def add_or_update(self, *items):
"""
Given an item, adds it to the index, or updates it if it's already
in the index.
"""
self.stdout.write("Adding or updating item(s): %s\n" % list(items))
# Use Celery to add or update the item asynchronously
if self.type == Opinion:
add_or_update_opinions.delay(items)
elif self.type == Audio:
add_or_update_audio_files.delay(items)
elif self.type == Person:
add_or_update_people.delay(items)
def add_or_update(self, *items):
"""
Given an item, adds it to the index, or updates it if it's already
in the index.
"""
self.stdout.write("Adding or updating item(s): %s\n" % list(items))
# Use Celery to add or update the item asynchronously
if self.type == Opinion:
add_or_update_opinions.delay(items)
elif self.type == Audio:
add_or_update_audio_files.delay(items)
elif self.type == Person:
add_or_update_people.delay(items)
def delete_model(self, request, obj):
obj.delete()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.person_id])
def save_model(self, request, obj, form, change):
obj.save()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.person_id])
def save_model(self, request, obj, form, change):
obj.save()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.person_id], force_commit=False)
def fix_fjc_positions(self, infile=None):
"""
Addresses issue #624.
We had some errant regexes in the district court assignments. This code
reassigns the court fields for these judges where the new regexes
differs from the old ones.
:param infile: The import file with fjc-data.xslx
:return: None
"""
if infile is None:
self.ensure_input_file()
infile = self.options['input_file']
textfields = [
'firstname', 'midname', 'lastname', 'gender',
'Place of Birth (City)', 'Place of Birth (State)',
'Place of Death (City)', 'Place of Death (State)'
]
df = pd.read_excel(infile, 0)
for x in textfields:
df[x] = df[x].replace(np.nan, '', regex=True)
df['Employment text field'].replace(to_replace=r';\sno',
value=r', no',
inplace=True,
regex=True)
for i, item in df.iterrows():
fjc_id = item['Judge Identification Number']
p = Person.objects.get(fjc_id=fjc_id)
logger.info("Doing person with FJC ID: %s, "
"https://courtlistener.com%s" %
(fjc_id, p.get_absolute_url()))
exclusions = []
for posnum in range(1, 7):
if posnum > 1:
pos_str = ' (%s)' % posnum
else:
pos_str = ''
if pd.isnull(item['Court Name' + pos_str]):
continue
courtid = get_fed_court_object(item['Court Name' + pos_str])
if courtid is None:
raise
date_termination = process_date_string(
item['Date of Termination' + pos_str])
date_start = process_date_string(item['Commission Date' +
pos_str])
date_recess_appointment = process_date_string(
item['Recess Appointment date' + pos_str])
if pd.isnull(
date_start) and not pd.isnull(date_recess_appointment):
date_start = date_recess_appointment
if pd.isnull(date_start):
# if still no start date, skip
continue
positions = (Position.objects.filter(
person=p,
date_start=date_start,
date_termination=date_termination,
position_type='jud').exclude(pk__in=exclusions))
position_count = positions.count()
if position_count < 1:
logger.info("Couldn't find position to match '%s' on '%s' "
"with exclusions: %s" %
(p, date_start, exclusions))
add_positions_from_row(item,
p,
self.debug,
fix_nums=[posnum])
if not self.debug:
add_or_update_people.delay([p.pk])
continue
elif position_count == 1:
# Good case. Press on!
position = positions[0]
exclusions.append(position.pk)
elif position_count > 1:
logger.info(
"Got too many results for '%s' on '%s'. Got %s" %
(p, date_start, position_count))
continue
if position.court.pk == courtid:
logger.info("Court IDs are both '%s'. No changes made." %
courtid)
else:
logger.info("Court IDs are different! Old: %s, New: %s" %
(position.court.pk, courtid))
court = Court.objects.get(pk=courtid)
position.court = court
if not self.debug:
position.save()
add_or_update_people.delay([p.pk])
def save_model(self, request, obj, form, change):
obj.save()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.pk], force_commit=False)
def fix_fjc_positions(self, infile=None):
"""
Addresses issue #624.
We had some errant regexes in the district court assignments. This code
reassigns the court fields for these judges where the new regexes
differs from the old ones.
:param infile: The import file with fjc-data.xslx
:return: None
"""
if infile is None:
self.ensure_input_file()
infile = self.options['input_file']
textfields = ['firstname', 'midname', 'lastname', 'gender',
'Place of Birth (City)', 'Place of Birth (State)',
'Place of Death (City)', 'Place of Death (State)']
df = pd.read_excel(infile, 0)
for x in textfields:
df[x] = df[x].replace(np.nan, '', regex=True)
df['Employment text field'].replace(to_replace=r';\sno', value=r', no',
inplace=True, regex=True)
for i, item in df.iterrows():
fjc_id = item['Judge Identification Number']
p = Person.objects.get(fjc_id=fjc_id)
print(
"Doing person with FJC ID: %s, https://courtlistener.com%s" %
(fjc_id, p.get_absolute_url())
)
exclusions = []
for posnum in range(1, 7):
if posnum > 1:
pos_str = ' (%s)' % posnum
else:
pos_str = ''
if pd.isnull(item['Court Name' + pos_str]):
continue
courtid = get_fed_court_object(item['Court Name' + pos_str])
if courtid is None:
raise
date_termination = process_date_string(
item['Date of Termination' + pos_str])
date_start = process_date_string(
item['Commission Date' + pos_str])
date_recess_appointment = process_date_string(
item['Recess Appointment date' + pos_str])
if pd.isnull(date_start) and not pd.isnull(
date_recess_appointment):
date_start = date_recess_appointment
if pd.isnull(date_start):
# if still no start date, skip
continue
positions = (Position.objects
.filter(person=p, date_start=date_start,
date_termination=date_termination,
position_type='jud')
.exclude(pk__in=exclusions))
position_count = positions.count()
if position_count < 1:
print "Couldn't find position to match '%s' on '%s' with " \
"exclusions: %s" % (p, date_start, exclusions)
add_positions_from_row(item, p, self.debug,
fix_nums=[posnum])
if not self.debug:
add_or_update_people.delay([p.pk])
continue
elif position_count == 1:
# Good case. Press on!
position = positions[0]
exclusions.append(position.pk)
elif position_count > 1:
print "Got too many results for '%s' on '%s'. Got %s" % \
(p, date_start, position_count)
continue
if position.court.pk == courtid:
print "Court IDs are both '%s'. No changes made." % courtid
else:
print "Court IDs are different! Old: %s, New: %s" % (
position.court.pk, courtid)
court = Court.objects.get(pk=courtid)
position.court = court
if not self.debug:
position.save()
add_or_update_people.delay([p.pk])
def delete_model(self, request, obj):
obj.delete()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.person_id])
def save_model(self, request, obj, form, change):
obj.save()
from cl.search.tasks import add_or_update_people
add_or_update_people.delay([obj.pk])
