def processDescription(self):
"""
technotes:
pdf - remove "pdf" but keep rest of description
jpg - remove entire description
manuscripts:
pdf - remove entire description
jpg - remove entire description
"""
rec = self.lib_dc_rec
desc_field = "dc:description"
issue_field = "library_dc:issue"
altTitle_field = "library_dc:altTitle"
for field in [desc_field, issue_field, altTitle_field]:
utils.validateField(field)
vals = rec.getFieldValues(desc_field)
new_descriptions = []
for desc in vals:
if desc is None:
continue
desc = desc.strip()
if desc.lower().endswith('pdf'):
desc = desc[:-3]
if self.collection == 'manuscripts':
continue
if self.collection == 'monographs' and 'ASR' in desc:
self.lib_dc_rec.addFieldValue(altTitle_field, desc.strip())
continue
elif desc and desc.lower().endswith('.jpg'):
continue
if desc:
new_descriptions.append(desc)
rec.setFieldValues(desc_field, new_descriptions)
def toTitleCase(self, field):
"""
put the field into title-case
"""
utils.validateField(field)
values = self.lib_dc_rec.getFieldValues(field)
newValues = []
for val in values:
newValues.append(titlecase(val.strip()))
if newValues:
self.lib_dc_rec.setFieldValues(field, newValues)
def showVocabs(rp):
print "\n%s" % rp.recId
rec = rp.lib_dc_rec
field_list = [
"library_dc:instName", 'library_dc:instDivision',
'library_dc:libraryType'
]
for field in field_list:
utils.validateField(field)
showFieldValues(field, rec.getFieldValues(field))
def periods(rp, field="dc:title"):
rec = rp.lib_dc_rec
utils.validateField(field)
field_vals = rec.getFieldValues(field)
period_vals = []
for val in field_vals:
if val[-1] == '.':
period_vals.append(val)
if period_vals:
print "\n%s (%s)" % (rp.recId, field)
for val in period_vals:
print "\t'%s'" % val
def dedup(self, field):
"""
eliminate duplicate values for each field.
THIS DOES NOT NEED TO BE USED
"""
utils.validateField(field)
nodups = []
rec = self.lib_dc_rec
vals = rec.getFieldValues(field)
for val in vals:
if not val in nodups:
nodups.append(val)
rec.setFieldValues(field, nodups)
def multiFields(rp):
rec = rp.lib_dc_rec
multis = {}
for field in field_list:
utils.validateField(field)
vals = rec.getFieldValues(field)
if len(vals) > 1:
multis[field] = vals
if multis:
print "\n", rp.recId
for field in multis.keys():
showFieldValues(field, multis[field])
def normalizeField(self, field, fn):
"""
function takes a single value and returns
the normalized value for this field
"""
rec = self.lib_dc_rec
utils.validateField(field)
vals = rec.getFieldValues(field)
normalized = []
for val in vals:
normalized_val = fn(val)
if normalized_val:
normalized.append(normalized_val)
rec.setFieldValues(field, normalized)
def removeDupValues(self, field1, field2):
"""
remove values in field1 that are found in field2
"""
utils.validateField(field1)
utils.validateField(field2)
rec = self.lib_dc_rec
vals1 = rec.getFieldValues(field1)
vals2 = map(string.upper, rec.getFieldValues(field2))
cleaned = []
for val in vals1:
if not val.upper() in vals2:
cleaned.append(val)
rec.setFieldValues(field1, cleaned)
def fieldsHaveMatchingValues(rp, field1, field2, verbose=True):
utils.validateField(field1)
utils.validateField(field2)
rec = rp.lib_dc_rec
vals1 = map(string.upper, rec.getFieldValues(field1))
vals2 = map(string.upper, rec.getFieldValues(field2))
for val in vals1:
if val in vals2:
if verbose:
print "\n", rp.recId
showFieldValues(field1, vals1)
showFieldValues(field2, vals2)
return 1
return 0
def normalizeDateDigitized(self):
"""
Ensure this field has a SINGLE VALUE
FOR NOW (as of 10/27/08) take LATEST date
"""
field = "library_dc:date_digitized"
rec = self.lib_dc_rec
utils.validateField(field)
vals = rec.getFieldValues(field)
touse = -1
if vals:
for val in vals:
year = int(val)
touse = max(touse, year)
rec.removeField(field)
rec.setFieldValue(field, str(touse))
def massageTitleAndAltTitle(self):
"""
often, we need to swap title and alt title values. but the
rules for this are different for each framework
we don't process 'technotes' or 'theses'
"""
rec = self.lib_dc_rec
title_field = 'dc:title'
altTitle_field = 'library_dc:altTitle'
utils.validateField(title_field)
utils.validateField(altTitle_field)
title_vals = rec.getFieldValues(title_field)
altTitle_vals = rec.getFieldValues(altTitle_field)
if title_vals and altTitle_vals:
# manuscripts case
if self.collection == "manuscripts" and len(
title_vals) == 1 and len(altTitle_vals) == 1:
title = title_vals[0]
alt = altTitle_vals[0]
if title.startswith(
alt[:-1]): # some altTitles end in period, some don't
rec.setFieldValue(title_field, alt)
rec.setFieldValue(altTitle_field, title)
if self.collection == "monographs":
## is there a title element containing 'ASR'?
ASRTitle = None
for title in title_vals:
if "ASR" in title:
ASRTitle = title
continue
## is there an altTitle containing 'Annual Scientific Report'?
ASRAltTitle = None
pat = 'Annual Scientific Report'
for altTitle in altTitle_vals:
if pat in altTitle:
ASRAltTitle = altTitle
continue
if ASRTitle and ASRAltTitle:
rec.removeField(altTitle_field)
rec.setFieldValue(title_field, ASRAltTitle)
rec.setFieldValue(altTitle_field, ASRTitle)
def showTitleStuff(rp):
print "\n%s" % rp.recId
rec = rp.lib_dc_rec
for field in field_list:
utils.validateField(field)
showFieldValues(field, rec.getFieldValues(field))