Add a Vagrantfile

Author: Matthew Peters

.gitignore

@@ -21,3 +21,5 @@ output/*
 # vim temp files
 *.swp
 
+.vagrant
+

.travis.yml

@@ -5,15 +5,7 @@ script: make test
 cache:
   - apt
   - pip
-install:
-  - sudo apt-get -y install libatlas-base-dev libatlas-dev lib{blas,lapack}-dev
-  - sudo apt-get -y install libxslt-dev libxml2-dev
-  # some conda magic to install numpy, scipy
-  # see http://sburns.org/2014/03/28/faster-travis-builds.html
-  - sudo pip install conda
-  - conda_deps='pip numpy scipy'
-  - conda create -p $HOME/py --yes $conda_deps "python=$TRAVIS_PYTHON_VERSION"
+install: ./provision.sh
+script:
   - export PATH=$HOME/py/bin:$PATH
-  - pip install Cython>=0.21.1
-  - pip install -r requirements.txt
-  - python setup.py build_ext --inplace
+  - make test

README.md

@@ -58,21 +58,45 @@ Otherwise, we try to guess the encoding from a `meta` tag or specified
 
 ## Installing
 
-The build requires numpy, lxml and a new version of Cython, so first make sure
-they are installed, then install Dragnet:
-
-```
-pip install numpy
-pip install --upgrade cython
-pip install lxml
-pip install dragnet
-```
-
 Dragnet is written in Python (developed with 2.7, not tested on 3)
 and built on the numpy/scipy/Cython numerical computing environment.
 In addition we use <a href="http://lxml.de/">lxml</a> (libxml2)
 for HTML parsing.
 
+We recommend installing from the master branch to ensure you have the latest
+version.
+
+### Installing vagrant:
+
+1. Install [vagrant](https://www.vagrantup.com/downloads.html).
+2. Install [Virtual Box](https://www.virtualbox.org/wiki/Downloads).
+3. Clone the master branch: `git clone [email protected]:seomoz/dragnet.git`
+4. Bring up the vagrant box: `vagrant up --provider=virtualbox`
+5. Log into the vagrant box:
+
+```bash
+vagrant ssh
+# inside the vagrant vm
+$ cd /vagrant
+# these should now pass
+$ make test
+```
+
+### Installing without vagrant
+
+1.  Install numpy, a new version of Cython and [lxml](http://lxml.de/installation.html).  You can install numpy and cython with pip, but lxml may require some
+additional packages (libxml2).
+2.  Clone the master branch: `git clone [email protected]:seomoz/dragnet.git`
+3.  Install the requirements: `sudo pip install -r dragnet/requirements.txt`
+4.  Build dragnet
+
+```bash
+$ cd dragnet
+$ sudo make install
+# these should now pass
+$ make test
+```
+
 # Contributing
 
 We love contributions!  We are especially looking for someone who would

Vagrantfile

@@ -0,0 +1,16 @@
+# -*- mode: ruby -*-
+# vi: set ft=ruby :
+
+Vagrant.configure(2) do |config|
+  config.vm.box = "ubuntu/trusty64"
+
+    # lxml has trouble building if the amount of memory is 512:
+    # http://stackoverflow.com/questions/16149613/installing-lxml-with-pip-in-virtualenv-ubuntu-12-10-error-command-gcc-failed
+  config.vm.provider "virtualbox" do |vb|
+    vb.memory = "1024"
+  end
+
+   config.vm.provision "shell", privileged: false, inline: <<-SHELL
+     cd /vagrant; ./provision.sh
+   SHELL
+end

provision.sh

@@ -0,0 +1,26 @@
+
+# die on error
+set -e
+
+sudo apt-get update
+
+sudo apt-get -y install libatlas-base-dev libatlas-dev lib{blas,lapack}-dev
+sudo apt-get -y install libxslt-dev libxml2-dev gcc g++
+
+# According to `conda list`, pip installing conda isn't the way to go. Instead, use this:
+curl -O http://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh
+bash Miniconda3-latest-Linux-x86_64.sh -b
+rm Miniconda3-latest-Linux-x86_64.sh
+export PATH=$HOME/miniconda3/bin:$PATH
+conda_deps='pip numpy scipy'
+conda create -m -p $HOME/py --yes $conda_deps python=2.7
+export PATH=$HOME/py/bin:$PATH
+
+# configure conda for future login
+echo "export PATH=$PATH" >> $HOME/.bashrc
+
+pip install "Cython>=0.21.1"
+pip install -r requirements.txt
+
+make install
+