def test_parse_with_qualifiers(self):
fd = parse('1985?')
assert fd.year == u'1985', fd
assert fd.qualifier == u'Uncertainty : 1985?', fd.qualifier
fd = parse('c.1780')
assert fd.year == u'1780', fd
assert fd.qualifier == u"Note 'circa' : c.1780", fd
fd = parse('c. 1780')
assert fd.year == u'1780', fd
assert fd.qualifier.startswith(u"Note 'circa'"), fd
def test_parse_with_qualifiers(self):
fd = parse('1985?')
assert fd.year == u'1985', fd
assert fd.qualifier == u'Uncertainty : 1985?', fd.qualifier
fd = parse('c.1780')
assert fd.year == u'1780', fd
assert fd.qualifier == u"Note 'circa' : c.1780", fd
fd = parse('c. 1780')
assert fd.year == u'1780', fd
assert fd.qualifier.startswith(u"Note 'circa'"), fd
def test_parse(self):
d1 = datetime.datetime(2000, 1, 23)
fd = parse(d1)
assert fd.year == '2000'
fd = parse('March 1762')
assert str(fd) == '1762-03'
fd = parse(1966)
assert str(fd) == '1966'
fd = parse('22/07/2010')
assert fd.month == '07', fd.month
def test_parse(self):
d1 = datetime.datetime(2000, 1, 23)
fd = parse(d1)
assert fd.year == '2000'
fd = parse('March 1762')
assert str(fd) == '1762-03'
fd = parse(1966)
assert str(fd) == '1966'
fd = parse('22/07/2010')
assert fd.month == '07', fd.month
def test_ambiguous(self):
# TODO: have to be careful here ...
fd = parse('1068/1069')
def test_parse_wildcards(self):
fd = parse('198?')
assert fd.year == '', fd.year # expect this to not parse
def test_parse_with_none(self):
d1 = parse(None)
assert d1 is None
def test_parse_ambiguous_day_month(self):
fd = parse('05/07/2010')
assert fd.month == '07', fd.month
assert fd.day == '05', fd.month
def get_combined_index_data(self):
combined_index_data = {}
title = self._get_text_or_none('.//oai:title')
if title:
combined_index_data['title'] = title
description = self._get_text_or_none('.//oai:description')
if description:
combined_index_data['description'] = description
date = None
date_granularity = 0
# Get date start
date_start_raw = self._get_text_or_none('.//oai:production.date.start')
# Skip date if 5 or more digits are concatenated, because we
# don't know how to parse those numbers
if date_start_raw:
if not re.search(r'\d{5}', date_start_raw):
if re.search(r'\d', date_start_raw):
date_start = parse(date_start_raw.strip())
# Get date end
date_end_raw = self._get_text_or_none('.//oai:production.date.end')
if re.search(r'\d', date_end_raw):
date_end = parse(date_end_raw.strip())
# Only save date if something got parsed
if date_start.isoformat():
# Determine granularity
date_granularity = len(re.findall(r'\d', date_start.isoformat()))
# Take average of start and end date if end date does not
# exist and they are not the same
if not date_end.isoformat():
date = date_start.as_datetime()
elif date_start.isoformat() == date_end.isoformat():
date = date_start.as_datetime()
else:
# Take the average of years using integers to also
# support before christ years (which datetime doesn't
# support)
if len(date_start.isoformat()) == 4:
average_year = int(date_start.year) + (int(date_end.year) - int(date_start.year)) / 2
date = parse(average_year).as_datetime()
# Take averages for dates with months and days using
# datetime
else:
date = date_start.as_datetime() + (date_end.as_datetime() - date_start.as_datetime()) / 2
combined_index_data['date'] = date
combined_index_data['date_granularity'] = date_granularity
authors = self._get_text_or_none('.//oai:creator')
if authors:
combined_index_data['authors'] = [authors]
mediums = self.original_item.findall('.//oai:imageUrl', namespaces=self.namespaces)
if mediums:
combined_index_data['media_urls'] = []
for medium in mediums:
#if medium.text.strip().split('.')[-1].lower() in self.media_mime_types:
combined_index_data['media_urls'].append(
{
'original_url': unicode(medium.text.strip()),
'content_type': self.media_mime_types[
unicode(medium.text).strip().split('.')[-1].lower()
]
}
)
return combined_index_data
def test_small_years_with_zeros(self):
in1 = '0023'
fd = parse(in1)
assert str(fd) == '0023', fd
assert fd.as_float() == 23, fd.as_float()
def test_ambiguous(self):
# TODO: have to be careful here ...
fd = parse('1068/1069')
def test_parse_wildcards(self):
fd = parse('198?')
assert fd.year == '', fd.year # expect this to not parse
def test_parse_with_none(self):
d1 = parse(None)
assert d1 is None
def test_parse_ambiguous_day_month(self):
fd = parse('05/07/2010')
assert fd.month == '07', fd.month
assert fd.day == '05', fd.month
def test_small_years_with_zeros(self):
in1 = '0023'
fd = parse(in1)
assert str(fd) == '0023', fd
assert fd.as_float() == 23, fd.as_float()
def test_years_with_alpha_prefix(self):
in1 = "p1980"
fd = parse(in1)
assert str(fd) == "1980", fd
def test_years_with_alpha_prefix(self):
in1 = "p1980"
fd = parse(in1)
assert str(fd) == "1980", fd
def get_combined_index_data(self):
combined_index_data = {}
title = self._get_text_or_none('.//oai:title')
if title:
combined_index_data['title'] = title
description = self._get_text_or_none('.//oai:description')
if description:
combined_index_data['description'] = description
date = None
date_granularity = 0
# Get date start
date_start_raw = self._get_text_or_none('.//oai:production.date.start')
# Skip date if 5 or more digits are concatenated, because we
# don't know how to parse those numbers
if date_start_raw:
if not re.search(r'\d{5}', date_start_raw):
if re.search(r'\d', date_start_raw):
date_start = parse(date_start_raw.strip())
# Get date end
date_end_raw = self._get_text_or_none(
'.//oai:production.date.end')
if re.search(r'\d', date_end_raw):
date_end = parse(date_end_raw.strip())
# Only save date if something got parsed
if date_start.isoformat():
# Determine granularity
date_granularity = len(
re.findall(r'\d', date_start.isoformat()))
# Take average of start and end date if end date does not
# exist and they are not the same
if not date_end.isoformat():
date = date_start.as_datetime()
elif date_start.isoformat() == date_end.isoformat():
date = date_start.as_datetime()
else:
# Take the average of years using integers to also
# support before christ years (which datetime doesn't
# support)
if len(date_start.isoformat()) == 4:
average_year = int(date_start.year) + (
int(date_end.year) - int(date_start.year)) / 2
date = parse(average_year).as_datetime()
# Take averages for dates with months and days using
# datetime
else:
date = date_start.as_datetime() + (
date_end.as_datetime() -
date_start.as_datetime()) / 2
combined_index_data['date'] = date
combined_index_data['date_granularity'] = date_granularity
authors = self._get_text_or_none('.//oai:creator')
if authors:
combined_index_data['authors'] = [authors]
mediums = self.original_item.findall('.//oai:imageUrl',
namespaces=self.namespaces)
if mediums:
combined_index_data['media_urls'] = []
for medium in mediums:
#if medium.text.strip().split('.')[-1].lower() in self.media_mime_types:
combined_index_data['media_urls'].append({
'original_url':
unicode(medium.text.strip()),
'content_type':
self.media_mime_types[unicode(
medium.text).strip().split('.')[-1].lower()]
})
return combined_index_data
