Print

Print



On Nov 4, 2015, at 10:35 AM, McCormick, Jeremy I. <[log in to unmask]> wrote:

The datacat has per file metadata so anything we want to calculate and store for each EVIO file.  I think maybe you're confusing it with the run database which is per run information.


Probably.

The SVT configuration bank contents likely belong in the run database if you want this stored (this is the XML config for the SVT DAQ right?).  I can do that if you provide me a code snippet for reading this information from an EVIO event.


We can do this when I get back.

The "svt_event_header_good" is not present in the EVIO files so that could be something for when we catalog the LCIO files.  Would it be useful to have a count of how many events are "good" in the file?


Right. Well itís already used in order to compute that but I donít know if there is someone planning to use any such catalog to figure that out in the future.

/Pelle



From: Hansson Adrian, Per Ola
Sent: Wednesday, November 4, 2015 9:35 AM
To: McCormick, Jeremy I.
Cc: hps-software
Subject: Re: file crawler for datacat
 

On Nov 3, 2015, at 5:04 PM, McCormick, Jeremy I. <[log in to unmask]> wrote:

Hi,

I'm in the process up updating the code that will extract file metadata for the data catalog.

Here's what I have included so far for EVIO files:

"standard" metadata expected by the file catalog...
run number
event count (should this only include physics events?)
file size

extras...
file sequence number
bad event count
first timestamp
last timestamp
first physics event number
last physics event number
END event present
PRESTART event present
EPICS event count
scaler bank count

I'm also including trigger bit counts (singles0, pairs0, etc.).

Anything else you would like to see in there while I'm working on this?


The SVT status and configuration evio banks should go in there I think. They are text banks that happen every 100k or 500k events so are pretty rare (in case your tests donít show them it might be because you are not running over enough events). 

Do we want to put any metadata from the SVT error checking in here?


Not sure exactly how this catalog would be used but it might make sense to include this flag from the event header; svt_event_header_good

If this header is not ok there will be other information in the header but I suspect that we donít want to catalog that? It can be a lot of information.


I know we wanted also the Faraday Cup charge per file, but I don't know if we want to instead use Sho's numbers here instead (probably more accurate but harder to integrate into my crawler which is only using the EVIO data).

I will move on to the recon/LCIO files next.

Thanks.

--Jeremy

########################################################################
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





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