Sample script to harvest metadata through CONTENTdm v6 API and format as csv for upload into ui-libraries fork of Omeka/Scripto. See ui-libraries/plugin-Scripto for documentation. Uses pycdm, a python library for working with the CONTENTdm v6 API (saverkamp/pycdm).

sample-cdm2scripto.py

import codecs
import csv
import datetime
import pycdm
from HTMLParser import HTMLParser


#get input: alias + items to retrieve
alias = raw_input('collection alias: ')
items = raw_input('item identifiers (separate by commas): ')
ptrs = items.split(',')
#current date-time for output filenames
today = datetime.datetime.now().strftime('%Y-%m-%d--%H-%M')

#create file-level metadata csv file
fileOutput = alias + today + '_File.csv'
fFile = codecs.open(fileOutput, 'wb', encoding='utf_8')
wtrFile = csv.writer(fFile, delimiter=',')
#header row for file-level csv file
fileHeaderRow = ['filename', 'title', 'identifier', 'source', 'status', 'transcription', 'Omeka file order']
wtrFile.writerow(fileHeaderRow)

#create item-level metadata csv file
itemOutput = alias + today + '_Item.csv'
fItem = codecs.open(itemOutput, 'wb', encoding='utf_8')
wtrItem = csv.writer(fItem, delimiter=',')
#header row for item-level csv file
itemHeaderRow = ['title', 'identifier', 'source', 'ispartof', 'relation', 'audience', 'files']
wtrItem.writerow(itemHeaderRow)

#get data for each item
for ptr in ptrs:
    #call api for item metadata
    item = pycdm.item(alias, ptr, 'on')
    #set item-level metadata
    #create unique item id for use outside CDM
    itemID = alias + '_' + ptr
    source = item.refurl
    itemtitle = item.info['title']
    #digital collection url
    ispartof = item.collection.url
    #default sorting number, maps to dc:Audience in Omeka
    sort = '000000'
    #collection guide url
    if ('findin' in item.info):
        relation = item.info['findin']
    elif ('collea' in item.info):
        relation = item.info['collea']
    #list for file locations
    files = []
    #set counter for file order
    order = 1
    #set file-level metadata
    for page in item.pages:
        #create unique page id for use outside CDM
        fileID = itemID + '_' + page.id
        pagelabel = page.label
        pageRefURL = page.refurl
        #set transcription, if available
        #assumes you have a field for full text with nickname 'full' or 'fula'
        if (('full' in page.info) and page.info['full']):
            transcription = str(page.info['full'].encode('ascii', 'ignore'))
            transcription = HTMLParser().unescape(transcription)
        elif (('fula' in page.info) and page.info['fula']):
            transcription = str(page.info['fula'].encode('ascii', 'ignore'))
            transcription = HTMLParser().unescape(transcription)
        else:
            transcription = ''
        #set transcription status
        if (transcription == ''):
            status = 'Not Started'
        else:
            status = 'Needs Review'
        url = page.fileurl
        files.append(url)
        #write file metadata to file-level csv file
        filerow = [url, pagelabel, fileID, pageRefURL, status, transcription.encode('ascii', 'ignore'), order]
        wtrFile.writerow(filerow)
        order += 1
    #write item metadata to item-level csv file
    files = ','.join(files)
    itemrow = [itemtitle, itemID, source, ispartof, relation, sort, files]
    wtrItem.writerow(itemrow)
    print ptr
fItem.close()
fFile.close()