PDF Search

In no particular order, here are the primary requirements: This is a new project - Basically a document (PDF) management website that uses OCR to catalog the metadata and contents of the documents and allows searching of all this data. I was thinking of Ruby on Rails, but again this project is in the planning stages so this could be done otherwise. - there are three basic levels of permission: guest, member, admin (although it might be more flexible to assign each user permissions for each action instead of having permissions grouped together) - allow users to self-manage their account: password reset, updated password and/or name and email address - allow admins to manage users: create/update/change permissions/remove users (self registration is not needed) - documents are multi-page PDF files containing articles (mostly text) - OCR and pulling text from the PDF itself should be used to store full text/metadata about each document in a searchable (elastic search?) format - document metadata fields should be defined through the website by a permitted user - document metadata should be modifiable by permitted users for manual corrections and setting data - documents can be added to the site (by permitted users) through multiple methods like upload in browser (single PDF or ZIP file), upload with FTP, send as attachment to special email address, or other methods to add in the future. Initially just the upload in the browser is important, but being able to add their existing documents will also be important (which is pretty similar in function to the “upload with FTP” method) - guests (anyone) should be able to search the documents based on their metadata. the search should have a “basic” form thats just a single text field that searches all/the important fields and there should also be an “advanced” form that allows search specific metadata fields (like author, date published, and other metadata fields) - permitted users should also be able to view the document in the browser (using [login to view URL] or similar) and download the raw PDF I was thinking of doing this in RoR using the following : twitter bootstrap, jquery, [login to view URL], devise, cancancan, docsplit. The specific implementation is not important as long as its relatively easy to setup and maintain. Client already has a Redhat-based system with Plesk, so ideally it would run on that, but they are not opposed to running a small, cheap VM somewhere as a dedicated application server (using something like digital ocean or linode). If you would, please let me know a project time table, hours expected of work, and cost for the project?