{"id":865,"date":"2013-05-04T12:00:17","date_gmt":"2013-05-04T19:00:17","guid":{"rendered":"http:\/\/bellm.org\/blog\/?p=865"},"modified":"2013-05-05T22:43:13","modified_gmt":"2013-05-06T05:43:13","slug":"idl-magics-for-the-ipython-notebook","status":"publish","type":"post","link":"http:\/\/bellm.org\/blog\/2013\/05\/04\/idl-magics-for-the-ipython-notebook\/","title":{"rendered":"IDL Magics for the IPython Notebook"},"content":{"rendered":"

The IPython Notebook<\/a> combines code, documentation, and computational results in one package that’s easy to share<\/a>.? It’s proving a great way to teach, as notebooks are easy for the instructor to write and for students to modify.? Notebooks also provide seamless integration with other programming environments<\/a> through extensions<\/a> providing “magic functions<\/a>“: short code prefixes beginning with %<\/code> or %%<\/code> that call other interpreters, like R or Octave.<\/p>\n

Since many astronomers who might want to transition to Python already know IDL<\/a>, I wanted to provide an %idl<\/code> magic function for IPython so they could easily call existing code.? The difficult interfacing between IDL and Python was already done by Anthony Smith<\/a>‘s pIDLy<\/a> package.? (As a bonus, it supports the free GDL<\/a> interpreter as well.)? I adapted the Octave magic code to provide %idl<\/code> magic functions with consistent syntax.? You can find the code on github as ipython-idlmagic<\/a>.<\/p>\n

The demonstration notebook below is available on github<\/a> as well, or you can view it with nbviewer<\/a>.<\/p>\n

\n
\n

Installation<\/h2>\n<\/div>\n
\n

To begin, we install pIDLy<\/a>:<\/p>\n

pip install pidly\r\n<\/code><\/pre>\n
Then we install idlmagic:<\/p>\n<\/div>\n
\n
\n
In?[?]:<\/div>\n
\n
\n
%<\/span>install_ext<\/span> https<\/span>:<\/span>\/\/<\/span>raw<\/span>.<\/span>github<\/span>.<\/span>com<\/span>\/<\/span>ebellm<\/span>\/<\/span>ipython<\/span>-<\/span>idlmagic<\/span>\/<\/span>master<\/span>\/<\/span>idlmagic<\/span>.<\/span>py<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Usage<\/h2>\n<\/div>\n
\n
When starting a new notebook, we load the magic:<\/p>\n<\/div>\n
\n
\n
In?[1]:<\/div>\n
\n
\n
%<\/span>load_ext<\/span> idlmagic<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
<\/div>\n
\n
IDL not found, using GDL<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
(I am using GDL<\/a> rather than IDL on this computer. idlmagic<\/code> will first look for the idl<\/code> interpreter on the search path and and fall back to gdl<\/code> if needed.)<\/p>\n<\/div>\n
\n
Line magics<\/h3>\n<\/div>\n
\n
The %idl<\/code> magic enables one-line execution of IDL commands in the IPython interpreter or notebook:<\/p>\n<\/div>\n
\n
\n
In?[2]:<\/div>\n
\n
\n
%<\/span>idl<\/span> print<\/span>,<\/span> findgen<\/span>(<\/span>5<\/span>)<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
<\/div>\n
\n
      0.00000      1.00000      2.00000      3.00000      4.00000<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
(Note that the %idl<\/code> line magic fails with TypeError: coercing to Unicode: need string or buffer, dict found<\/code> in current release versions of IPython (0.13.2 and below) due to a known bug<\/a>; the github development version of IPython works as expected.)<\/p>\n<\/div>\n
\n
Cell magics<\/h3>\n<\/div>\n
\n
Multi-line input can be entered with the %%idl<\/code> cell magic:<\/p>\n<\/div>\n
\n
\n
In?[3]:<\/div>\n
\n
\n
%%<\/span>idl<\/span>\r\nx<\/span> =<\/span> findgen<\/span>(<\/span>5<\/span>)<\/span>\r\ny<\/span> =<\/span> x<\/span>^<\/span>2.<\/span>\r\n;<\/span> comments<\/span> are<\/span> supported<\/span>\r\nprint<\/span>,<\/span> $<\/span> ;<\/span> as<\/span> are<\/span> line<\/span> continuations<\/span>\r\nmean<\/span>(<\/span>y<\/span>)<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
<\/div>\n
\n
% Compiled module: MEAN.\r\n      6.00000<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Passing variables between Python and IDL<\/h3>\n<\/div>\n
\n
The mechanisms for passing variables to and from IDL are based on those in the built-in %R<\/code> and %octave<\/code> magics.<\/p>\n
Variables may be pushed from Python into IDL with %idl_push<\/code>:<\/p>\n<\/div>\n
\n
\n
In?[4]:<\/div>\n
\n
\n
msg<\/span> =<\/span> '  padded   string   '<\/span>\r\nimport<\/span> numpy<\/span> as<\/span> np<\/span>\r\narr<\/span> =<\/span> np<\/span>.<\/span>arange<\/span>(<\/span>5<\/span>)<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[5]:<\/div>\n
\n
\n
%<\/span>idl_push<\/span> msg<\/span> arr<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[6]:<\/div>\n
\n
\n
%%<\/span>idl<\/span>\r\nprint<\/span>,<\/span> strcompress<\/span>(<\/span>msg<\/span>,<\/span>\/<\/span>REMOVE_ALL<\/span>)<\/span>\r\nprint<\/span>,<\/span> reverse<\/span>(<\/span>arr<\/span>)<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
<\/div>\n
\n
paddedstring\r\n              4                     3                     2\r\n              1                     0<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Similarly, variables can be pulled from IDL back to Python with %idl_pull<\/code>:<\/p>\n<\/div>\n
\n
\n
In?[7]:<\/div>\n
\n
\n
%<\/span>idl<\/span> arr<\/span> +=<\/span> 1<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[8]:<\/div>\n
\n
\n
%<\/span>idl_pull<\/span> arr<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[9]:<\/div>\n
\n
\n
arr<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
Out[9]:<\/div>\n
\n
array([1, 2, 3, 4, 5])<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Variables can also be pushed and pulled from IDL inline using the -i<\/code> (or --input<\/code>) and -o<\/code> (or --output<\/code>) flags:<\/p>\n<\/div>\n
\n
\n
In?[10]:<\/div>\n
\n
\n
Z<\/span> =<\/span> np<\/span>.<\/span>array<\/span>([<\/span>1<\/span>,<\/span> 4<\/span>,<\/span> 5<\/span>,<\/span> 10<\/span>])<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[11]:<\/div>\n
\n
\n
%<\/span>idl<\/span> -<\/span>i<\/span> Z<\/span> -<\/span>o<\/span> W<\/span> W<\/span> =<\/span> sqrt<\/span>(<\/span>Z<\/span>)<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
In?[12]:<\/div>\n
\n
\n
W<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
Out[12]:<\/div>\n
\n
array([ 1.        ,  2.        ,  2.23606801,  3.1622777 ])<\/pre>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Plotting<\/h3>\n<\/div>\n
\n
Inline plots are displayed automatically by the IPython notebook. IDL Direct graphics are used. The optional -s width,height<\/code> argument (or --size<\/code>; default: 600,375<\/code>) specifies the size of the resulting png image.<\/p>\n<\/div>\n
\n
\n
In?[13]:<\/div>\n
\n
\n
%%<\/span>idl<\/span> -<\/span>s<\/span> 400<\/span>,<\/span>400<\/span>\r\nplot<\/span>,<\/span>findgen<\/span>(<\/span>10<\/span>),<\/span>xtitle<\/span>=<\/span>'X'<\/span>,<\/span>ytitle<\/span>=<\/span>'Y'<\/span><\/pre>\n<\/div>\n<\/div>\n<\/div>\n
\n
\n
\n
<\/div>\n
\n
% Compiled module: WRITE_PNG.<\/pre>\n<\/div>\n<\/div>\n
\n
<\/div>\n
\"\"<\/div>\n<\/div>\n<\/div>\n<\/div>\n<\/div>\n
\n
Known issues and limitations<\/h2>\n<\/div>\n
\n