Added download by id and support to send upload stream in req.filedata

[sachinhub]

Hi,

While using this module, we felt a need to add support for following functionalities -

1. Download by id in addition to download by name.
2. While uploading, req object is expected to contain the stream data. However, we might read the stream before calling the API for processing like virus scanning. Hence added a support for sending upload stream in req.filedata variable as well.

[codecov-io]

Current coverage is 86.82%

Merging #7 into master will not affect coverage as of 1b8c829

@@            master      #7   diff @@
======================================
  Files            1       1       
  Stmts          129     129       
  Branches        14      14       
  Methods         30      30       
======================================
  Hit            112     112       
  Partial          0       0       
  Missed          17      17       

Review entire Coverage Diff as of 1b8c829

Powered by Codecov. Updated on successful CI builds.

[jdrouet]

@sachinhub thank you for the pull request but there is several problems. First, the code you have done is not in coffeescript. The second problem is that you don't make tests. Next, the prototype you need deffer from the standard loopback component storage (which is not really a problem). Could you fix it ?

[jdrouet]

When you store it in temporary location, you store it on the filesystem ? What is the difference between storing it in the filesystem and storing it in database ?

[sachinhub]

Yes @jdrouet. We store in a folder on file system and then execute a virus scan script on it. The folder is quarantine.

Once the virus scan command returns okay, we upload it in database. We dont want to upload in database until it passes virus scanning tests.

[jdrouet]

@sachinhub I think that you need to use the standard component storage first then upload it to database