Name | Last commit date | Author |
---|---|---|
master | 2020-04-19 01:13:36 +0100 | mitxela |
Name | Size |
---|---|
README.md | 1.032K |
playback.py | 1.677K |
process.py | 825B |
requirements.txt | 54B |
Reads through a git repo and turns each commit into a series of vim keystrokes. The result is a pseudo-timelapse which almost looks like it's being typed out by a real person. Results vary depending on how frequent the commits were. It's also still only doing linewise diffs, so even a single byte change means the whole line is re-typed.
Only works on a single file. It is a good idea to rebase any merge commits so that the history reads as if it was written by a single author.
pip install -r requirements.txt
process.py
reads the git history and outputs the patches as a list of tuples representing changes. If the repo contains multiple files, specify a filename to filter as an argument. E.g. ./process.py gbjs.htm > patches.txt
Multiple filenames can be provided if the file was renamed.
playback.py
reads the patches file and turns it into keystrokes. Run the script and then focus on a new terminal window. Hold esc to abort.
These scripts were used to produce this video.