def __init__(self, function):
self.function = function
self.total_pages = 0
self.pages_complete = 0
self.series_client = RSSeriesSearchClient()
self.search_client = RSSearchClient()
db = self.get_db()
self.functions = db.functions
self.series = db.series
self.series.create_index('identifier', unique=True)
def __init__(self, harvest, **kwargs):
self.total_pages = None
self.pages_complete = 0
self.client = RSSearchClient()
self.prepare_harvest(access='Closed')
db = self.get_db()
self.items = db.items
self.series = db.series
self.harvests = db.harvests
self.set_harvest(harvest)
def __init__(self, function):
self.function = function
self.total_pages = 0
self.pages_complete = 0
self.series_client = RSSeriesSearchClient()
self.search_client = RSSearchClient()
db = self.get_db()
self.functions = db.functions
self.series = db.series
self.series.create_index('identifier', unique=True)
class SearchHarvester():
"""
Harvest the details of 'Closed' files from RecordSearch.
Saves to MongoDB.
harvester = SearchHarvester(harvest='2015-12-31')
harvester.start_harvest()
"""
def __init__(self, harvest, **kwargs):
self.total_pages = None
self.pages_complete = 0
self.client = RSSearchClient()
self.prepare_harvest(access='Closed')
db = self.get_db()
self.items = db.items
self.series = db.series
self.harvests = db.harvests
self.set_harvest(harvest)
def set_harvest(self, harvest):
# self.harvests.create_index('harvest_date', unique=True)
dates = harvest.split('-')
harvest_date = datetime(int(dates[0]), int(dates[1]), int(dates[2]), 0, 0, 0)
try:
self.harvests.insert({'harvest_date': harvest_date})
except DuplicateKeyError:
pass
self.harvest_date = harvest_date
def get_db(self):
dbclient = MongoClient(MONGOLAB_URL)
db = dbclient.get_default_database()
return db
def get_total(self):
return self.client.total_results
def prepare_harvest(self, **kwargs):
self.client.search(**kwargs)
total_results = self.client.total_results
print '{} items'.format(total_results)
self.total_pages = (int(total_results) / self.client.results_per_page) + 1
print '{} pages'.format(self.total_pages)
def start_harvest(self, page=None):
item_client = RSItemClient()
series_client = RSSeriesClient()
# Refresh series with each harvest
# self.series.remove({})
# self.items.remove({})
if not page:
page = self.pages_complete + 1
else:
self.pages_complete = page - 1
while self.pages_complete < self.total_pages:
response = self.client.search(access='Closed', page=page, sort='9')
for result in response['results']:
exists = self.items.find_one({'_id': result['identifier'], 'harvests': self.harvest_date})
if not exists:
item = item_client.get_summary(entity_id=result['identifier'])
item['_id'] = item['identifier']
item['random_id'] = [random.random(), 0]
# Normalise reasons
item['reasons'] = []
# item['year'] = item['contents_dates']['end_date']['date'].year
for reason in item['access_reason']:
matched = False
for exception, pattern in EXCEPTIONS:
if re.match(pattern, reason['reason']):
item['reasons'].append(exception)
matched = True
if not matched:
item['reasons'].append(reason['reason'])
# Get series and agency info
print item['series']
series = self.series.find_one({'identifier': item['series']})
if not series:
try:
series = series_client.get_summary(entity_id=item['series'], include_access_status=False)
# agencies = series_client.get_controlling_agencies(entity_id=item['series'])
# series['controlling_agencies'] = agencies
self.series.insert(series)
except UsageError:
series = None
if series:
item['series_title'] = series['title']
if series['controlling_agencies']:
item['agencies'] = []
for agency in series['controlling_agencies']:
if not agency['end_date'] or not agency['end_date']['date']:
item['agencies'].append(agency)
item['harvests'] = [self.harvest_date]
self.items.insert_one(item)
print item['identifier']
self.pages_complete += 1
page += 1
print '{} pages complete'.format(self.pages_complete)
time.sleep(1)
class SeriesHarvester():
"""
Works through agencies associated with a harvested function.
Grabs series in the function date range and gets some details.
"""
def __init__(self, function):
self.function = function
self.total_pages = 0
self.pages_complete = 0
self.series_client = RSSeriesSearchClient()
self.search_client = RSSearchClient()
db = self.get_db()
self.functions = db.functions
self.series = db.series
self.series.create_index('identifier', unique=True)
def get_db(self):
dbclient = MongoClient(MONGOLAB_URL)
db = dbclient.get_default_database()
return db
def start_harvest(self):
# Get agencies in function
function = self.functions.find_one({'function': self.function})
agencies = []
for agency in function['agencies']:
if agency['agency_id'] == 'CA 51':
self.pages_complete = 0
page = 1
if not agency['function_start']['date']:
function_start = '1900'
else:
function_start = str(agency['function_start']['date'].year)
if not agency['function_end']['date']:
function_end = datetime.datetime.now().year
else:
function_end = str(agency['function_end']['date'].year)
series_list = []
self.prepare_harvest(agency, function_start, function_end)
while self.pages_complete < self.total_pages:
response = self.series_client.search_series(
page=page,
agency_recording=agency['agency_id'],
date_from=function_start,
date_to=function_end)
for result in response['results']:
if result['items_digitised'] == '20000+':
result['items_digitised'] = 20000
elif result['items_digitised'] is None:
result['items_digitised'] = 0
series = {
'series_id':
result['identifier'],
'title':
result['title'],
'items_described':
int(result['items_described']['described_number']),
'items_digitised':
result['items_digitised']
}
try:
self.search_client.search(
digitised=False,
series=result['identifier'],
date_from=function_start,
date_to=function_end)
series['items_described_in_period'] = int(
self.search_client.total_results)
except TooManyError:
series['items_described_in_period'] = 20000
try:
self.search_client.search(
digitised=False,
series=result['identifier'],
date_from=function_start,
date_to=function_end,
digital=['on'])
series['items_digitised_in_period'] = int(
self.search_client.total_results)
except TooManyError:
series['items_digitised_in_period'] = 20000
pp.pprint(series)
series_list.append(series)
try:
self.series.insert(result)
except DuplicateKeyError:
pass
self.pages_complete += 1
page += 1
time.sleep(1)
agency['series'] = series_list
agencies.append(agency)
self.functions.update_one({'function': self.function},
{'$set': {
'agencies': agencies
}})
def get_total(self):
return self.client.total_results
def prepare_harvest(self, agency, function_start, function_end):
self.pages_complete = 0
print agency['agency_id']
self.series_client.search_series(results_per_page=0,
agency_recording=agency['agency_id'],
date_from=function_start,
date_to=function_end)
total_results = self.series_client.total_results
if total_results is not None:
print '{} series'.format(total_results)
self.total_pages = (int(total_results) /
self.series_client.results_per_page) + 1
print '{} pages'.format(self.total_pages)
else:
print 'No series'
class SeriesHarvester():
"""
Works through agencies associated with a harvested function.
Grabs series in the function date range and gets some details.
"""
def __init__(self, function):
self.function = function
self.total_pages = 0
self.pages_complete = 0
self.series_client = RSSeriesSearchClient()
self.search_client = RSSearchClient()
db = self.get_db()
self.functions = db.functions
self.series = db.series
self.series.create_index('identifier', unique=True)
def get_db(self):
dbclient = MongoClient(MONGOLAB_URL)
db = dbclient.get_default_database()
return db
def start_harvest(self):
# Get agencies in function
function = self.functions.find_one({'function': self.function})
agencies = []
for agency in function['agencies']:
if agency['agency_id'] == 'CA 51':
self.pages_complete = 0
page = 1
if not agency['function_start']['date']:
function_start = '1900'
else:
function_start = str(agency['function_start']['date'].year)
if not agency['function_end']['date']:
function_end = datetime.datetime.now().year
else:
function_end = str(agency['function_end']['date'].year)
series_list = []
self.prepare_harvest(agency, function_start, function_end)
while self.pages_complete < self.total_pages:
response = self.series_client.search_series(page=page, agency_recording=agency['agency_id'], date_from=function_start, date_to=function_end)
for result in response['results']:
if result['items_digitised'] == '20000+':
result['items_digitised'] = 20000
elif result['items_digitised'] is None:
result['items_digitised'] = 0
series = {
'series_id': result['identifier'],
'title': result['title'],
'items_described': int(result['items_described']['described_number']),
'items_digitised': result['items_digitised']
}
try:
self.search_client.search(digitised=False, series=result['identifier'], date_from=function_start, date_to=function_end)
series['items_described_in_period'] = int(self.search_client.total_results)
except TooManyError:
series['items_described_in_period'] = 20000
try:
self.search_client.search(digitised=False, series=result['identifier'], date_from=function_start, date_to=function_end, digital=['on'])
series['items_digitised_in_period'] = int(self.search_client.total_results)
except TooManyError:
series['items_digitised_in_period'] = 20000
pp.pprint(series)
series_list.append(series)
try:
self.series.insert(result)
except DuplicateKeyError:
pass
self.pages_complete += 1
page += 1
time.sleep(1)
agency['series'] = series_list
agencies.append(agency)
self.functions.update_one({'function': self.function}, {'$set': {'agencies': agencies}})
def get_total(self):
return self.client.total_results
def prepare_harvest(self, agency, function_start, function_end):
self.pages_complete = 0
print agency['agency_id']
self.series_client.search_series(results_per_page=0, agency_recording=agency['agency_id'], date_from=function_start, date_to=function_end)
total_results = self.series_client.total_results
if total_results is not None:
print '{} series'.format(total_results)
self.total_pages = (int(total_results) / self.series_client.results_per_page) + 1
print '{} pages'.format(self.total_pages)
else:
print 'No series'
