.164853 Sec., and the shim 
finished reporting the fill in a formatted form to the 
downstream client through logging 0.000472 Sec. after that. 
Sub Second fills, and sub milliSecond reporting. We are really 
pleased with the speed here, in light of our focus on 'design 
for scalability to hundreds of simultaneous lines of inquiry'.
 	http://www.trading-shim.org/screenshots/

We moved to TWS 862.9a, so that we are MUCH closer to IB's 
latest (we see some strange failures with TWS 864.4 and had to 
drop back from it for now).  More importantly to stomping some 
remaining bugs, I have re-synchronized my build and test 
process to include all of his, to permit me to get reproducers 
which are more useful to him.

Not whining at all -- I think success is close, and that it is 
well useful to know the Debian derived approach on this.

I am active with the LSB, and one of the issues which is not 
under any central nameing authority (and probably never will 
be) is how to name packages properly -- the namespace 
possibilities and practice are just too diverse for any rule 
to please everyone, and there is no disincentive to differing.

Getting a lsb-mysql-dev[el] packaging is even farther out, I 
think.

Thanks for the added knowledge.

-- Russ Herrold