The server uses the node.js express framework and, in addition to various middleware tasks, communicates with the poetic transformations database API.
The server is also responsible for generating two CSV files relating to the auto-generated spine index applied to the xml files hard coded into the 'data' folder. Should these files be updated, the spine index generation should be re-run and the CSV files will be re-populated.
The spine index is auto-generated using a cosine-similarity comparison (based on word similarity between each line of the poem).
All xquerys are saved for reference in the queries folder.
James Cummings
Tiago Sousa Garcia
To use Docker:
Build a container using
docker build -t pt/server .
(this will tag the container as pt/server, this can be changed as preferred).
Run the container in detached mode using docker run -d -p 3000:3000 pt/server
local server to docker exist DB_CONNECTION_STRING= "http://admin:@localhost:8080/exist/rest"
or
docker server to docker exist DB_CONNECTION_STRING = 'http://admin:@172.17.0.2:8080/exist/rest'
DB_CONNECTION_ATTEMPTS = 10 DB_CONNECTION_TIMEOUT = 10000 SENTRY_DSN
On startup the server tries to put data files into the database and errors if the database is not up or does not return the expected responses. This is enough to enable viewing the files.
To enable searching there are some extra steps: After the server has finished adding data, you need to navigate in the browser to the exist-db GUI (go to database URL and click 'exide'). The, navigate in the file structure panel on the left to db/system/config/db/transformations/collection.xconf. Open this file and then click 'save'. You should see a popup asking you to click 'OK'. Do this and the search feature should now work.