Documents the process and automation for bringing new content into the mapping environment from the authoring environment for the prod deployment.
The drip feed, or the automated updating of SNOMEDCT content in the Mapping Tool from the authoring environment, runs weekly and consists of the following steps
- Obtain the latest daily build
- Process the daily build
- Remove and recalculate the SNOMEDCT tree positions
- Recompute workflow.
The source code for the loadDelta.csh script is in the config project for the prod deployment - config/prod/src/main/resources/bin.
The drip feed wrapper script (see the next section) is automatically run via the crontab application. A sample cron job for running the drip feed every sunday at 7pm is shown below:
Running the Drip Feed Wrapper Script
The wrapper script loadDelta.csh can be run by:
The wrapper script automatically performs all of the tasks described in the next section.
Manually Performing the Drip Feed
To manually run the drip feed, perform each of the following steps.
Restart the server
To prevent indexing and workflow errors due to editing concurrent with the drip feed, stop the tomcat server
Obtain the latest daily build
The daily build data is retrieved via the following maven get command. Note that the version must be specified and is subject to change.
The daily build data can also be retrieved from the ihtsdo aceworkspace repository at https://mgr.servers.aceworkspace.net/apps/ihtsdo-archiva/repository/all/org/ihtsdo/intl/release/process/wb-release-process/. At the time of writing, the repository interface is subject to a bug where the contents of the directories are not automatically shown. To view the daily builds:
- Enter an invalid URL by appending some string to the address, e.g. https://mgr.servers.aceworkspace.net/apps/ihtsdo-archiva/repository/all/org/ihtsdo/intl/release/process/wb-release-process/abc
- When prompted, log in
- Return to the valid URL specified above
- If no content appears, hit Refresh to show the daily builds
Once the data is obtained, unzip it and place it in the desired folder. The wrapper script automatically places the unzipped data in /home/ihtsdo/.m2/repository/org/ihtsdo/intl/release/process/wb-release-process/1.20-SNAPSHOT/wb-release-process-1.20-SNAPSHOT-delta/destination/Delta.
Run the delta loader mojo
The delta loader mojo is named TerminologyRF2DeltaLoader. The mojo is specifically configured for SNOMEDCT, and requires specification of parameters including the preferred term type id and refset id, which are set in the configuration properties file.
To run the delta loader, execute the command below. Note again that if the daily build version changes, this command must be altered.
Clear and recompute SNOMEDCT tree positions
See Terminology Tree Position Management for instructions on removing and calculating tree positions.
See Workflow Tools for instructions on computing workflow.
Start the server
Restart the tomcat7 server: