a collection of experimental/one-off projects conducted by BU Librarians and friends.
important note: code has been preserved as is, and is not maintained or expected to be completely runnable
playhouse
is inhabited by Boston University Librarians and friends interested in
using python, javascript, php, and other languages to develop simple scripts and
complex applications to support library work.
this repository contains various projects of interest, often involving consuming a number of apis, extracting and enriching data from a given source, and using one or more libraries or digital humanities techniques to draw additional value or information from it.
these projects are not actively maintained or consumed by the library, but are preserved as technical/inspirational examples, proofs of concept, and potential placeholders for future development/exploration.
filename | project description |
---|---|
doi-get-bibtex-from-title.ipynb |
obtain a bibtex record for a given doi (doi.org) after getting its doi from crossref |
loc-harverst-modify-newspapers.ipynb |
harvest and modify bibliographic records of digitized newspapers from LoC |
loc-visualize-article-sentiment.ipynb |
analyze and visualize polarity data of LoC articles from 'Chronicling America' collection. |
nyt-article-congress-api.ipynb |
interact with NYT articlesearch and politics APIs |
nsf-query-boston-awards.ipynb |
interrogate NSF awards API for those given to Boston academics |
oclc-find-related-issns.ipynb |
populate list of ISSNs related to given marcxml files, via OCLC xISSN API |
osf-find-by-related-orcid.ipynb |
generate a list of records that were created by selected orcid researchers |
the majority of these projects were generated within/using
jupyter notebooks
,
which allow python code to be documented/described, run, and displayed in place.
jupyter notebook
is the core requirement for running and interacting with these
projects. it is included in the greater conda
package/environment/dependency manager, and can be installed alongside the
conda install, or
done separately.
in order to best view/edit these projects, it's recommended that you run the following command in the terminal at the base of this project's directory:
$ jupyter notebook /playhouse/src/
this should take over the command window with 'NotebookApp' logs, opening the browser automatically to http://localhost:8888/tree where you should see an interactive display of the files included in the directory.
to view (and/or edit) a notebook, simply select it from the list and it should open in a new tab. see the jupyter docs for more usage instructions.
note: untested legacy docs moved from install_a_working_python_3.5_environment.md
The following instructions will guide you through creating a fresh environment into which to install python 3.5, anaconda, numpy, pandas, matplotlib, and seaborn along with several modules that are commonly used at BU Libraries: pymarc, marcx, textblob, and gensim. The commands below should accomplish that.
This assumes you have installed anaconda.
$ source deactivate
$ conda create --name py3x python=3.5 anaconda numpy pandas matplotlib seaborn
$ source activate py3x
$ pip install https://github.com/edsu/pymarc/archive/python3.zip
$ pip install https://github.com/miku/marcx/archive/master.zip
$ pip install textblob
$ conda install gensim