Reactivate CIRCE2

[kilian]

CIRCE and CIRCE2 are essential for ILC physics. Reactivate this part which was working with W1.

[Juergen Reuter]

If possible (and available) also here the noweb sources with comments should be included. For both programs we should upgrade to the newest design proposals for ILC and CLIC.

[Juergen Reuter]

Thereby, fixing ugliness in the Makefiles of CIRCE/CIRCE2: In src/circe/Makefile.am we find

LIB=../@LIB@

IMHO, automake files should never use @variable@ but $(variable), but

variable=../$(variable)

is bad. Find a cleaner idiomatic expression. NB: this last point has low priority.

[Juergen Reuter]

I decide in favour of not including circe_dummy.f90 and circe2_dummy.f90 files into the WHIZARD 2 distribution. There is no need to exclude somehow CIRCE or CIRCE2 from the distribution.

[Juergen Reuter]

Small progress in r2171. But I guess we have to rewrite CIRCE completely in F90 as it bursts over with common blocks. :(

[Juergen Reuter]

Ok we postpone this.

[Juergen Reuter]

One of the most important tasks for 2.0.2 after polarization has been done.

[Juergen Reuter]

OK, the basic structure of CIRCE1 and CIRCE2 are autotoolized as of revision r2586. libcirce1 and libcirce2 are being built, the documentations are being made, and make distcheck works in both cases. The extra targets in the old Makefiles seem to be VERY much deprecated, and will not be enabled by myself.

[kilian]

CIRCE1 works with r2653. What doesn't work and needs fixing is chaining structure functions, e.g., circe1 => isr.

CIRCE2 not yet.

[kilian]

Chaining fixed in r2655.

[kilian]

With r2663, a few remaining bugs are fixed (the decay chain was activated even if there was no decay possible).

[Juergen Reuter]

CIRCE2 is still missing. :( It should be quasi-analogous to CIRCE1, EXCEPT for the polarization. I do not dare attach the polarizations into the state matrix constructions, WK should do this!

[Juergen Reuter]

Since WK is back from the uncivilized realms I'm taking over.

[Juergen Reuter]

Using CIRCE1 in WHIZARD2 issues warnings like:

 circe:error:   invalid `acc':        0
 circe:message: keeping `acc' at TESLA
 circe:warning: *************************************
 circe:warning: * This release is not official yet, *
 circe:warning: * do not use it in publications!    *
 circe:warning: *************************************

After a proper validation this should be removed.

[Juergen Reuter]

The paths for the data files are now correctly set within WHIZARD. Now the program terminates with the following:

| Initializating integration for process ilc_ww:
******************************************************************************
******************************************************************************
*** WHIZARD BUG: Structure functions: colliding particles can't be determined
******************************************************************************
******************************************************************************
WHIZARD run aborted.

[Juergen Reuter]

After the histogram files get installed and are properly communicated to the WHIZARD core, I give the ticket back to WK (for connecting the structure functions) and to TO (for clean-up and generating the new designs).

[kilian]

Let's see what's left to do ...

[kilian]

CIRCE2 is reactivated in r2786.