#56 clearer README, auto-install dependencies, build numpy from source in Docker

CHANGELOG.md

@@ -7,7 +7,8 @@
 - Unit tests
 
 ### CHANGED
-- Made the interface for most metadata public.
+- Switched to setuptools to automatically install missing dependencies
+- Made the interface for most metadata public
 - Refactored some poorly-named things
 
 ## [1.0.0] - 2015-05-23

Dockerfile

@@ -3,14 +3,17 @@ MAINTAINER Jim Rybarski <jim@rybarski.com>
 
 RUN mkdir -p /var/nds2
 RUN apt-get update && apt-get install -y --no-install-recommends \
-    python-numpy \
-    python3-numpy \
+    build-essential \
+    libatlas3-base \
+    liblapack-dev \
+    libblas-dev \
+    python \
+    python3 \
+    python-dev \
+    python3-dev \
     python-pip \
     python3-pip
 
-RUN pip install six
-RUN pip3 install six
-
 COPY . /opt/nd2reader
 WORKDIR /opt/nd2reader
 RUN python setup.py install

Makefile

@@ -1,4 +1,4 @@
-.PHONY: build shell
+.PHONY: build py2shell py3shell test
 
 build:
 	docker build -t jimrybarski/nd2reader .

README.md

@@ -10,16 +10,19 @@
 
 ### Installation
 
-Just use pip (`numpy` is required):
+Dependencies will automatically be installed if you don't have them. That said, for optimal performance, you should
+install the following packages before installing nd2reader:
 
-`pip install numpy nd2reader`
+#### Ubuntu
+`apt-get install python-numpy python-six` (Python 2.x)
+`apt-get install python3-numpy python3-six` (Python 3.x)
 
-If you want to install via git, clone the repo and run:
+#### Other operating systems
+These have not been tested yet.
 
-```
-pip install numpy
-python setup.py install
-```
+nd2reader is compatible with both Python 2.x and 3.x. I recommend installing using pip:
+
+`pip install nd2reader` (Python 2.x) or `pip3 install nd2reader` (Python 3.x)
 
 ### ND2s
 
@@ -50,7 +53,7 @@ You can also get some metadata about the nd2 programatically:
 
 ### Images
 
-`nd2reader` will always return an `Image` object, which contains some metadata about the image as well as the
+Every method returns an `Image` object, which contains some metadata about the image as well as the
 raw pixel data itself. Images are always a 16-bit grayscale image. The `data` attribute holds the numpy array
 with the image data:
 
@@ -162,6 +165,10 @@ You can also see how many images are in your image set:
 7
 ```
 
+### Protips
+
+nd2reader is about 14 times faster under Python 3.4 compared to Python 2.7. If you know why, please get in touch!
+
 ### Bug Reports and Features
 
 If this fails to work exactly as expected, please open a Github issue. If you get an unhandled exception, please

setup.py

@@ -1,10 +1,14 @@
-from distutils.core import setup
+from setuptools import setup
 
 VERSION = "1.1.0"
 
 setup(
     name="nd2reader",
     packages=['nd2reader', 'nd2reader.model'],
+    install_requires=[
+        'numpy>=1.6.2, <2.0',
+        'six>=1.4, <2.0'
+    ],
     version=VERSION,
     description='A tool for reading ND2 files produced by NIS Elements',
     author='Jim Rybarski',