def generate_data(self):
"""Make a CSV of the data extracted from the database.
CSV will have the following format:
Court, Name, Title, Count, 2000, 2011...
{
'ca2': {
"harold baller": {
"Mag judge": {
"years": {
"1999': 22,
"2000': 14,
},
'total count': 36,
},
}
}
}
"""
courts = Court.objects.filter(
jurisdiction__in=Court.FEDERAL_JURISDICTIONS,
)
out = {}
for court in courts:
out[court.pk] = {}
dockets = (court.dockets
.exclude(Q(assigned_to_str='') & Q(referred_to_str=''))
.filter(source__in=Docket.RECAP_SOURCES)
.only('assigned_to_str', 'referred_to_str',
'date_filed'))
logger.info("Processing %s dockets in %s" % (dockets.count(),
court.pk))
for docket in dockets:
for judge_type in ['assigned', 'referred']:
judge = getattr(docket, '%s_to_str' % judge_type)
if not judge:
continue
name, title = normalize_judge_string(unidecode(judge))
if not name:
continue
if name not in out[court.pk]:
# No entry for this person.
out[court.pk][name] = {
title: Counter([docket.date_filed.year]),
}
else:
# Person already exists.
if title not in out[court.pk][name]:
# Title not yet found.
out[court.pk][name][title] = Counter(
[docket.date_filed.year])
else:
# Title already exists.
out[court.pk][name][title][
docket.date_filed.year] += 1
self.export_files(out)
def generate_data(self):
"""Make a CSV of the data extracted from the database.
CSV will have the following format:
Court, Name, Title, Count, 2000, 2011...
{
'ca2': {
"harold baller": {
"Mag judge": {
"years": {
"1999': 22,
"2000': 14,
},
'total count': 36,
},
}
}
}
"""
courts = Court.objects.filter(
jurisdiction__in=Court.FEDERAL_JURISDICTIONS, )
out = {}
for court in courts:
out[court.pk] = {}
dockets = (court.dockets.exclude(
Q(assigned_to_str="") & Q(referred_to_str="")).filter(
source__in=Docket.RECAP_SOURCES).only(
"assigned_to_str", "referred_to_str", "date_filed"))
logger.info("Processing %s dockets in %s" %
(dockets.count(), court.pk))
for docket in dockets:
for judge_type in ["assigned", "referred"]:
judge = getattr(docket, "%s_to_str" % judge_type)
if not judge:
continue
name, title = normalize_judge_string(unidecode(judge))
if not name:
continue
if name not in out[court.pk]:
# No entry for this person.
out[court.pk][name] = {
title: Counter([docket.date_filed.year]),
}
else:
# Person already exists.
if title not in out[court.pk][name]:
# Title not yet found.
out[court.pk][name][title] = Counter(
[docket.date_filed.year])
else:
# Title already exists.
out[court.pk][name][title][
docket.date_filed.year] += 1
self.export_files(out)
def test_title_name_splitter(self):
pairs = [
{
"q": "Magistrate Judge George T. Swartz",
"a": ("George T. Swartz", "mag"),
},
{"q": "J. Frederick Motz", "a": ("Frederick Motz", "jud"),},
{
"q": "Honorable Susan W. Wright",
"a": ("Susan W. Wright", "jud"),
},
]
for pair in pairs:
self.assertEqual(pair["a"], normalize_judge_string(pair["q"]))
def test_title_name_splitter(self):
pairs = [{
'q': 'Magistrate Judge George T. Swartz',
'a': ('George T. Swartz', 'mag'),
},
{
'q': 'J. Frederick Motz',
'a': ('Frederick Motz', 'jud'),
},
{
'q': 'Honorable Susan W. Wright',
'a': ('Susan W. Wright', 'jud'),
},
]
for pair in pairs:
self.assertEqual(pair['a'], normalize_judge_string(pair['q']))