HI Mikhail,

See below.

On Wed, Sep 10, 2014 at 10:03 AM, [log in to unmask] <[log in to unmask]> wrote:

Hello,

I am trying to read some "final"/reconstructed LCIO or ROOT files in order to understand how to proceed with my task. For the beginning I took few files mentioned at:
https://confluence.slac.stanford.edu/display/hpsg/Finding+Monte+Carlo+data+at+SLAC

e.g. this one:
ap2.2gev075mev_001_SLIC-v2r11p1_geant4-v9r3p2_QGSP_BERT_HPS-Proposal2014-v3-2pt2_readout_recon.slcio

(the corresponding ROOT files do not load with the last LCIO library saying: "The StreamerInfo of class HpsEvent read from file has the same version (=1) as the active class but a different checksum. You should update the version to ClassDef(HpsEvent,2). Do not try to write objects with the current class definition, the files will not be readable.")

This is simply pointing to the fact that the DST file (ROOT file) that is being read was generated by and older/outdated version of the DST maker. The DST file will need to be regenerated using the latest version of the DST maker.

At the moment, every time an update to the DST maker or the HpsEvent API is made, the DST files need to be regenerated. Otherwise, you will get an error similar to what is shown above. I can "fix" this error by increment the class version number every time a change occurs, but this might cause conflicts elsewhere.

Once we decide to make a 1.0 release of the DST maker, this should no longer be an issue since most people will just use the tagged release. Until then, any update to the DST maker or the HpsEvent API will also require regenerating all files.

I converted the working file into ROOT DST by means of the "dst_maker". By the way most of the other files available on that page, e.g.:

ap075mev_egs_tri_2.2gev_0.00125x0_200na_5e5b_30mr_001_SLIC-v2r11p1_geant4-v9r3p2_QGSP_BERT_HPS-Proposal2014-v3-2pt2_readout_recon.slcio

egs_tri_2.2gev_0.00125x0_200na_5e5b_30mr_001_SLIC-v2r11p1_geant4-v9r3p2_QGSP_BERT_HPS-Proposal2014-v3-2pt2_readout_recon.slcio

do not convert, causing a segmentation fault of "dst_maker".

Can you send me the error? The LCIO file that you are trying to generate a DST file from is very old and is likely missing several collections required by the DST maker. This might be causing the seg faults but I wont be able to verify this until I see the errors.

Now, I tried to look into the obtained ROOT file, there are some tracks and EC hits, but there are no "_particles" (except for, I guess, input mc_particles).

This is likely due to the fact that the LCIO recon file is missing all of the ReconParticle collections. The DST wont have any particle collections if these are missing.

Am I missing something or using obsolete "examples" of MC data? Or the construction of reconstructed particles does not work yet?

Another question, on the page:
https://confluence.slac.stanford.edu/display/hpsg/HPS+Data+Summary+Tapes

you provided a list of DST variables, however if one could spend a few minutes explaining what are for example:
tracks: d0, phi, omega, tan_lambda, z0,
svt_hits: covariance matrix cxy,czz...,
mc_particles: endpoint,
gbl_tracks: kappa, theta, phi, d0, z0, seed_kappa, seed_theta, seed_phi, seed_d0, seed_z0, chi^2, ndf, momentum, covariance

it would be much more helpful.

I can update the page with a summary of what these variables are, but it likely wont happen until the end of the week.

Best Regards,
Mikhail.

Use REPLY-ALL to reply to list

To unsubscribe from the HPS-SOFTWARE list, click the following link:
https://listserv.slac.stanford.edu/cgi-bin/wa?SUBED1=HPS-SOFTWARE&A=1