def get_recids_to_load():
"""
Generates the final list of record IDs to load.
Returns a list of tuples like: (recid, date)
"""
recids_given = task_get_option("recids", default=[])
query_given = task_get_option("query")
reportnumbers_given = task_get_option("reportnumbers")
if query_given:
write_message("Performing given search query: %s" % (query_given,))
result = perform_request_search(p=query_given,
of='id',
rg=0,
wl=0)
recids_given.extend(result)
if reportnumbers_given:
write_message("Searching for records referring to given reportnumbers")
for reportnumber in reportnumbers_given:
result = perform_request_search(p='reportnumber:%s' % (reportnumber,),
of='id',
rg=0,
wl=0)
recids_given.extend(result)
recids_given = [(recid, None) for recid in recids_given]
last_id, last_date = fetch_last_updated(name="bibcatalog")
records_found = []
if task_get_option("new", default=False):
records_found.extend(get_all_new_records(since=last_date, last_id=last_id))
if task_get_option("modified", default=False):
records_found.extend(get_all_modified_records(since=last_date, last_id=last_id))
for recid, date in records_found:
recids_given.append((recid, date))
return recids_given
def get_recids_to_load():
"""
Generates the final list of record IDs to load.
Returns a list of tuples like: (recid, date)
"""
recids_given = task_get_option("recids", default=[])
query_given = task_get_option("query")
reportnumbers_given = task_get_option("reportnumbers")
if query_given:
write_message("Performing given search query: %s" % (query_given,))
result = perform_request_search(p=query_given,
of='id',
rg=0,
wl=0)
recids_given.extend(result)
if reportnumbers_given:
write_message("Searching for records referring to given reportnumbers")
for reportnumber in reportnumbers_given:
result = perform_request_search(p='reportnumber:%s' % (reportnumber,),
of='id',
rg=0,
wl=0)
recids_given.extend(result)
recids_given = [(recid, None) for recid in recids_given]
last_id, last_date = fetch_last_updated(name="bibcatalog")
records_found = []
if task_get_option("new", default=False):
records_found.extend(get_all_new_records(since=last_date, last_id=last_id))
if task_get_option("modified", default=False):
records_found.extend(get_all_modified_records(since=last_date, last_id=last_id))
for recid, date in records_found:
recids_given.append((recid, date))
return recids_given
def task_run_core(name=NAME):
"""Entry point for the arxiv-pdf-checker task"""
# First gather recids to process
recids = task_get_option("recids")
if recids:
start_date = None
recids = [(recid, None) for recid in recids]
else:
start_date = datetime.now()
dummy, last_date = fetch_last_updated(name)
recids = fetch_updated_arxiv_records(last_date)
updated_recids = set()
try:
for count, (recid, dummy) in enumerate(recids):
if count % 50 == 0:
msg = "Done %s of %s" % (count, len(recids))
write_message(msg)
task_update_progress(msg)
# BibTask sleep
task_sleep_now_if_required(can_stop_too=True)
write_message("processing %s" % recid, verbose=9)
try:
if process_one(recid):
updated_recids.add(recid)
time.sleep(6)
except AlreadyHarvested:
write_message("already harvested successfully")
time.sleep(6)
except FoundExistingPdf:
write_message("pdf already attached (matching md5)")
time.sleep(6)
except PdfNotAvailable:
write_message("no pdf available")
time.sleep(20)
except InvenioFileDownloadError, e:
write_message("failed to download: %s" % e)
time.sleep(20)
finally:
# We want to process updated records even in case we are interrupted
msg = "Updated %s records" % len(updated_recids)
write_message(msg)
task_update_progress(msg)
write_message(repr(updated_recids))
# For all updated records, we want to sync the 8564 tags
# and reextract references
if updated_recids:
submit_fixmarc_task(updated_recids)
submit_refextract_task(updated_recids)
# Store last run date of the daemon
# not if it ran on specific recids from the command line with --id
# but only if it ran on the modified records
if start_date:
store_last_updated(0, start_date, name)
return True
def task_run_core(name=NAME):
"""Entry point for the arxiv-pdf-checker task"""
# First gather recids to process
recids = task_get_option('recids')
if recids:
start_date = None
recids = [(recid, None) for recid in recids]
else:
start_date = datetime.now()
dummy, last_date = fetch_last_updated(name)
recids = fetch_updated_arxiv_records(last_date)
updated_recids = set()
try:
for count, (recid, dummy) in enumerate(recids):
if count % 50 == 0:
msg = 'Done %s of %s' % (count, len(recids))
write_message(msg)
task_update_progress(msg)
# BibTask sleep
task_sleep_now_if_required(can_stop_too=True)
write_message('processing %s' % recid, verbose=9)
try:
if process_one(recid):
updated_recids.add(recid)
time.sleep(6)
except AlreadyHarvested:
write_message('already harvested successfully')
time.sleep(6)
except FoundExistingPdf:
write_message('pdf already attached (matching md5)')
time.sleep(6)
except PdfNotAvailable:
write_message("no pdf available")
time.sleep(20)
except InvenioFileDownloadError, e:
write_message("failed to download: %s" % e)
time.sleep(20)
finally:
# We want to process updated records even in case we are interrupted
msg = 'Updated %s records' % len(updated_recids)
write_message(msg)
task_update_progress(msg)
write_message(repr(updated_recids))
# For all updated records, we want to sync the 8564 tags
# and reextract references
if updated_recids:
submit_fixmarc_task(updated_recids)
submit_refextract_task(updated_recids)
# Store last run date of the daemon
# not if it ran on specific recids from the command line with --id
# but only if it ran on the modified records
if start_date:
store_last_updated(0, start_date, name)
return True