CSV to EAD instructions

Requirements

• Python 3.x (2.7 support removed due to the addition of tkinter for error messages)

Use

Download or clone the script.

At the command line: python csv_to_ead.py followed by the filename(s) of the csv files you wish to convert (multiple filenames can be provided, separated by a space, and processed at once)

The script expects a UTF-8 encoded csv file and will validate the headers before it processes the contents. The only required fields are level_type and level, which need to be present in either the first or second column (it doesn't matter which one is in which column, they just both need to be present for the script to function). The script will validate against the following column headers and map to the corresponding elements in the EAD:

• level_type (ead:c @level=collection, series, subseries, etc.) required
• level (depth of c: 1, 2, 3, etc.) required
• unittitle (ead:unittitle, may contain <title render="*"> elements)
• date (ead:unitdate)
• date_begin (ead:unitdate @normal before the /)
• date_end (ead:unitdate @normal after the /)
• box (value of ead:container with @type="Box")
• folder (value of ead:container with @type="Folder")
• digital_file (value of ead:container with @type="Digital_file")
• oversize (value of ead:container with @type="Oversize")
• instance_type (ead:container @label value)
• general_note (ead:odd)
• accessrestrict (ead:accessrestrict)
• userestrict will validate but output has not been implemented yet
• scopecontent will validate but output has not been implemented yet

The script does not construct a fully valid EAD, just the container list (the <dsc> section of the finding aid).