search

OpenPhilology / phaidra

eLearning for historical languages.
19 stars 6 forks source link

Remove CypherAlign.py and instead provide proper backups of the database. #41

Closed mlent closed 9 years ago

mlent commented 10 years ago

Right now, whenever an aspect of the graph database changes, we have to run this huge script, which takes exponentially longer the more texts are added.

We should instead get a dump of the database after you've made implemented changes via the parser, which we can directly restore from.

mlent commented 10 years ago

@mariamoritz so far, this has worked just fine for me. Can you add functionality in, such that it checks before creating new nodes?

E.g. If I run this script several times, I end up with several documents, all the same extra translation of the document. It would be better if it looked for the document/node first before creating.