0

I have a requirement to develop a Document Repository which will maintain all documents related to different listed Companies. Each document will be related to a Company. It has to be REST API. Documents can be in pdf, html, word or excel format. Along with storing documents, I need to store metadata as well like CompanyID, Doc format, timestamp, doc language etc. As the number of document will grow in years to come, its important that the application is scalable.

Also need to translate non-English doc and store it translated English version in some parent-child relation which is easy for retrieval.

Any insights on the approach, libraries/jars to use and best practices and references are welcome.

nanosoft
  • 2,913
  • 4
  • 41
  • 61

1 Answers1

0

The base 64 encoded content of the file could be included as the part of your payload along with file metadata.

Posting a File and Associated Data to a RESTful WebService preferably as JSON

Once the file reach to your end, you could save the file either local to your hard disk or save the same base64 encoded content as in your data store(user blob/clob).

Community
  • 1
  • 1
jean joseph
  • 82
  • 1
  • 2
  • which datastore is MORE efficient. Would NOSQL datastore would be more efficient? Off couse simply saving in local hard drive wont be efficient for retrieval. – nanosoft Apr 18 '17 at 11:19
  • we have implemented the same with Riak. Make sure that you create seperate table/document/bucket for metadata and file content. – jean joseph Apr 19 '17 at 05:58