I am going to collect documents/assets of several types (webpages, PDFs, pictures, videos) that I want to be able to manage.
Some of those documents are a single file (e.g. PDF), some may span several files (e.g. a saved webpage) and in the latter case I want to administer the document as one entity, rather than just a set of files.
I want to be able to attach metadata to the documents (in particular tags, but also other informations (essentially key-value pairs).
It must run on Linux and be open-source and it should be "hackable", i.e., I would prefer a solution based on a scripting language rather than something compiled.
Ideally everything would be file-based rather than using a database but that is not a strict requirement.