[Halld-offline] Geometry from CCDB

[Sean Dobbs]

Hi Offliners,

To follow up on the target position shift that's been discussed in the
analysis meetings and elsewhere, Simon has verified with the survey data
that there was a shift between the 2016 and 2017 running periods.  Note
that thanks to Simon, the correct target positions have been in the CCDB
since June.  I've updated the CCDB geometry files to include all of the
latest changes, including the FCAL geometry updates, retaining the correct
target locations.

I'm not sure if there was a formal announcement, but JANA does support
reading the HDDS geometry from CCDB.  I propose that this be the preferred
way of running, since this naturally takes into account the run dependence
of the geometry.

To do this, just set JANA_GEOMETRY_URL as follows:

bash:  export JANA_GEOMETRY_URL=“ccdb:///GEOMETRY/main_HDDS.xml"
csh:  setenv JANA_GEOMETRY_URL ccdb:///GEOMETRY/main_HDDS.xml

This will suffice for reconstruction. Note that the current HDDS master
tracks the current (2017) configuration, and the "runperiod-2016-02" branch
will track the 2016 configuration.

For simulations, you will need to have different versions of sim-recon
built for the HDDS versions for the 2016 and 2017 data.  This is
unavoidable when using hdgeant.

I'd like to have some feedback from experts before rolling this out to the
full collaboration, so please try this over the next couple of weeks and
let me know if you run into any problems.

Cheers,
Sean
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.jlab.org/pipermail/halld-offline/attachments/20171120/a494d034/attachment.html>