Web Development

Fedora Commons Ingest Tool

Fedora Commons is an open source repository for digital collections. This tool automates the ingest of digitized items and metadata into a Fedora Commons 4 instance.

All object relationships conform to the Portland Common Data Model.

All PHP code uses the PSR-2 (Coding Style) and PSR-4 (Autoloading) standards. This library was released under the GPLv3 License.

The library is installed via Composer:
$ composer require unhplace/fedora-ingest

Usage


          use UNHPlace\FedoraIngest\FedoraCollection;


          
          // Create a new collection with a named slug

          $col = FedoraCollection('col-1');



          // Set the local path for binaries

          $col->setBinaryPath('data');



          // CSV source

          $col->ingestCsv('csv/metadata.csv');



          // FGDC XML source

          $col->ingestFgdcXml('xml/*.xml');



          // Print the Linked Data URIf

          echo $col->getUri();

CSV keys are Dublin Core metadata elements, e.g., dcterms:identifier, dcterms:title, dcterms:coverage, etc.

The project was funded by the Institute of Museum and Library Services, National Leadership Grants for Libraries Program.

fedora-ingest (github.com)

Drupal & Solr

Screenshot of UNH Digital Collections search results for NH Cities & Towns — The University of new Hampshire Digital Collections web site, c. 2017

The UNH Library Digital Colllections (2017) used the Solr search server to index their scanned documents, full text, and metadata. Drupal has a community-supported module to provide search facets using Solr fields.

This project required developing a Solr schema and appropriately weighting index columns to support keyword search.

A custom Drupal module provided a connection between Fedora Commons and the Drupal theme.

Library Archives

Screenshot of the Find page, containing a set of icons and a variety of collections — Find, 2017

Screenshot of the Digltal Collections page. Each collection has a book cover, vintage photograph, or example map — Digital Collections, 2017

Screenshot of search results page with options to filter by category, type, date, or sort by date — NH Cities & Towns, 2017

Screenshot of The Guide to Amy Cheney Beach Papers, 1835-1956, containing a photo of Beach and collection contents — Guide to Amy Cheney Beach Papers…, 2017

Archival finding aids were formatted in XML then translated to HTML5 using Extensible Stylesheet Language Transformations (XSLT). This was packaged as a Drupal module, which also caches the computationally expensive translation.

CSS Flexbox

Screenshot of a Layout Example site with vegetable-themed sample content.

This modified work sample employs a responsive design using only HTML & CSS, using Flexible Box Layout (Flexbox) instead of external libraries.

Flexbox Layout

Zwitscher-Maschine / Twittering Machine

This experiment used a Markov chain to generate a new tweet based on 100 tweets from Twitter. Users could search for source tweets using keywords and emojis 😎. Written in PHP.

Zwitscher-Maschine / Twittering Machine