This repository contains scripts for generating and manipulating OKS database files.
Commandline utility to visually inspect and verify configurations databases and the objects they contain. Documenation available here.
Add Resource objects to or remove from the disabled relationship of a Session
Merge the contents of several database files, putting all objects into a single output file. Output file will only include schemas.
Merge the contents of several database files, preserving included databases. Output file will contain only objects defined in files given on command line.
Copy the input file(s) to the specified directory, also moving any included files and updating include paths, to create a clone of the configuration databases.
Retrieve the DAQ applications defined in the given configuration
Ensure that database files are in the "DBE format", alphabetized and with correct spacing
Enable or disable TPG for a Session's ReadoutApplications
Attempt to determine if a given Session configuration is valid and does not contain common errors
Attempt to replicate OKS' Data Base editor within Python. Full details are here. Current implementation is very incomplete so use with caution.
A script that generates an 'empty' OKS database, just containging the include files for the core schema and any other schema/data files you specify on the commad line.
Convert a JSON readout map file from dunedaq v4 to an OKS file.
Create an OKS configuration file defining ReadoutApplications for all readout groups defined in a readout map.
Read the DUNE-DAQ asset file database and return a path to a referenced asset file
A collection of methods to generate segments and sessions.
### generate_hwmap.py
Create a set of DetectorToDaqConnection objects, GeoIDs, and streams for the given number of links and applications.
Utilities for parsing OKS databases. Currently contains an include file search routine.
In daqconf it's possible to create a diagram of a DAQ configuration in the DOT graph description language, which can then be fed to the dot program to generate a viewable graphic. This is accomplished using the create_config_plot application. You can run create_config_plot -h to see how it's used, but to explain it simply, you provide it with a database file, the name of a session in that database, and then the name of an object associated with the session. It will create a plot using that object as the root of the plot. Note the object either needs to be the session itself, or a segment or an application in it. E.g., with the daqsystemtest repo you can generate a plot of the mlt application in its ehn1-local-2x3-config test DAQ session via:
create_config_plot -f config/daqsystemtest/example-configs.data.xml -s ehn1-local-2x3-config -r mlt
and if you want to plot the entire session, you can run
create_config_plot -f config/daqsystemtest/example-configs.data.xml -s ehn1-local-2x3-config -r ehn1-local-2x3-config
or a shorthand version of that command is
create_config_plot -f config/daqsystemtest/example-configs.data.xml -s ehn1-local-2x3-config
...where if no root object is specified the default is for create_config_plot to take the session passed to -s and use it as the root object. Also note that in the event that there's a single session in the database, the -s <session-name> argument isn't required. Any of the above commands will create (or clobber) a file called config.dot. If you wish to give the file a different name you can use the -o option, e.g., -o mypreferredname.dot. Once you have the DOT file, you can generate a graphic by doing the following:
dot -Tsvg -o mypreferredname.svg config.dot
...which can then be displayed, e.g., in a browser window via the link file:///path/to/file/mypreferredname.svg.
Note that while dot is available on the np04 cluster's system, it's not necessarily available on other hosts. If you're on another host, you may need to execute spack load graphviz in your work area to get the dot executable. Note that in this scenario that after you're done using dot if you wish to develop code you'll want to spack unload graphviz; more generally, your build environment isn't guaranteed to continue working after you run spack load <package>.