brycepg/fto-scraper

web scraping to get interval statistics for a game

fto-scraper's Issues

With 1 data point, fto-graph.py issues warning and produces empty graph

For example, with fto.csv having these contents:

Date,Population,Birth Queue,Pregnant Mothers
06/28/16-03,297,318,6

This is the console output:

$ ./fto-graph.py fto.csv
/usr/lib/python2.7/dist-packages/matplotlib/axes/_base.py:2782: UserWarning: Attempting to set identical left==right results
in singular transformations; automatically expanding.
left=736143.125, right=736143.125
  'left=%s, right=%s') % (left, right))

And this is output.png:

Other relevant information:

$ python --version
Python 2.7.10rc1
$ python -c 'print __import__("matplotlib").__version__'
1.5.2rc2
$ python -c 'print __import__("pandas").__version__'
0.18.0+git114-g6c692ae

The expected behaviour would be that it produces a graph with a single data point, or at least does not write a warning to the console. I realise this isn't a very important issue, but I thought it would be good to note it.

Color Y-axis titles instead of legend

Raw pregnant mothers offset by +2 since late February 2017.

fto_graph.py automatically adjusts for a bug where the reported number of pregnant mothers is 1 greater than the true value. I've been told that this is because a mother was killed while pregnant.

It recently came to my attention that another pregnant mother was killed between 2017-02-24 (the last time that the population rose while the number of pregnant mothers was 2) and 2017-02-28 (when the mother's murderer was reported to have been killed). After this, it can be seen that the number of pregnant mothers reported by the game has never fallen below 2.

This means that the number of pregnant mothers being graphed is now incorrect. This could be fixed simply by applying a piecewise constant offset to the data, based on the date of collection, but I'm not sure if this is the best solution.

Running the script fails with network time outs

error_output.txt