lolologist 0.5.5

A utility that automatically generates an image macro from your webcam whenever you commit to a git repository.

lolologist
==========
[](https://travis-ci.org/arusahni/lolologist)

A python-based tool that automatically generates an image macro from your webcam whenever you commit to a git repository. Inspired by [lolcommits](https://github.com/mroth/lolcommits).

Installing
------------

### Linux

*BEFORE* installing lolologist, make sure that the following packages are installed locally:

* python-dev
* libfreetype6-dev
* libjpeg-dev
* mplayer
* pandoc (if doing local development)

If pillow is already installed and you do not have the packages above, you need to uninstall it and install those packages. This is because when pillow is installed, it compiles optional features based on the availability of those packages. You would do this with commands:

```console
sudo pip uninstall pillow
sudo apt-get install python-dev libfreetype6-dev libjpeg-dev

# Optional, for a better font
sudo apt-get install ttf-mscorefonts-installer
```

To install lolologist, after confirming that the above prerequisites are installed, run `sudo python setup.py install`, or install via pip.

### OS X

*BEFORE* installing lolologist, make sure that the following Homebrew packages (or their equivalents) are installed locally.

* freetype
* jpeg
* imagesnap

Also, allow the locate daemon to run. This is necessary for font detection.

```console
sudo launchctl load -w /System/Library/LaunchDaemons/com.apple.locate.plist
```

To install lolologist, after confirming that the above prerequisites are installed, run `sudo python setup.py install`, or install via pip.

Using
-----
1. Within the root of your git repository, type `lolologist register`. This should add a githook that will trigger the program every time you commit.
2. Commit!

The path to your photo will be printed in the commit output. The path is configurable - see the `Output*` options in the configuration section below.

Fonts
-----
Due to licensing concerns, I can't distribute lolologist with the iconic Impact TrueType font. To account for this, lolologist will use your system's Impact if it exists, or fall back to an open font.

If Impact isn't installed on your system, [download and install it](http://www.fontpalace.com/font-details/Impact), and then run either `lolologist setfont` or `lolologist setfont <path-to-font>` to load it.

Configuration
-------------
The utility can be configured through the `.lolologistrc` file, usually found in your home directory. If the file doesn't exist, feel free to create it. The following fields are accepted:

| Field | Description |
| ----------------- | -------------------------------------------------------------------------- |
| `Camera` | The video device to use. (e.g. for Linux: `/dev/video1`, for OS X: `iSight`) |
| `FontPath` | The full path to the Impact font's TTF file |
| `Lolspeak` | `on` if commit messages should be translated to lolspeak, `off` otherwise |
| `OutputDirectory` | The format string for the directory into which all images will be placed |
| `OutputFilename` | The format string for the name of the generated file |
| `OutputFormat` | The type of image to generate (e.g. `jpg`) |
| `UploadImages` | `on` if macros should be uploaded to the internet, `off` otherwise |
| `UploadUrl` | The URL to post the generated image macro to |

Pythonic format strings are accepted for the outpute file name, with the caveat that *percent signs have to be escaped with another percent sign*.

For example, if you wanted to group images by the commit year and month, you could use the following:

```ini
OutputDirectory = ~/.lolologist/{project}/{time:%%Y}/{time:%%m}
```

Possible format parameters:

| Parameter | Type | Description |
| ---------- | :--------: | ------------------------------------------ |
| `project` | *string* | The name of the git repository's directory |
| `revision` | *string* | A ten character ref sha |
| `message` | *string* | The entire commit message. |
| `time` | *datetime* | The time of the commit. |

=======

Acknowledgements
----------------
Thanks go out to Stephen Newey for his [PyLOLz module](https://code.google.com/p/pylolz/).