Welcome to Franklin’s documentation!¶
franklin
is a set of tools for accessing research articles and
their associated bibtex references.
The following console scripts are available.
fetch-doi
- Retrieve PDFs and bibtex entries.abbreviate-journals
- Parse a bibtex file and abbreviate the journal titles.bibtex-cleanup
- [coming soon] Parse a bibtex file and clean it up.
Installation¶
The easiest way to install franklin is through the Python package index.
$ pip install franklin
For development, a developer installation through pip is recommended.
$ git fetch https://github.com/canismarko/franklin.git
$ pip install franklin/requirements.txt
$ pip install -e franklin/
Command-Line Scripts¶
Fetch DOI¶
The fetch DOI script will retrieve the bibtex entry and, if possible, the PDF of an article.
Retrieving the PDF requires either an open-access article or institutional access (e.g. through VPN). The following publishers are currently supported:
American Chemical Society (ACS)
The Electrochemical Society (ECS)
Elsevier
Springer
The Royal Society of Chemistry (RSC)
To request support for additional publishers, please submit an issue. If possible, include the DOI of an open-access article for testing.
Abbreviate Journals¶
The abbreviate-journals
command line tool can be used to look up
abbreviated journal names from several sources.
First, a dictionary of local journal abbreviations is queried. This
can be skipped using the --no-native
option.
Second, the chemical abstract services source index is
queried. Since CASSI does not provide an API, this involves directly
parsing HTML pages and so is not perfectly reliable. This option is
provided in good faith to comply with the usage guidelines set forth
by CASSI, since the retrieved data are not saved in a directory and
are retrieved one at a time. Please review the CASSI usage guidelines
and refrain from using this option if you do not agree to these
guidelines. This behavior can be disable by providing the
--no-cassi
option.
As a last resort, franklin will attempt to determine the journal
abbreviation directly the the ISSN list of title word abbreviations
(LTWA). This will always produce a result, but is currently very slow
and not well tested. This option can be disabled with the
--no-ltwa
option.