The following is a summary of changes and improvements to
namedropper. New features in each version should be listed, with
any necessary information about installation or upgrade notes.
A user can run a script to get summary information about the number of
tagged names in an EAD document, in order to do simple comparison of
tagged and untagged documents.
Updates to lookup-names script
When a user runs the lookup-names script to generate a CSV file, the resulting output
includes resource type for person, place, or organization so that results can be
filtered and organized by broad types.
When users interrupts the lookup-names script while it is running, it stops
processing gracefully and reports on what was done so that user can get an idea
of the output without waiting for the script to complete on a long document.
When a user runs the lookup names script with options that generate no results,
the script does not create a CSV file or an enhanced xml file (even if those options
were specified) and prints a message explaining why, so that the user is not confused
by empty or unchanged files.
When users run the lookup-names script to generate annotated XML, they can optionally
add tags with Oxygen history tracking comments so that changes can be reviewed and
accepted or rejected in Oxygen.
Bug fix: When a user runs a lookup-names script on an XML file that does not have
all of its component parts, it should not crash.
Bug fix: When annotating XML, the script will no longer crash if –types is not restricted
to Person,Place,Organisation (or some subset of those three), and will warn about
recognized entities that cannot be inserted into the output XML.
Bug fix: When annotating XML, tags will not be inserted where they are not schema valid
(schema validation currently only supported for EAD).
Bug fix: If output XML is requested but an HTTP Proxy is not configured, the script will halt and
information about setting a proxy, instead of crashing when attempting to validate the output XML.
A command-line user running the lookup-names script can have the input
document type auto-detected, so they don’t have to specify an input type
every time they use the script.
A command line user can run a script to look up recognized person names from
a TEI or EAD XML document in a name authority system so that recognized
names can be linked to other data.
A command line user can run a script to generate a new version of an EAD XML
document with tagged named entities, in order to automatically link
mentioned entities to other data sources.
A command line user can run a script to generate a new version of a TEI XML
document with tagged named entities, in order to automatically link
mentioned entities to other data sources.
A command line user can optionally export identified resources and
associated data to a CSV file, so they can review the results in more
detail.
A command line user can run a script to output recognized names in an EAD
XML document in order to evaluate automated name recognition and
disambiguation.
A command line user can run a script to output recognized names in a TEI XML
document in order to evaluate automated name recognition and disambiguation.