covid-19-canvon
Make plots from publically available data about covid-19 (Corona Virus Disease started 2019).
Makefile/Perl/gnuplot CSV-to-PNG solution by Fabian "canvon" Pietsch.
Also builds slide.html files (via Perl Template::Toolkit "TT2")
which are HTML/CSS/JavaScript using the Bootstrap & jQuery libraries.
git clone from https://github.com/canvon/covid-19-canvon,
then fill with data & plots via make invocations,
see below (overview).
Plot runs are sometimes uploaded to some sub-directory of
https://files.canvon.de/covid-19-canvon/plots/; there,
the slide.html (mentioned above) can be browsed/linked
for easy swipingclicking through all slides (of a plot run) --
especially on smartphone/touch devices like Android.
An example is linked from the "rendering" badge above.
Overview
To get data for a data source Foo and make some plots,
do the following (on a Linux machine, e.g., Debian 9,
with GNU Make, Perl (+ Template::Toolkit "TT2") &
gnuplot installed):
(For valid values of Foo,
see Supported data sources.)
covid-19-canvon$ cd Foo
covid-19-canvon/Foo$ make update
[...]
covid-19-canvon/Foo$ make update # A second time if it failed expectedly.
[...]
covid-19-canvon/Foo$ make plot-examples
[...]
This should get the source data and make some plots. See the in-terminal output for where things will have been created, or follow the previous sentence's links for details.
Get source data
As mentioned above (overview), choose
a supported data source,
enter its sub-directory and run make update (possibly two times).
This should download the source .csv file, or clone the
git repository containing the source data files with history,
as appropriate.
You would also run make update again every day (or so), to update to
the newest data.
In the data source's sub-directory, you should then have data to plot,
usually converted to a file full_data_for_gnuplot.csv; every such
generated file will also be held as a copy in archive/generated_data/
(including a time stamp, e.g., full_data_for_gnuplot-20200331_135019.csv
for a file generated 31st March, 2020, at 13:50:19 (PM)).
The original source data will also be archived,
either in archive/source_data/ or as part of a (e.g., git) repository
cloned to get the source data (so it'll already track its own history).
Make plots
You could do your own plotting, now, using the CSV (Comma-Separated Values)
file(s) from a data source, or the pre-processed CSV for use with gnuplot.
There are ready-made gnuplot scripts, though, with a build system integration,
in a data source's sub-directory's examples/ sub-directory. They are supposed
to be gnuplot-loaded or called from the Makefile directory.
As mentioned above (overview), choose
a supported data source,
enter its sub-directory and run: make plot-examples
This should plot the previously prepared data
from full_data_for_gnuplot.csv into several output files,
in a per-plot-run archive sub-directory.
In the data source's sub-directory, you should then have
the plot run's output .png image and slide.html web page
files in archive/plots/YYYYMMDD_HHMMSS/,
e.g., JohnsHopkinsUniversity/archive/plots/20200331_135027/,
ready-to-upload as a bulk directory to your own web-space
(or for simple local browsing).