def rendermedia(filecontent):
result = deURN(filecontent)
rows = result.split('\n')
FIELDS = rows[0].strip().split('\t')
rows = rows[1:]
result = []
for counter, row in enumerate(rows):
row = row.strip() # seems there may be a stray \r still at the end of the string.
if row == '' or row[0] == '#': continue
row = row.split('\t')
media = {'otherfields': []}
media['counter'] = counter
for i,r in enumerate(row):
if FIELDS[i] == 'objectnumber':
media['accession'] = row[i]
elif FIELDS[i] == 'name':
media['mainentry'] = row[i]
# media['otherfields'].append({'label': 'File', 'value': row[i]})
elif FIELDS[i] == 'objectCSID':
media['csid'] = row[i]
elif FIELDS[i] == 'mediaCSID':
media['media'] = row[i]
elif FIELDS[i] == 'blobCSID':
media['blobs'] = [ row[i] ]
elif FIELDS[i] == 'creator':
media['otherfields'].append({'label': 'Creator', 'value': row[i]})
elif FIELDS[i] == 'description':
media['otherfields'].append({'label': 'Description', 'value': row[i]})
elif FIELDS[i] == 'date':
media['otherfields'].append({'label': 'Image Date', 'value': row[i]})
result.append(media)
return result
def reformat(filecontent):
result = deURN(filecontent)
result = result.replace('\n', '<tr><td>')
result = result.replace('\t', '<td>')
result = result.replace('|', '<td>')
result = result.replace('False', '<span class="error">False</span>')
result += '</table>'
return '<table width="100%"><tr><td>\n' + result
def reformat(filecontent):
result = deURN(filecontent)
result = result.replace('\n','<tr><td>')
result = result.replace('\t','<td>')
result = result.replace('|','<td>')
result = result.replace('False','<span class="error">False</span>')
result += '</table>'
return '<table width="100%"><tr><td>\n' + result
def extractTag(xml, tag):
element = xml.find('.//%s' % tag)
try:
if "urn:" in element.text:
element_text = deURN(str(element.text))
else:
element_text = element.text
except:
element_text = ''
return element_text
def rendermedia(filecontent):
result = deURN(filecontent)
rows = result.split('\n')
FIELDS = rows[0].strip().split('\t')
rows = rows[1:]
result = []
for counter, row in enumerate(rows):
row = row.strip(
) # seems there may be a stray \r still at the end of the string.
if row == '' or row[0] == '#': continue
row = row.split('\t')
media = {'otherfields': []}
media['counter'] = counter
for i, r in enumerate(row):
if FIELDS[i] == 'objectnumber':
media['accession'] = row[i]
elif FIELDS[i] == 'name':
media['mainentry'] = row[i]
# media['otherfields'].append({'label': 'File', 'value': row[i]})
elif FIELDS[i] == 'objectCSID':
media['csid'] = row[i]
elif FIELDS[i] == 'mediaCSID':
media['media'] = row[i]
elif FIELDS[i] == 'blobCSID':
media['blobs'] = [row[i]]
elif FIELDS[i] == 'creator':
media['otherfields'].append({
'label': 'Creator',
'value': row[i]
})
elif FIELDS[i] == 'description':
media['otherfields'].append({
'label': 'Description',
'value': row[i]
})
elif FIELDS[i] == 'date':
media['otherfields'].append({
'label': 'Image Date',
'value': row[i]
})
result.append(media)
return result
def extract_refname(xml, term, pgSz, record_type):
try:
cspaceXML = fromstring(xml)
totalItems = int(cspaceXML.find('.//totalItems').text)
if totalItems == 0:
return 'ZeroResults X X X X'.split(' '), totalItems
items = cspaceXML.findall('.//list-item')
for i in items:
csid = i.find('.//csid')
csid = csid.text
try:
refName = extract_tag(i, 'refName')
updated_at = extract_tag(i, 'updatedAt')
if record_type == 'collectionobjects':
try:
termDisplayName = deURN(refName)
except:
raise
else:
try:
termDisplayName = extract_tag(i, 'termDisplayName')
except:
try:
termDisplayName = extract_tag(i, 'displayName')
except:
raise
except:
loginfo(
'csvimport',
'could not get termDisplayName or refName or updatedAt from %s'
% csid, {}, {})
return 'Failed X X X X'.split(' '), totalItems
if normalize(term) == normalize(termDisplayName):
return ['OK', csid,
str(termDisplayName), refName, updated_at], totalItems
if totalItems > pgSz:
return 'MaybeMissed X X X X'.split(' '), totalItems
return ['NoMatch', '', term, term, 'X'], totalItems
except:
raise
return 'xmlParseFailed X X X X'.split(' '), totalItems
def updateXML(fieldset, updateItems, xml):
message = ''
# Fields vary with fieldsets
if fieldset == 'keyinfo':
fieldList = ('pahmaFieldCollectionPlace', 'assocPeople', 'objectName',
'pahmaEthnographicFileCode')
elif fieldset == 'namedesc':
fieldList = ('briefDescription', 'objectName')
elif fieldset == 'registration':
# nb: 'pahmaAltNumType' is handled with 'pahmaAltNum'
fieldList = ('objectName', 'pahmaAltNum', 'fieldCollector')
elif fieldset == 'hsrinfo':
fieldList = ('objectName', 'pahmaFieldCollectionPlace',
'briefDescription')
elif fieldset == 'objtypecm':
fieldList = ('objectName', 'collection', 'responsibleDepartment',
'pahmaFieldCollectionPlace', 'pahmaTmsLegacyDepartment')
elif fieldset == 'collection':
fieldList = ('objectName', 'collection')
elif fieldset == 'placeanddate':
fieldList = ('objectName', 'pahmaFieldLocVerbatim',
'pahmaFieldCollectionDate')
elif fieldset == 'places':
fieldList = ('pahmaFieldLocVerbatim', 'pahmaFieldCollectionPlace',
'objectProductionPlace', 'contentPlace')
elif fieldset == 'dates':
fieldList = ('objectProductionDate', 'pahmaFieldCollectionDate',
'contentDate', 'briefDescription')
elif fieldset == 'mattax':
fieldList = ('material', 'taxon', 'briefDescription')
elif fieldset == 'student':
fieldList = ('taxon', 'fieldLocCountry', 'fieldLocState',
'fieldLocCounty')
elif fieldset == 'fullmonty':
fieldList = ('assocPeople', 'briefDescription', 'collection',
'contentDate', 'contentPlace', 'fieldCollector',
'material', 'objectName', 'objectName',
'objectProductionDate', 'objectProductionPlace',
'objectProductionPerson', 'pahmaAltNum',
'pahmaEthnographicFileCode', 'pahmaFieldCollectionDate',
'pahmaFieldCollectionPlace', 'pahmaFieldLocVerbatim',
'pahmaObjectStatus', 'responsibleDepartment', 'taxon',
'material')
root = etree.fromstring(xml)
# add the user's changes to the XML
for relationType in fieldList:
# sys.stderr.write('tag1: %s\n' % relationType)
# this app does not insert empty values into anything!
if not relationType in updateItems.keys(
) or updateItems[relationType] == '':
continue
listSuffix = 'List'
extra = ''
if relationType in [
'assocPeople', 'pahmaAltNum', 'pahmaFieldCollectionDate',
'objectProductionDate', 'objectProductionPlace',
'objectProductionPerson', 'contentDate', 'material', 'taxon',
'fieldLocCountry', 'fieldLocState', 'fieldLocCounty'
]:
extra = 'Group'
elif relationType in [
'briefDescription', 'fieldCollector', 'responsibleDepartment',
'contentPlace'
]:
listSuffix = 's'
if relationType in [
'collection', 'pahmaFieldLocVerbatim', 'contentDate',
'pahmaTmsLegacyDepartment'
]:
listSuffix = ''
else:
pass
# html += ">>> ",'.//'+relationType+extra+'List'
# sys.stderr.write('tag2: %s\n' % (relationType + extra + listSuffix))
if relationType == 'taxon':
tmprelationType = 'taxonomicIdent'
elif relationType in [
'fieldLocCountry', 'fieldLocState', 'fieldLocCounty'
]:
tmprelationType = 'locality'
else:
tmprelationType = relationType
metadata = root.findall('.//' + tmprelationType + extra + listSuffix)
if 'objectNumber' in updateItems and updateItems['objectNumber'] == '':
updateItems['objectNumber'] = root.find('.//objectNumber').text
try:
metadata = metadata[0] # there had better be only one!
except:
# hmmm ... we didn't find this element in the record. Make a note a carry on!
# message += 'No "' + relationType + extra + listSuffix + '" element found to update.'
continue
# html += ">>> ",relationType,':',updateItems[relationType]
if relationType in [
'assocPeople', 'objectName', 'pahmaAltNum', 'material',
'taxon', 'objectProductionPerson', 'objectProductionPlace',
'fieldLocCountry', 'fieldLocState', 'fieldLocCounty'
]:
# group = metadata.findall('.//'+relationType+'Group')
# sys.stderr.write(' updateItem: ' + relationType + ':: ' + updateItems[relationType] + '\n' )
Entries = metadata.findall('.//' + relationType)
if not alreadyExists(updateItems[relationType], Entries):
try:
newElement = metadata.findall('.//' + tmprelationType +
'Group')[0]
except:
newElement = etree.Element(tmprelationType + 'Group')
leafElement = etree.Element(relationType)
leafElement.text = updateItems[relationType]
newElement.append(leafElement)
if relationType in ['assocPeople', 'pahmaAltNum']:
apgType = etree.Element(relationType + 'Type')
# this needs to be a refname for PAHMA's assocpeopletype...
apgType.text = updateItems[
relationType +
'Type'] if relationType == 'pahmaAltNum' else "urn:cspace:pahma.cspace.berkeley.edu:vocabularies:name(assocpeople):item:name(assocpeopletype06)'made by'"
# sys.stderr.write(relationType + 'Type:' + updateItems[relationType + 'Type'])
newElement.append(apgType)
if (len(Entries) == 1 and Entries[0].text is None
) or tmprelationType == 'locality':
# sys.stderr.write('reusing empty element: %s\n' % Entries[0].tag)
# sys.stderr.write('ents : %s\n' % Entries[0].text)
for child in metadata:
# html += '<br>tag: ', child.tag
if child.tag == tmprelationType + 'Group':
# html += '<br> found it! ',child.tag
metadata.remove(child)
metadata.insert(0, newElement)
else:
metadata.insert(0, newElement)
else:
if IsAlreadyPreferred(updateItems[relationType], Entries):
continue
else:
# exists, but not preferred. make it the preferred: remove it from where it is, insert it as 1st
for child in metadata:
if child.tag == tmprelationType + 'Group':
checkval = child.find('.//' + relationType)
if checkval.text == updateItems[relationType]:
savechild = child
metadata.remove(child)
metadata.insert(0, savechild)
pass
# for AltNums, we need to update the AltNumType even if the AltNum hasn't changed
if relationType == 'pahmaAltNum':
apgType = metadata.find('.//' + relationType + 'Type')
apgType.text = updateItems[relationType + 'Type']
# sys.stderr.write(' updated: pahmaAltNumType to' + updateItems[relationType + 'Type'] + '\n' )
elif relationType in [
'briefDescription', 'fieldCollector', 'responsibleDepartment',
'contentPlace'
]:
Entries = metadata.findall('.//' + relationType)
# for e in Entries:
# html += '%s, %s<br>' % (e.tag, e.text)
# sys.stderr.write(' e: %s\n' % e.text)
if alreadyExists(updateItems[relationType], Entries):
if IsAlreadyPreferred(updateItems[relationType], Entries):
# message += "%s exists as %s, already preferred;" % (updateItems[relationType],relationType)
pass
else:
# exists, but not preferred. make it the preferred: remove it from where it is, insert it as 1st
for child in Entries:
sys.stderr.write(' c: %s\n' % child.tag)
if child.text == updateItems[relationType]:
new_element = child
metadata.remove(child)
# message += '%s removed. len = %s<br/>' % (child.text, len(Entries))
metadata.insert(0, new_element)
message += "'%s' exists in %s, now preferred.<br/>" % (
deURN(updateItems[relationType]), relationType)
# html += 'already exists: %s<br>' % updateItems[relationType]
# check if the existing element is empty; if so, use it, don't add a new element
else:
if len(Entries) == 1 and Entries[0].text is None:
# message += "removed %s ;<br/>" % (Entries[0].tag)
metadata.remove(Entries[0])
new_element = etree.Element(relationType)
new_element.text = updateItems[relationType]
metadata.insert(0, new_element)
message += "added '%s' as the preferred term in %s.<br/>" % (
deURN(updateItems[relationType]), relationType)
elif relationType in [
'objectProductionDate', 'pahmaFieldCollectionDate',
'contentDate'
]:
# we'll be replacing the entire structured date group
newDateGroup = etree.Element('%sGroup' % relationType)
new_element = etree.Element('dateDisplayDate')
new_element.text = updateItems[relationType]
newDateGroup.insert(0, new_element)
DateGroup = metadata.find('.//%sGroup' % relationType)
if DateGroup is not None:
metadata.remove(DateGroup)
# one of many special cases...
if relationType == 'contentDate':
DateGroup = metadata.findall('.//*')
[metadata.remove(d) for d in DateGroup]
metadata.insert(0, new_element)
else:
metadata.insert(0, newDateGroup)
else:
# check if value is already present. if so, skip
if alreadyExists(updateItems[relationType],
metadata.findall('.//' + relationType)):
if IsAlreadyPreferred(updateItems[relationType],
metadata.findall('.//' + relationType)):
continue
else:
message += "'%s' already exists as an NPT in %s: This value has been inserted as the PT and in doing so is now duplicated.<br/>" % (
deURN(updateItems[relationType]), relationType)
pass
newElement = etree.Element(relationType)
newElement.text = updateItems[relationType]
metadata.insert(0, newElement)
objectCount = root.find('.//numberOfObjects')
if 'objectCount' in updateItems:
if objectCount is None:
objectCount = etree.Element('numberOfObjects')
collectionobjects_common = root.find(
'.//{http://collectionspace.org/services/collectionobject}collectionobjects_common'
)
collectionobjects_common.insert(0, objectCount)
objectCount.text = updateItems['objectCount']
inventoryCount = root.find('.//inventoryCount')
if 'inventoryCount' in updateItems:
if inventoryCount is None:
inventoryCount = etree.Element('inventoryCount')
collectionobjects_pahma = root.find(
'.//{http://collectionspace.org/services/collectionobject/local/pahma}collectionobjects_pahma'
)
collectionobjects_pahma.insert(0, inventoryCount)
inventoryCount.text = updateItems['inventoryCount']
for fld in 'pahmaTmsLegacyDepartment pahmaFieldLocVerbatim'.split(' '):
if fld in updateItems and updateItems[fld] != '':
fldtoupdate = root.find('.//' + fld)
if fldtoupdate is None:
fldtoupdate = etree.Element(fld)
collectionobjects_pahma = root.find(
'.//{http://collectionspace.org/services/collectionobject/local/pahma}collectionobjects_pahma'
)
collectionobjects_pahma.insert(0, fldtoupdate)
fldtoupdate.text = updateItems[fld]
collection = root.find('.//collection')
if 'collection' in updateItems:
if collection is None:
collection = etree.Element('collection')
collectionobjects_common = root.find(
'.//{http://collectionspace.org/services/collectionobject}collectionobjects_common'
)
collectionobjects_common.insert(0, collection)
message += " %s added as <%s>.<br/>" % (deURN(
updateItems['collection']), 'collection')
collection.text = updateItems['collection']
payload = '<?xml version="1.0" encoding="UTF-8"?>\n' + etree.tostring(
root, encoding='unicode')
# update collectionobject..
# html += "<br>pretending to post update to %s to REST API..." % updateItems['objectCsid']
# elapsedtimetotal = time.time()
# messages = []
# messages.append("posting to %s REST API..." % uri)
# print(payload)
# messages.append(payload)
return message, payload
