Create and use de-identified research databases. Preprocess, extract text, anonymise/de-identify, link, apply natural language processing, query for research, manage consent for contact.
Following on from #126 , this has the following changes:
Only rebuild the Docker image if an image doesn't exist with the same tag, unless the installer is called with the update option
When building the demo MySQL containers, provide a port on the host so that they can easily be viewed
Create a script to unset all of the CRATE_DOCKER_XXX and CRATE_INSTALLER_XXX settings
Split the Dockerfile into stages. This should in theory lead to quicker build times but in practice Docker doesn't cache beyond the copying of source files stage (ADD instruction). It's possibly because something in the source tree always changes and invalidates any caches beyond this stage. We could be more selective as to what gets copied.
Use ARG not export for variables in the Dockerfile. export does not persist between stages.
Following on from #126 , this has the following changes:
CRATE_DOCKER_XXX
andCRATE_INSTALLER_XXX
settingsADD
instruction). It's possibly because something in the source tree always changes and invalidates any caches beyond this stage. We could be more selective as to what gets copied.ARG
notexport
for variables in the Dockerfile.export
does not persist between stages.as_root
argument toinstaller.py exec