Add build instructions / deps

README.md

@@ -25,7 +25,24 @@ used by C-bliss, since it uses
 different, more accurate values, based on
 [actual literature](https://lelele.io/thesis.pdf). It is also faster.
 
+## Dependencies
+
+To use bliss-rs, you'll need a few packages: a C linker, `ffmpeg`, and the associated
+development packages (libavformat, libavutil, libavcodec, libavfilter, libavdevice),
+as well as the clang development packages. These steps are necessary for e.g. run the
+examples below.
+
+On Ubuntu:
+
+    $ sudo apt install build-essential pkg-config libavutil-dev libavformat-dev \
+    libavfilter-dev libavdevice-dev libclang-dev 
+
+On Archlinux:
+
+    $ sudo pacman -S base-devel clang ffmpeg sqlite
+
 ## Examples
+
 For simple analysis / distance computing, take a look at `examples/distance.rs` and
 `examples/analyze.rs`.
 
@@ -47,6 +64,7 @@ this behavior by using `-a <path>` to store this file in a specific place.
 Ready to use code examples:
 
 ### Compute the distance between two songs
+
 ```
 use bliss_audio::decoder::bliss_ffmpeg::FFmpeg as Decoder;
 use bliss_audio::decoder::Decoder as DecoderTrait;
@@ -62,6 +80,7 @@ fn main() -> Result<(), BlissError> {
 ```
 
 ### Make a playlist from a song
+
 ```
 use bliss_audio::decoder::bliss_ffmpeg::FFmpeg as Decoder;
 use bliss_audio::decoder::Decoder as DecoderTrait;

TODO.md

@@ -10,8 +10,9 @@ ask questions if you want to tackle an item.
 
 ## Actual TODO
 
-- The album playlist should take into account multi-CDs albums
 - Add a list of dependencies / installation guide
+- bliss: A way to learn a metric with a "user survey" on their own libraries using code from the thesis
+  (probably reuse the interactive-playlist in blissify?)
 - Maybe add playlist functions for single songs as convenience methods?
 - Regularly update the python bindings with the new code
 - Check the chroma feature for anomalies (the last 4 numbers look anomalous in most of my cases -
@@ -32,8 +33,6 @@ ask questions if you want to tackle an item.
   (can probably recycle the "album" feature)
 - Use genre clustering (cf already existing code) to find an appropriate M matrix, and put it as alternative
   Hopefully will make playlists not drift
-- bliss: A way to learn a metric with a "user survey" on their own libraries using code from the thesis
-  (probably reuse the interactive-playlist in blissify?)
 - Improve bliss-python somehow / use it in a small demo project maybe?
   A blissify in python?
 
@@ -44,4 +43,4 @@ ask questions if you want to tackle an item.
   - Make the tests that don't need it not dependent of ffmpeg
   - Make the Song::from_path a trait that is by default implemented with the
     ffmpeg feature (so you can theoretically implement the library trait without ffmpeg)
-
+- The album playlist should take into account multi-CDs albums