def from_dict(dct: dict) -> Expungement:
dct.update({
"attorney": Attorney.from_dict(dct["attorney"]),
"client": Person.from_dict(dct["client"]),
"cases": [Case.from_dict(c) for c in dct["cases"]],
})
return Expungement(**dct)
def test_serializing_person(example_person):
ser = to_serializable(example_person)
assert ser["first_name"] == example_person.first_name
assert ser["aliases"] == example_person.aliases
deser = PersonSerializer(data=ser)
assert deser.is_valid(), deser.error_messages
deser = deser.validated_data
deser = Person.from_dict(deser)
assert isinstance(deser, Person)
assert deser == example_person
def test_person_from_dict(example_person):
ser = to_serializable(example_person)
per2 = Person.from_dict(ser)
assert example_person.last_name == per2.last_name
assert isinstance(per2.date_of_birth, date)
def parse_mdj_pdf_text(txt: str) -> Tuple[Person, List[Case], List[str]]:
"""
Parse MDJ docket, given the formatted text of the pdf.
This function uses the original Expungement Generator's technique: regexes and nested loops,
iterating over the lines of the docket.
see https://github.com/NateV/Expungement-Generator/blob/master/Expungement-Generator/Record.php:64
"""
already_searched_aliases = False
case_info = dict()
case_info["charges"] = []
person_info = dict()
person_info["aliases"] = []
lines = txt.split("\n")
for idx, line in enumerate(lines):
m = PATTERNS.mdj_district_number.search(line)
if m:
# what's the mdj district number for?
case_info["mdj_district_number"] = m.group(1)
m = PATTERNS.mdj_county_and_disposition.search(line)
if m:
case_info["county"] = m.group(1)
case_info["disposition_date"] = m.group(2)
m = PATTERNS.docket_number.search(line)
if m:
case_info["docket_number"] = m.group(1)
m = PATTERNS.otn.search(line)
if m:
case_info["otn"] = m.group(1)
m = PATTERNS.dc_number.search(line)
if m:
case_info["dc_num"] = m.group(1)
m = PATTERNS.arrest_agency_and_date.search(line)
if m:
case_info["arresting_agency"] = m.group(1)
try:
case_info["arrest_date"] = m.group(2)
except:
pass
m = PATTERNS.complaint_date.search(line)
if m:
case_info["complaint_date"] = m.group(1)
m = PATTERNS.affiant.search(line)
if m:
# TODO - mdj docket parse should reverse order of names of affiant
case_info["affiant"] = m.group(1)
# MHollander said:
# the judge name can appear in multiple places. Start by checking to see if the
# judge's name appears in the Judge Assigned field. If it does, then set it.
# Later on, we'll check in the "Final Issuing Authority" field. If it appears there
# and doesn't show up as "migrated," we'll reassign the judge name.
m = PATTERNS.judge_assigned.search(line)
if m:
judge = m.group(1).strip()
next_line = lines[idx + 1]
overflow_match = PATTERNS.judge_assigned_overflow.search(next_line)
if overflow_match:
judge = f"{judge} {overflow_match.group(1).strip()}"
if "igrated" not in judge:
case_info["judge"] = judge
m = PATTERNS.judge.search(line)
if m:
if len(m.group(1)) > 0 and "igrated" not in m.group(1):
case_info["judge"] = m.group(1)
m = PATTERNS.dob.search(line)
if m:
person_info["date_of_birth"] = m.group(1)
m = PATTERNS.name.search(line)
if m:
person_info["first_name"] = m.group(2)
person_info["last_name"] = m.group(1)
person_info["aliases"].append(f"{m.group(1)}, {m.group(2)}")
m = PATTERNS.alias_names_start.search(line)
if already_searched_aliases is False and m:
idx2 = idx + 1
end_of_aliases = False
while not end_of_aliases:
if PATTERNS.end_of_page.search(lines[idx2]):
continue
if re.search(r"\w", lines[idx2]):
person_info["aliases"].append(lines[idx2].strip())
idx2 += 1
end_of_aliases = PATTERNS.alias_names_end.search(lines[idx2])
if end_of_aliases:
already_searched_aliases = True
m = PATTERNS.charges.search(line) # Arrest.php;595
if m:
charge_info = dict()
charge_info["statute"] = m.group(1)
charge_info["grade"] = m.group(3)
charge_info["offense"] = m.group(4)
charge_info["disposition"] = m.group(6)
m2 = PATTERNS.charges_search_overflow.search(lines[idx + 1])
if m2:
charge_info[
"offense"
] = f"{charge_info['offense'].strip()} {m2.group(1).strip()}"
## disposition date is on the next line
if "disposition_date" in case_info.keys():
charge_info["disposition_date"] = case_info["disposition_date"]
case_info["charges"].append(charge_info)
m = PATTERNS.bail.search(line)
if m:
# TODO charges won't use the detailed bail info yet.
case_info["bail_charged"] = m.group(1)
case_info["bail_paid"] = m.group(2)
case_info["bail_adjusted"] = m.group(3)
case_info["bail_total"] = m.group(5)
m = PATTERNS.costs.search(line)
if m:
case_info["total_fines"] = m.group(1)
case_info["fines_paid"] = m.group(2)
case_info["costs_adjusted"] = m.group(3)
case_info["costs_total"] = m.group(5)
case_info = {
k: (v.strip() if isinstance(v, str) else v) for k, v in case_info.items()
}
person_info = {
k: (v.strip() if isinstance(v, str) else v) for k, v in person_info.items()
}
person = Person.from_dict(person_info)
case = Case.from_dict(case_info)
logger.info("Finished parsing MDJ docket")
return person, [case], []