https://github.com/james91b/ida_ipython

An IDA Pro Plugin for embedding an IPython Kernel
https://github.com/james91b/ida_ipython

Last synced: 20 days ago
JSON representation

An IDA Pro Plugin for embedding an IPython Kernel

• Host: GitHub
• URL: https://github.com/james91b/ida_ipython
• Owner: james91b
• License: mit
• Created: 2015-02-15T08:59:11.000Z (almost 10 years ago)
• Default Branch: master
• Last Pushed: 2019-07-13T08:00:28.000Z (over 5 years ago)
• Last Synced: 2024-07-31T23:46:07.157Z (4 months ago)
• Language: Jupyter Notebook
• Size: 1.88 MB
• Stars: 250
• Watchers: 25
• Forks: 46
• Open Issues: 7
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

        
# What's New

- Improve python plugin load order (prevents crashes when python does not load correctly)

- Add in safe IDA process termination (Special thanks to @tmr232 for this)

# What and Why?

This is a plugin to embed an IPython kernel in IDA Pro. The Python ecosystem has amazing libraries (and communities) for scientific computing. IPython itself is great for exploratory data analysis. Using tools such as the IPython notebook make it easy to share code and explanations with rich media. IPython makes using IDAPython and interacting with IDA programmatically really fun and easy.

## Example Uses

## QT Console

You can just use IPython qtconsole for a better interactive python shell for IDA.

![Image of Basic QT Usage](qtbasic.gif)

You can also use the QT console to graph things. This is an example creating a bar chart for the occurrences of each instruction mnemonic in a function (in notepad.exe).

![Image of QT with graph](qtwithgraph.gif)

### Notebooks

Another useful case is using IPython notebooks.

- [Function Entropy](http://nbviewer.ipython.org/github/james91b/ida_ipython/blob/master/notebook/examples/Function%20Entropy.ipynb) - Here is an example where we compute the entropy (using scipy stats module) of each function in notepad.exe and graph the result.

- [Cython and IDA](http://nbviewer.ipython.org/github/james91b/ida_ipython/blob/master/notebook/examples/Cython%20and%20IDA.ipynb) - Here is an example where we use the cython cell magic to call IDA Api's that are not exposed via IDAPython.

- [Sark Snapshots](http://nbviewer.ipython.org/github/james91b/ida_ipython/blob/master/notebook/examples/Sark%20Snapshots.ipynb) - Example of screen snapshots using Sark.

More examples..soon...

# How the plugin works

IDA is predominantly single threaded application, so we cannot safely run the kernel in a separate thread. So instead of using another thread a hook is created on the QT process events function and the `do_one_iteration` method of the ipython kernel is executed each frame.

# Installation

I suggest using the [Anaconda](http://continuum.io/downloads) distribution of Python as it comes with all the required python libraries pre-built and installed. To get IDA to use Anaconda, simply set the PYTHONHOME enviroment variable. Alternatively you can install IPython and the dependencies separately.

This plugin should work on all 6.X x86 QT versions of IDA on Windows.

## Basic Installation and QTConsole

1. Download and extract the [release](https://github.com/james91b/ida_ipython/releases/latest)

2. Copy the contents of the `plugins` and `python` directories under IDA's installation directory.

4. Launch IDA.

5. At the command line (Windows), start an IPython qtconsole with the kernel instance (outputted in the IDA console) e.g `ipython qtconsole --existing kernel-4264.json`

## Using the Notebook

1. Copy `idc` directory to your IDA directory. (the `nothing.idc` script is used to pass command line parameters to the plugin)

2. Change the paths to the `idaq.exe` and `idaq64.exe` executables in the `kernel.json` under the `notebook\kernels\ida32`

    and `notebook\kernels\ida64` directories respectively

3. Install the kernels using `jupyter-kernelspec install` (e.g. `jupyter-kernelspec install --user notebook\kernels\ida64`)

4. When starting a notebook, choose the `IDA32` or `IDA64` kernels, depending on your desired IDA version.

# How to Build

1. Install cmake

2. At the command line cd to the root directory and run the following

3. `mkdir build`

4. `cd build`

5. `cmake -G "Visual Studio 11" -DPYTHON_DIR="" -DIDA_SDK="" -DIDA_DIR="" ..`

e.g.

`cmake -G "Visual Studio 11" -DPYTHON_DIR="C:\Anaconda" -DIDA_SDK="C:\dev\IDA\idasdks\idasdk64" -DIDA_DIR="C:/Program Files (x86)/IDA 6.4" ..`

6. `cmake --build . --config Release`

So far only tested with "Visual Studio 11" compiler.

# Changelog

0.5

- Improve python plugin load order (prevents crashes when python does not load correctly)

- Add in safe IDA process termination (Special thanks to @tmr232 for this)

0.4

- Added IDA 6.9 support

0.3

- 2015-10-04: Running the plugin from the plugins menu or view menu will now launch an IPython QTConsole. Consoles are terminated on exit.

- 2015-10-04: Only capture standard output and error when executing the kernel

- 2015-10-21: Improve error reporting

0.2

- 2015-09-30: Added support for Jupyter (replaces original support for IPython).

- 2015-10-01: Added support for ida64.

0.1

- First release

# To do/Future Ideas

- More examples

- Create a library for cell/line magic functions specific to IDA