2. Fill out templates

For your collection, create spreadsheets for the following types of entities according to our template guidelines. These templates assume that you will create the main resource for the collection manually.

Available CSV templates¶

Template limitations¶

Not all metadata fields available in template.¶

While we tried to have the CSVs templates include as many metadata fields as possible, there are many we were not able to add at this time. If a field isn't in template, you can only add it by changing both the templates and the relevant scripts.

Some metadata fields are not repeatable in the template¶

Some metadata fields are not repeatable in the template, even if ArchivesSpace allows repeatability. For some paired fields, which only make sense together, like authority_id and source_id -- we have chosen not to make repeatable to make the template easier to use. We understand this isn't ideal and may change it in the future.

Formatting and encoding in CSVs¶

Encoding¶

If given the option by your spreadsheet editor, always encode your CSV as UTF-8.
If possible, format all of your cells as Plain Text or Text and quote all text before saving/exporting as a CSV. This will prevent automatic formatting in most spreadsheet editors.

Diacritics¶

Do not simply enter diacritics and other special characters into your spreadsheets, as these cannot be properly read by ArchivesSpace for some horrible reason. Instead, use HTML character codes, which can be found online and have the pattern $#[number].

Column rules¶

Columns can be in any order in the template.
Unless required, unused columns do not need to be included in your template. However, keeping unused columns will not negatively impact the scripts.
Column headings must be spelled exactly as prescribed in the template with the same capitalization.

Row rules¶

Rows can be in any order in the template.
Delete all blank rows before ingest.

Data Types¶

The data types given in the CSV template indicate what type of value is needed by the Python scripts to successfully ingest new records into ArchivesSpace. It does not, however, always enforce ArchivesSpace rules, formatting, or best practices. For instance, if best practices determine that URL is always desired in a specific field, and you enter blah mondays are the worst instead, the script will mostly likely still work, creating a new record with the value of blah mondays are the worst in that field.

String¶

Any characters can be entered in this cell, including numbers, letters, and symbols. For instance, a URL, barcode, or a paragraph of text would be accepted as string values.

Examples:

http://id.worldcat.org/fast/834963
41151034444400
Digital photographs of blue jay

Repeatable string fields are separated by a single pipe |.

Integers¶

Any positive whole numbers, like 1, 2, 3, 4, etc.

Examples:

4
14000
11

Controlled List¶

Only values from the specified controlled list can be used. To find see the accepted values, either navigate to System → Controlled Value lists in the staff interface of ArchivesSpace or view your enumerations.csv -- the name of the Controlled List is in column Enumeration and the accepted values are in column Value code.

Enumeration code	Enumeration	Value code	Value	Position
subject_source	Subject Source	aat	Art & Architecture Thesaurus	0
subject_source	Subject Source	fast	fast	1
subject_source	Subject Source	gmgpc	TGM II, Genre and physical characteristic terms	2
subject_source	Subject Source	lcsh	Library of Congress Subject Headings	3
subject_source	Subject Source	local	Local sources	4
subject_source	Subject Source	mesh	Medical Subject Headings	5

For instance, if the Controlled List given for a column is Subject Source, acceptable values would include:

aat
fast
local

Values like the following, containing typos or different capitalization, will not work properly.

FAST
Medical Subject Headings
gmgpc
Aat

Boolean¶

The only accepted values in these fields are TRUE or FALSE.

Ref¶

For ingest, the only accepted values in these fields are URIs (aka Refs) from your local ArchivesSpace instance. Do not include the "base" component of the URI.

Examples (after replacement in step 7 & 9):

/agents/corporate_entities/879
/subjects/339|/subjects/142
/repositories/3/archival_objects/84724
instance_type==mixed_materials;;ref==/repositories/3/top_containers/14557;;type_2==folder

If the URI has not yet been created, please type the exact name of the subject or agent instead. For top containers in archival object instances, use the barcode in the ref field if the top container does not exist yet. The names and barcodes will be replaced by their URIs in step 7 and 9.

Examples (before replacement in step 7 & 8): - Johns Hopkins University. Academic Council of Homewood Faculties - Academic libraries|ephemera (general object genre) - Departmental Review Committee, 1972-1973 - instance_type==mixed_materials;;ref==31151034438071;;type_2==folder

Repeatable ref fields are separated by a single pipe |. Be sure there are no spaces surrounding the pipe, this will break the script.

Subfield type¶

This data type is specific to the Big Blue Jay workflow templates and helps us represent fields with multiple subfields. Each subfield is linked to its value by "==". For instance, the linked_agents field has three subfields: role, relator, and ref. This means a corporate entity with the role of creator and the relator would be formatted as followed:

role==creator;;relator==pht;;ref==/agents/corporate_entities/388

If the field needs to be repeated, each field is separated by double pipes ||. For instance:

role==creator;;relator==pht;;ref==/agents/corporate_entities/388||role==creator;;relator==edt;;ref==/agents/corporate_entities/698

The subfields can be in any order in the spreadsheet. Subfields that are not required do not need to be included in spreadsheet.