Need help with plaintext database driver?

Question

linux

Greetings.
There are a huge number of text files in a large tree of subdirectories.
A fairly large number of different software accesses these files, they read them, create new directories, write or update existing files with replacement, create new ones.
There is a task to transfer these files to the NoSQL database (redis / mongo).
Putting files in the database is not a problem, the problem is getting the software written many years ago to read/write.
I'm thinking of writing a layer that will look something like a /proc file system.
Thus, with minimal cost, it will be possible to adapt existing software.
What are the ready-made solutions (are there any)? How it is possible to implement similar in Linux (only fuse?) How it is correct to name it, what to esteem on this subject in Google?

Answer 1

There was a discussion of a similar question here For CouchDB, there is couchdb-fuse
And, yes, you need to look for mongodb fuse

Answer 2

Not only fuse. More via webdav is not uncommon