DevIO

I/O tools for reading and writing MCH data in custom formats

This is mostly for debug and/or usage with Run2 (converted) data.

• Digit I/O
  • Digit sampler
  • Digit sink
  • Digit ID converter (from Run2 to Run3)
  • Digit random generator
  • Digit file dumper
• Precluster I/O
  • Precluster sink
• Cluster I/O
  • Cluster sampler
  • Cluster sink
• Track I/O
  • Track sampler
  • Track sink

Digit I/O

The MCH digits can be stored in binary format for debug purpose.

The file they are stored in can currently be of two different formats : with or without associated ROFRecord. In both cases the digits themselves (currently) have the same version (named D0). The file version number changes whenever the file structure changes or the format of the objects it contain changes.

File version	Digit version	contains ROF	ROF version	support
0	0	no	(0)	yes
1	0	yes	1	yes
2	1	yes	1	planned
X	X	yes	X	xxxxxxx

Digit sampler

Reads a binary file containing digits. The binary format should be embedded in the file directly. If the reader cannot identify the format it will assume it is the V0 version, which might or not might be the case...

o2-mch-digits-sampler-workflow
  --infile arg       input file name
  [--max-nof-tfs]    max number of timeframes to process
  [--first-tf]       first timeframe to process
  [--max-nof-rofs]   max number of ROFs to process
  [--repack-rofs]    number of rofs to repack into a timeframe
                     (aka min number of rofs per timeframe)
  [--print-digits]   print digits
  [--print-tfs]      print number of digits and rofs per tf

Digit sink

o2-mch-digits-sink-workflow
  [--outfile] arg (=digits.out)       output file name
  [--max-nof-tfs] arg (=2147483647)   max number of timeframes to process
  [--first-tf] arg (=0)               first timeframe to process
  [--no-file]                         no output to file
  [--binary-file-format] arg (=v1)    digit binary format to use
  [--print-digits]                    print digits
  [--print-tfs]                       print number of digits and rofs per tf
  [--txt]                             output digits in text format
  [--max-size] arg (=2147483647)      max output size (in KB)

Take as input a list of digits in timeframes (and optionally their associated Orbits) and dump them into a file (binary or text if --txt option is used) or to screen (if --no-file option is used)

Digit ID converter (from Run2 to Run3)

o2-mch-digits-r23-workflow
  [--verbose]        print ids being converted

A device that reads digits with ids in Run2 convention and outputs digits with ids in Run3 convention. Note that expected input spec is MCH/DIGITSRUN2 and the output one is MCH/DIGITS

Digit random generator

o2-mch-digits-random-generator-workflow
  [--first-tf] arg (=0)               first timeframe to generate
  [--max-nof-tfs] arg (=2147483647)   max number of timeframes to generate
  [--nof-rofs-per-tf]                 number of ROFs to generate in each timeframe
  [--occupancy]                       fraction of pads to generate digits for
  [--print-digits]                    print digits
  [--print-tfs]                       print number of digits and rofs per tf

Generate a fixed number of digits (equal to the total number of MCH pads times the occupancy) per ROFRecord for a fixed number of ROFRecord per time frames, for a fixed number of timeframes.

Note that the occupancy must be strictly positive and less than or equal to 1.

Example of use : generate digits (with 1% occupancy, 128 ROFs per TF, 10 TFs) and dump them into debug binary form :

o2-mch-digits-random-generator-workflow -b
  --max-nof-tfs 10
  --nof-rofs-per-tf 128
  --occupancy 0.01 |
o2-mch-digits-writer-workflow -b
  --print-tfs
  --binary-file-format 1
  --outfile digits.v1.out

$ ls -alrth digits.v1.out
-rw-r--r--  1 laurent  staff   258M 15 avr 10:30 digits.v1.out
 
$ o2-mch-digits-file-dumper --infile digits.v1.out -c -d
[ file version 1 digit version 0 size 20 rof version 1 size 16 hasRof true run2ids false ] formatWord 1224998065220435759
nTFs 10 nROFs 1280 nDigits 13506560

The generated digit file can then be injected into other workflows using o2-mch-digits-file-reader-workflow described above. (digits do not need to be written to disc, the o2-mch-digits-random-generator-workflow can of course also be used as the first stage of a multi device workflow using pipes).

Digit file dumper

The digit file dumper is not, in contrast to all the program above, a DPL worfklow, but a "regular" simple program, that dumps digit information on screen.

o2-mch-digits-dumper
options:
  -h [ --help ]          produce help message
  -i [ --infile ] arg    input file name
  -c [ --count ]         count items (rofs, tfs, etc...)
  -d [ --describe ]      describe file format
  -p [ --print-digits ]  print digits
  -t [ --print-tfs ]     print number of digits and rofs per tf
  --max-nof-tfs arg      max number of timeframes to process
  --first-tf arg         first timeframe to process

Precluster I/O

Precluster sink

o2-mch-preclusters-sink-workflow --outfile "preclusters.out"

Take as input the list of all preclusters (PreCluster) in the current time frame, the list of all associated digits (Digit) and the list of ROF records (ROFRecord) pointing to the preclusters associated to each interaction, with the data description "PRECLUSTERS", "PRECLUSTERDIGITS" and "PRECLUSTERROFS", respectively, and write them event-by-event in the binary file preclusters.out with the following format for each event:

• number of preclusters (int)
• number of associated digits (int)
• list of preclusters (PreCluster)
• list of associated digits (Digit)

Option --txt allows to write the preclusters in the output file in text format.

Option --useRun2DigitUID allows to convert the run3 pad ID stored in the digit data member mPadID into a digit UID in run2 format.

Cluster I/O

Cluster sampler

o2-mch-clusters-sampler-workflow --infile "clusters.in" [--global]

where clusters.in is a binary file containing for each event:

• number of clusters (int)
• number of associated digits (int)
• list of clusters (Cluster)
• list of associated digits (Digit)

Send the list of all clusters (Cluster) in the current time frame, with the data description "CLUSTERS" (or "GLOBALCLUSTERS" if --global option is used), the list of ROF records (ROFRecord) pointing to the clusters associated to each interaction, with the data description "CLUSTERROFS", and the list of digits (Digit) associated to clusters, with the data description "CLUSTERDIGITS".

Option --nEventsPerTF xxx allows to set the number of events (i.e. ROF records) to send per time frame (default = 1).

Option --no-digits allows to do not send the associated digits.

Cluster sink

o2-mch-clusters-sink-workflow --outfile "clusters.out" [--txt] [--no-digits] [--global]

Take as input the list of all clusters (Cluster) in the current time frame, the list of all associated digits (Digit), unless --no-digits option is used, and the list of ROF records (ROFRecord) pointing to the clusters associated to each interaction, with the data description "CLUSTERS" (or "GLOBALCLUSTERS" if --global option is used), "CLUSTERDIGITS" and "CLUSTERROFS", respectively, and write them event-by-event in the binary file clusters.out with the following format for each event:

• number of clusters (int)
• number of associated digits (int) (= 0 if --no-digits is used)
• list of clusters (Cluster)
• list of associated digits (Digit) (unless option --no-digits is used)

Option --txt allows to write the clusters in the output file in text format.

Option --useRun2DigitUID allows to convert the run3 pad ID stored in the digit data member mPadID into a digit UID in run2 format.

Track I/O

Track sampler

o2-mch-tracks-sampler-workflow --infile "tracks.in"

where tracks.in is a binary file with the same format as the one written by the workflow o2-mch-tracks-sink-workflow

Send the list of all MCH tracks (TrackMCH) in the current time frame, the list of all associated clusters (Cluster) and the list of ROF records (ROFRecord) pointing to the tracks associated to each interaction in three separate messages with the data description "TRACKS", "TRACKCLUSTERS" and "TRACKROFS", respectively.

Option --forTrackFitter allows to send the messages with the data description "TRACKSIN", "TRACKCLUSTERSIN" and TRACKROFSIN, respectively, as expected by the workflow o2-mch-tracks-to-tracks-workflow.

Option --nEventsPerTF xxx allows to set the number of events (i.e. ROF records) to send per time frame (default = 1).

Track sink

o2-mch-tracks-sink-workflow --outfile "tracks.out"

Take as input the list of all tracks at vertex (TrackAtVtxStruct) in the current time frame, the list of all MCH tracks (TrackMCH), the list of all associated clusters (Cluster) and the list of ROF records (ROFRecord) pointing to the MCH tracks associated to each interaction, with the data description "TRACKSATVERTEX", "TRACKS", "TRACKCLUSTERS" and "TRACKROFS", respectively, and write them event-by-event in the binary file tracks.out with the following format for each event:

• number of tracks at vertex (int)
• number of MCH tracks (int)
• number of associated clusters (int)
• list of tracks at vertex (TrackAtVtxStruct)
• list of MCH tracks (TrackMCH)
• list of associated clusters (Cluster)

Option --tracksAtVertexOnly allows to take as input and write only the tracks at vertex (number of MCH tracks and number of associated clusters = 0).

Option --mchTracksOnly allows to take as input and write only the MCH tracks and associated clusters (number of tracks at vertex = 0).