This page details our design plans.

Indexer

• Crawler
  • finds files
  • notices file changes
  • creates change events
• ID based on filetype
• File movement analysis

Analyzer

• Error checker
• Codecs/Quality
• Content Info (Episode name/number, artist, etc)
• Full-file MD5 (for external comparisons)

Problems

Local

1. what is a "file"?
2. what is *in* a file?

Network

1. find a better/replacement file?

A File Is

Filename
File/Byte Stream
Content

Process

1. Indexer - /some/path/file
   • via (dbus/inotify/crawl)
   • dumps data (file list)
   • file movement checks - eliminates duplicate byte streams

Tables

Files

id
first seen time
last seen time
index run time
path
filename
byte stream id

Byte Stream

id
file type
file type id
hash

Content

byte stream id
length
number_warnings
number_errors
v_format
v_bitrate
v_width
v_height
v_fps
v_aspect
v_bitrate
v_codec
a_rate
a_nch
a_codec