Publications of the Journal of Open Source Software (JOSS1) are
equally available as .html
and .pdf
, altogether with their
corresponding .xml
files, on a separate public repository2.
This not only offers an alternative access to the publications at
once, equally, it is possible to retrieve their bibliographic data
rapidly for further processing in literature reference programs like
zotero3
The uncompressed .zip
archive is cloned onto a machine running
Python3
. The branch was chosen both because of the anticipated
retirement of Python2
, as well for the better support of Unicode
characters. Both scripts run from the CLI, and expect the .zip
archive in the same folder, e.g.
python bibtool_Joss_pocket.py joss-papers-master.zip
to retrieve the necessary raw data, which is either the .html
or .crossfile.xml
– but not the substantial larger .pdf
.
The smaller script, joss-papers-master.zip
, retrieves the small
«cite as» block provided on each .html
of the corresponding
publication, in either .txt
or .csv
format. The larger
bibtool_Joss_bibtex.py
attempts to retrieve basic bibliographic
information and creation of a bibtex file (.bib
) understood by
many literature reference programs. In addition, a less verbose
.csv
(separator: semicolon) is created, offering futher processing
with any editor, or spreadsheet.
After running these scripts, the scripts leave you with the original
.zip
archive (still containing the .html
and .pdf
about the
publications) and the extracted literature data. Intermediate files
are deleted automatically.
(c) 2019 Norwid Behrnd, GPLv3.