As a repository end user, I want to be alerted to matches in full-text in my search results, so that I can find material that matches my interests more quickly
As a repository administrator, I want a user interface to manage full-text indexing of specific works, so that I can ensure only materials with OCR data undergo the additional processing
Use one or more of the following options to provide acceptance criteria.
Notes
This epic follows work done previously to research and prototype options for enabling full text works. This epic seeks to actually implement a solution. The scope of this epic would include enhancements to the full text indexing process in Curate as well as enabling search and display of full text matches within Lux (https://digital.library.emory.edu). Searching within an individual work in the Universal Viewer will be a separate effort.
As background, our digitized text material which has already had OCR processing consists of 2 main types of outputs, "Kirtas" and "LIMB". Kirtas is the legacy output, and some examples include the Yellowbacks. LIMB is the newer, current output and includes collections such as the Yearbooks.
Both will always contain:
a PDF of the entire volume
a plain text .txt for each page containing textual content (File Use: "Transcript File")
Kirtas volumes also contain:
An OCR xml file for the entire volume
LIMB volumes also contain:
An OCR xml file (either ALTO or ABYY) for each page containing textual content (File Use: "Extracted")
[x] A UI component is available to Curate admin users so they can run the reindexing process on selected works/IDs only
[x] The existing display implemented in Curate search results is added to Lux. Please revise the existing search results label "Keyword matches" to "Full-text matches".
[x] The highlighting of matching full text search terms in search results currently implemented for Curate should also display in Lux.
[x] The new full text SOLR field should be available in the Common Fields and All Fields search options in Lux.
Optional if time permits
[ ] The full text indexing process can be added as part of current bulk-import process for new works as they are ingested
Given/When/Then
Given (some context) and (some other optional context)
When (some action is carried out)
Then (a set of observable consequences should occur)
Story
As a repository end user, I want to be alerted to matches in full-text in my search results, so that I can find material that matches my interests more quickly
As a repository administrator, I want a user interface to manage full-text indexing of specific works, so that I can ensure only materials with OCR data undergo the additional processing
See also the enhancement request from the Rose Library.
Acceptance Criteria
Use one or more of the following options to provide acceptance criteria.
Notes
This epic follows work done previously to research and prototype options for enabling full text works. This epic seeks to actually implement a solution. The scope of this epic would include enhancements to the full text indexing process in Curate as well as enabling search and display of full text matches within Lux (https://digital.library.emory.edu). Searching within an individual work in the Universal Viewer will be a separate effort.
As background, our digitized text material which has already had OCR processing consists of 2 main types of outputs, "Kirtas" and "LIMB". Kirtas is the legacy output, and some examples include the Yellowbacks. LIMB is the newer, current output and includes collections such as the Yearbooks.
Both will always contain:
Kirtas volumes also contain:
LIMB volumes also contain:
Links to Additional Information
Add links here
Checklist
Optional if time permits
Given/When/Then