This project implements a text writing progress tracker based on existing git commit messages. Results include textual and visualised information about daily average, remarkable changes and progress. It will also provide you with some motivational comments to keep going.
It analyses all git log messages from a given file that contain a keyword followed by a number with a space in between. The default use case is to track writing progress in fractions of pages with the number representing the current page at the time of the commit. Those keywords must of course be manually introduced to the commits.
So far keywords can only be changed directly in (line 9 of) the source code.
An example when using the default keywords ('S.', 'Seite') and calling the program from a git repository with a git log containing 167 commits that include the keyword number pattern using:
python ../gitalyse/gitalyse.py filename -r -t 300 -n 5
results in the following output:
Detected 167 commits, failed to detect 46 commits.
Between:
9dc2bc4 2017-10-16T18:20:19+02:00 filename: S. 303
77b76ee 2016-10-01T20:05:05+02:00 filename S. 6
380 days total passed while writing 297 pages
That is an average of 0.781578947368421 pages per day
For the last 5 commits:
9dc2bc4 2017-10-16T18:20:19+02:00 filename: S. 303
b03af46 2017-10-16T01:13:52+02:00 filename: S. 295
5a021af 2017-10-15T21:39:39+02:00 filename: S. 294
ce5ffae 2017-10-15T17:46:49+02:00 filename: S. 291
9fba437 2017-10-14T17:14:39+02:00 filename: S. 291
3 days total passed while writing 12 pages
That is an average of 4.0 pages per day
- great, keep going! :D
and the following visualisation:
Using the help by calling python gitalyser.py -h is highly
recommended to explore additional features.
- python >= 3.6 (not tested for lower versions)
- python-dateutil
- python-matplotlib (optional)
- git (and commits with the specified keyword number patterns)
Limiting the amount of characters on a page to a fixed number (e.g. 1800), setting the font size to a fixed value and defining a fixed font is highly recommended. Primarily the first as otherwise pages may vary (e.g. due to software updates) and become infeasible for analysis. Changing the keyword and not relying on pages would be an alternative but has not been a use case yet.
This project is licensed under the MIT License.