The Readout GUI works with the event logger to maintain a directory structure of event files and associated data. You can associate an arbitrary set of files and directory subtrees with each run. Through heavy use of symbolic links, the Readout GUI provides you with two views of your data.
Events view. The events view presents all of the event files in a single directory. This view allows you to sequentially analyze several runs without needing to move around in the directory hierarchy.
Runs view. This view collects all of the data associated with each run into a single directory.
The top of the directory subtree that is managed by Readout Gui is determined by a symbolic link; ~/stagearea or, alternatively, the EVENTS environment variable if it is defined. [1] This must be created by you when you set the account up for data taking and should point to the event area that you were assigned.
When Readout GUI runs the first time it will, if necessary, create the directory structure it need underneath the directory pointed to by ~/stagearea.
The directory tree created looks like this:
Figure 7-4. Readout GUI Directory tree
stagearea +
+----> experiment+
| +---> current
| +---> run1
| +---> run2
... ...
+----> complete
Let's start with stagearea/experiment. The current directory is one location you can put the ReadoutCallouts.tcl customization file. (new behavior in 12.0x) When event recording begins, the readout Gui creates a run directory with the same name as the requested run number. Events are recorded in that directory.
In 11.x, event files were segmented into approximately 2Gbyte chunks. This was done to meet maximum file size requirements of older file systems that are no longer prevalent. In 12.x the default segment size is 1,000 Tbytes which is effectively unlimited. In 11.x, event files were recorded in stagearea/experiment/current. In 12.x they are recorded directly in their run directories.
To retain compatiblity with 11.x software that assumes the structure of the stage area, while event recording is underway, event files in the current run directory are given symbolic links in the stagearea/experiment/current directory. These links are removed at the end of the run.
When event recording is completed for a run, the symbolic links to event files are removed from stagearea/experiment/current. Furthermore a link following copy of the files in that directory are made into the run directory allowing you to associate arbitrary data with each run. Link followed copy means that if there are symbolic links, the item those links point to are copied rather than the links. This allows you a hook to copy data that isn't in the stagearea directory tree.
Finally, when all this copying is done, the events view is updated by creating symbolic links to the run directory event files in stagearea/complete.
| [1] | Environment variables can override this link however and are useful in the event you are using a single account to do multiple tests for which you want to segregate the data. |