def test_parse_plea_type_nolo_contendere(self):
plea1 = 'PLEA OF NOLO CONTENDERE'
plea2 = "DEFENDANT ENTERED PLEA OF : NOLO-CONTENDERE SEQ 2"
plea3 = "DEFENDANT ENTERED PLEA OF NOLO CONTENDERE SEQ: 1,2,3,4,5"
assert ScraperUtils.parse_plea_type(plea1) == 'Nolo Contendere'
assert ScraperUtils.parse_plea_type(plea2) == 'Nolo Contendere'
assert ScraperUtils.parse_plea_type(plea3) == 'Nolo Contendere'
def test_parse_plea_type_not_guilty(self):
plea1 = 'PLEA OF NOT GUILTY'
plea2 = "PLEA OF NOT GUILTY/DENIAL, WAIVER OF ARRAIGNMENT, DEMAND FOR NOTICE OF EXPERT TESTIMONY, DEMAND FOR DISCOVERY, DEMAND FOR STATEMENT OF PARTICULARS, DEMAND FOR JURY TRIAL, DESIGNATION OF E-MAIL ADDRESSES PURSUANT TO RULE 2.516 1/28/2020"
plea3 = 'EP - NOTICE OF APPEARANCE, AND ENTRY OF CONDITIONAL PLEA OF NOT GUILTY AND DEMAND FOR JURY TRIAL'
assert ScraperUtils.parse_plea_type(plea1) == 'Not Guilty'
assert ScraperUtils.parse_plea_type(plea2) == 'Not Guilty'
assert ScraperUtils.parse_plea_type(plea3) == 'Not Guilty'
def test_parse_defense_attorneys_invalid(self):
invalid_test = ['', '', '']
invalid_test2 = []
invalid_test3 = None
assert ScraperUtils.parse_attorneys(invalid_test) is None
assert ScraperUtils.parse_attorneys(invalid_test2) is None
assert ScraperUtils.parse_attorneys(invalid_test3) is None
def test_parse_charge_statute_incomplete(self):
charge1 = '(32234 2a)'
charge2 = 'DRIVING WHILE LICENSE SUSPENDED OR REVOKED'
assert ScraperUtils.parse_charge_statute(charge1) == (None, '32234 2a')
assert ScraperUtils.parse_charge_statute(charge2) == (charge2, None)
assert ScraperUtils.parse_charge_statute(' ') == (None, None)
assert ScraperUtils.parse_charge_statute(None) == (None, None)
def test_parse_charge_statute(self):
charge1 = ' FLEEING OR ATTEMPTING TO ELUDE (HIGH SPEED RECKLESS) (3161935 3) '
charge2 = 'DRIVING WHILE LICENSE SUSPENDED OR REVOKED (32234 2a)'
charge3 = ' FELON IN POSSESSION OF AMMUNITION (ACTUAL POSSESSION) (79023) '
charge4 = 'FAIL TO DISPLAY REGISTRATION - POSSESSION REQUIRED (320.0605(1)) '
assert ScraperUtils.parse_charge_statute(charge1) == (
'FLEEING OR ATTEMPTING TO ELUDE (HIGH SPEED RECKLESS)',
'3161935 3')
assert ScraperUtils.parse_charge_statute(charge2) == (
'DRIVING WHILE LICENSE SUSPENDED OR REVOKED', '32234 2a')
assert ScraperUtils.parse_charge_statute(charge3) == (
'FELON IN POSSESSION OF AMMUNITION (ACTUAL POSSESSION)', '79023')
assert ScraperUtils.parse_charge_statute(charge4) == (
'FAIL TO DISPLAY REGISTRATION - POSSESSION REQUIRED',
'320.0605(1)')
def test_parse_defense_attorneys(self):
attorneys1 = [
'DEFENSE ATTORNEY: DOE, JANE EMILY ASSIGNED',
'DEFENSE ATTORNEY: DOE, JOHN MICHAEL ASSIGNED',
'DEFENSE ATTORNEY: SELF, SELF ASSIGNED'
]
public_defenders1 = [
'COURT APPOINTED ATTORNEY: DOE, JOHN MICHAEL ASSIGNED'
]
assert ScraperUtils.parse_attorneys(attorneys1) == [
'DOE, JANE EMILY', 'DOE, JOHN MICHAEL', 'SELF, SELF'
]
assert ScraperUtils.parse_attorneys(public_defenders1) == [
'DOE, JOHN MICHAEL'
]
def test_parse_out_path_valid(self):
# Function should not affect valid length filenames and paths.
normal_filename = 'document'
parsedPath = ScraperUtils.parse_out_path('C:\\Example\Path',
normal_filename, 'pdf')
assert parsedPath == os.path.join(
'C:\\Example\Path', '{}.{}'.format(normal_filename, 'pdf'))
def begin_scrape():
"""
Starts the scraping process. Continues from the last scraped record if the scraper was stopped before.
:return:
"""
global driver
# Find the progress of any past scraping runs to continue from then
try:
last_case_number = ScraperUtils.get_last_csv_row(output_file).split(',')[3]
print("Continuing from last scrape (Case number: {})".format(last_case_number))
last_year = 2000 + int(str(last_case_number)[:2]) # I know there's faster ways of doing this. It only runs once ;)
if not last_case_number.isnumeric():
last_case_number = last_case_number[:-4]
last_case = int(str(last_case_number)[-6:])
settings['end-year'] = last_year
continuing = True
except FileNotFoundError:
# No existing scraping CSV
continuing = False
pass
# Scrape from the most recent year to the oldest.
for year in range(settings['end-year'], settings['start-year'], -1):
if continuing:
N = last_case + 1
else:
N = 1
print("Scraping year {} from case {}".format(year, N))
YY = year % 100
record_missing_count = 0
# Increment case numbers until the threshold missing cases is met, then advance to the next year.
while record_missing_count < settings['missing-thresh']:
# Generate the case number to scrape
case_number = f'{YY:02}' + f'{N:06}'
search_result = search_portal(case_number)
if search_result:
record_missing_count = 0
# if multiple associated cases are found,
# scrape all of them
if len(search_result) > 1:
for case in search_result:
search_portal(case)
scrape_record(case)
# only a single case, no multiple associated cases found
else:
scrape_record(case_number)
else:
record_missing_count += 1
N += 1
continuing = False
print("Scraping for year {} is complete".format(year))
def test_parse_out_path_shortening(self):
# 260 characters long before the extension. This is an invalid filename in Windows.
filename_too_long = '01234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789'
parsed_path = ScraperUtils.parse_out_path(os.getcwd(),
filename_too_long, 'txt')
try:
open(parsed_path, 'w')
os.remove(parsed_path)
except OSError:
pytest.fail('parse_out_path() generates an invalid file path.')
def test_parse_plea_case_numbers_no_charge_numbers(self):
plea = "DEFENDANT ENTERED PLEA OF NOLO CONTENDERE SEQ: 1,2,3,4,5"
assert ScraperUtils.parse_plea_case_numbers(plea, []) == []
def test_parse_out_path_illegal_characters(self):
filename_invalid_chars = 't<>:e"/s\\t|?n*ame'
assert ScraperUtils.parse_out_path('', filename_invalid_chars,
'pdf') == os.path.join(
'', 'testname.pdf')
def test_parse_plea_case_numbers_blank(self):
assert ScraperUtils.parse_plea_case_numbers("", [1, 2, 3]) == []
def test_filename_blank(self):
filename = None
parsed_path = ScraperUtils.parse_out_path(r'C:\\Example\\Path',
filename, 'pdf')
assert parsed_path == os.path.join(r'C:\\Example\\Path', '.pdf')
def test_parse_name(self):
name1 = "DOE, JANE EMILY"
name2 = "DOE, JOHN"
assert ScraperUtils.parse_name(name1) == ('JANE', 'EMILY', 'DOE')
assert ScraperUtils.parse_name(name2) == ('JOHN', None, 'DOE')
def test_parse_plea_type_guilty(self):
plea1 = 'PLEA OF GUILTY'
plea2 = 'DEFENDANT ENTERED PLEA OF : GUILTY SEQ 2'
assert ScraperUtils.parse_plea_type(plea1) == 'Guilty'
assert ScraperUtils.parse_plea_type(plea2) == 'Guilty'
def test_parse_name_error(self):
name1 = None
name2 = ''
assert ScraperUtils.parse_name(name1) == (None, None, None)
assert ScraperUtils.parse_name(name2) == (None, None, None)
def search_portal(case_number):
"""
Performs a search of the portal from its home page, including selecting the case number input, solving the captcha
and pressing Search. Also handles the captcha being solved incorrectly
:param case_number: Case to search
:return: A set of case number(s).
"""
# Load portal search page
load_page(f"{settings['portal-base']}/Home.aspx/Search", 'Search',
settings['verbose'])
# Give some time for the captcha to load, as it does not load instantly.
time.sleep(0.8)
# Select Case Number textbox and enter case number
select_case_input()
case_input = driver.find_element_by_id('caseNumber')
case_input.click()
case_input.send_keys(case_number)
# Solve captcha if it is required
try:
# Get Captcha. This is kinda nasty, but if there's no Captcha, then
# this will throw (which is a good thing in this case) and we can
# move on with processing.
captcha_image_elem = driver.find_element_by_xpath(
'//*/img[@alt="Captcha"]')
captcha_buffer = captcha_image_elem.screenshot_as_png
if settings['solve-captchas']:
solved_captcha = captcha_solver.solve_captcha(captcha_buffer)
captcha_textbox = driver.find_element_by_xpath(
'//*/input[@name="captcha"]')
captcha_textbox.click()
captcha_textbox.send_keys(solved_captcha.answer)
# Do search
search_button = driver.find_element_by_id('searchButton')
search_button.click()
else:
print(f"Captcha encountered trying to view case ID {case_number}.")
print(
"Please solve the captcha and click the search button to proceed."
)
while True:
try:
WebDriverWait(
driver, 6 * 60 *
60).until(lambda x: case_number in driver.title)
print("continuing...")
break
except TimeoutException:
print("still waiting for user to solve the captcha...")
except NoSuchElementException:
# No captcha on the page, continue.
solved_captcha = None
# Do search
search_button = driver.find_element_by_id('searchButton')
search_button.click()
# If the title stays as 'Search': Captcha solving failed
# If the title contains the case number or 'Search Results': Captcha solving succeeded
# If a timeout occurs, retry 'connect-thresh' times.
for i in range(settings['connect-thresh']):
try:
# Wait for page to load
WebDriverWait(
driver,
5).until(lambda x: 'Search' in driver.title or case_number in
driver.title or 'Search Results:' in driver.title)
# Page loaded
if driver.title == 'Search':
# Clicking search did not change the page. This could be because of a failed captcha attempt.
try:
# Check if 'Invalid Captcha' dialog is showing
driver.find_element_by_xpath(
'//div[@class="alert alert-error"]')
print("Captcha was solved incorrectly")
if settings['solve-captchas'] and solved_captcha:
solved_captcha.notify_incorrect()
except NoSuchElementException:
pass
# Clear cookies so a new captcha is presented upon refresh
driver.delete_all_cookies()
# Try solving the captcha again.
search_portal(case_number)
elif 'Search Results: CaseNumber:' in driver.title:
# Captcha solved correctly
if settings['solve-captchas'] and solved_captcha:
solved_captcha.notify_correct()
case_count = ScraperUtils.get_search_case_count(
driver, settings['county'])
# Case number search found multiple cases.
if case_count > 1:
return ScraperUtils.get_associated_cases(driver)
# Case number search found no cases
else:
return set()
elif case_number in driver.title:
# Captcha solved correctly
if settings['solve-captchas'] and solved_captcha:
solved_captcha.notify_correct()
# Case number search did find a single court case.
return {case_number}
except TimeoutException:
if i == settings['connect-thresh'] - 1:
raise RuntimeError(
'Case page could not be loaded after {} attempts, or unexpected page title: {}'
.format(settings['connect-thresh'], driver.title))
else:
search_portal(case_number)
def search_portal(case_number):
"""
Performs a search of the portal from its home page, including selecting the case number input, solving the captcha
and pressing Search. Also handles the captcha being solved incorrectly
:param case_number: Case to search
:return: A set of case number(s).
"""
# Load portal search page
load_page(f"{settings['portal-base']}/Home.aspx/Search", 'Search', settings['verbose'])
# Give some time for the captcha to load, as it does not load instantly.
time.sleep(0.8)
# Select Case Number textbox and enter case number
select_case_input()
case_input = driver.find_element_by_id('caseNumber')
case_input.click()
case_input.send_keys(case_number)
if settings['solve-captchas']:
# Solve captcha if it is required
try:
# Get Captcha
captcha_image_elem = driver.find_element_by_xpath(
'//*/img[@alt="Captcha"]')
captcha_buffer = captcha_image_elem.screenshot_as_png
captcha_answer = captcha_solver.solve_captcha(captcha_buffer)
captcha_textbox = driver.find_element_by_xpath(
'//*/input[@name="captcha"]')
captcha_textbox.click()
captcha_textbox.send_keys(captcha_answer)
except NoSuchElementException:
# No captcha on the page, continue.
pass
# Do search
search_button = driver.find_element_by_id('searchButton')
search_button.click()
else:
raise Exception("Automated captcha solving is disabled by default. Please seek advice before using this feature.")
# If the title stays as 'Search': Captcha solving failed
# If the title contains the case number or 'Search Results': Captcha solving succeeded
# If a timeout occurs, retry 'connect-thresh' times.
for i in range(settings['connect-thresh']):
try:
# Wait for page to load
WebDriverWait(driver, 5).until(
lambda x: 'Search' in driver.title or case_number in driver.title or 'Search Results:' in driver.title)
# Page loaded
if driver.title == 'Search':
# Clicking search did not change the page. This could be because of a failed captcha attempt.
try:
# Check if 'Invalid Captcha' dialog is showing
driver.find_element_by_xpath(
'//div[@class="alert alert-error"]')
print("Captcha was solved incorrectly")
captcha_solver.notify_last_captcha_fail()
except NoSuchElementException:
pass
# Clear cookies so a new captcha is presented upon refresh
driver.delete_all_cookies()
# Try solving the captcha again.
search_portal(case_number)
elif 'Search Results: CaseNumber:' in driver.title:
# Captcha solved correctly
captcha_solver.notify_last_captcha_success()
# Figure out the numer of cases returned
case_detail_tbl = driver.find_element_by_tag_name('table').text.split('\n')
case_count_idx = case_detail_tbl.index('CASES FOUND') + 1
case_count = int(case_detail_tbl[case_count_idx])
# Case number search found multiple cases.
if case_count > 1:
return ScraperUtils.get_associated_cases(driver)
# Case number search found no cases
else:
return set()
elif case_number in driver.title:
# Captcha solved correctly
captcha_solver.notify_last_captcha_success()
# Case number search did find a single court case.
return {case_number}
except TimeoutException:
if i == settings['connect-thresh'] - 1:
raise RuntimeError('Case page could not be loaded after {} attempts, or unexpected page title: {}'.format(settings['connect-thresh'], driver.title))
else:
search_portal(case_number)
def test_parse_plea_case_numbers__no_charge_mentioned(self):
plea = "PLEA OF NOT GUILTY"
assert ScraperUtils.parse_plea_case_numbers(plea, [1]) == []
def test_parse_plea_case_numbers_one_charge_mentioned(self):
plea = "DEFENDANT ENTERED PLEA OF : NOLO-CONTENDERE SEQ 2"
assert ScraperUtils.parse_plea_case_numbers(plea, [1, 2]) == [2]
def test_parse_out_path_correct_length(self):
# 260 characters long before the extension. This is an invalid filename in Windows.
filename_too_long = '01234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789'
parsed_path = ScraperUtils.parse_out_path(os.getcwd(),
filename_too_long, 'txt')
assert len(parsed_path) <= 256
def test_parse_out_path_filename_extension_shortening(self):
# 252 characters long, but with the .pdf extension it becomes 256 characters long - one too many.
filename = '012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901234567890123456789012345678901'
parsed_filename = ScraperUtils.parse_out_path('', filename, 'pdf')
assert parsed_filename == '{}.{}'.format(filename[:-1], 'pdf')
def test_parse_plea_case_numbers_multiple_case_numbers(self):
plea = "DEFENDANT ENTERED PLEA OF NOLO CONTENDERE SEQ: 1,2,3,4,5"
assert ScraperUtils.parse_plea_case_numbers(
plea, [1, 2, 3, 4, 5, 6]) == [1, 2, 3, 4, 5]
def scrape_record(case_number):
"""
Scrapes a record once the case has been opened.
:param case_number: The current case's case number.
"""
# Wait for court summary to load
for i in range(settings['connect-thresh']):
try:
WebDriverWait(driver, 5).until(EC.presence_of_element_located((By.ID, 'summaryAccordion')))
except TimeoutException:
if i == settings['connect-thresh'] - 1:
raise RuntimeError('Summary details did not load for case {}.'.format(case_number))
else:
driver.refresh()
# Get relevant page content
summary_table_col1 = driver.find_elements_by_xpath('//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[1]/dl/dd')
summary_table_col2 = driver.find_elements_by_xpath('//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[2]/dl/dd')
summary_table_col3 = driver.find_elements_by_xpath('//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[3]/dl/dd')
# Wait for court dockets to load
for i in range(settings['connect-thresh']):
try:
WebDriverWait(driver, 5).until(EC.presence_of_element_located((By.ID, 'gridDocketsView')))
except TimeoutException:
if i == settings['connect-thresh'] - 1:
raise RuntimeError('Dockets did not load for case {}.'.format(case_number))
else:
driver.refresh()
charges_table = driver.find_elements_by_xpath('//*[@id="gridCharges"]/tbody/tr')
docket_public_defender = driver.find_elements_by_xpath(
"//*[contains(text(), 'COURT APPOINTED ATTORNEY') and contains(text(), 'ASSIGNED')]")
docket_attorney = driver.find_elements_by_xpath("//*[contains(text(), 'DEFENSE') and contains(text(), 'ASSIGNED')]")
docket_pleas = driver.find_elements_by_xpath("//*[contains(text(), 'PLEA OF')]")
docket_attachments = driver.find_elements_by_class_name('casedocketimage')
_id = str(uuid.uuid4())
_state = settings['state-code']
_county = settings['county']
CaseNum = summary_table_col2[1].text.strip()
AgencyReportNum = summary_table_col1[4].text.strip()
ArrestDate = None # Can't be found on this portal
FilingDate = summary_table_col1[2].text.strip()
OffenseDate = None # Can't be found on this portal
DivisionName = summary_table_col3[3].text.strip()
CaseStatus = summary_table_col3[1].text.strip()
if settings['collect-pii']:
# Create list of assigned defense attorney(s)
defense_attorney_text = list(map(lambda x: x.text, docket_attorney))
DefenseAttorney = ScraperUtils.parse_attorneys(defense_attorney_text)
# Create list of assigned public defenders / appointed attorneys
public_defender_text = list(map(lambda x: x.text, docket_public_defender))
PublicDefender = ScraperUtils.parse_attorneys(public_defender_text)
# Get Judge
Judge = summary_table_col1[0].text.strip()
# Download docket attachments.
# Todo(OscarVanL): This could be parallelized to speed up scraping if save-attachments is set to 'all'.
if settings['save-attachments']:
for attachment_link in docket_attachments:
attachment_text = attachment_link.find_element_by_xpath('./../../td[3]').text.strip()
if settings['save-attachments'] == 'filing':
if not ('CITATION FILED' in attachment_text or 'CASE FILED' in attachment_text):
# Attachment is not a filing, don't download it.
continue
ScraperUtils.save_attached_pdf(driver, output_attachments, '{}-{}'.format(case_number, attachment_text),
settings['portal-base'], attachment_link, 20, settings['verbose'])
else:
DefenseAttorney = []
PublicDefender = []
Judge = None
Charges = {}
for charge in charges_table:
charge_details = charge.find_elements_by_tag_name('td')
count = int(charge_details[0].text.strip())
long_desc = charge_details[1].text.strip()
# Statute is contained within brackets
if '(' in long_desc and ')' in long_desc:
statute = long_desc[long_desc.find('(') + 1:long_desc.find(')')]
else:
statute = None
description = long_desc.split('(')[0]
level = charge_details[2].text.strip()
degree = charge_details[3].text.strip()
# plea = charge_details[4].text.strip() # Plea is not filled out on this portal.
disposition = charge_details[5].text.strip()
disposition_date = charge_details[6].text.strip()
offense_date = None # Not shown on this portal
citation_number = None # Not shown on this portal
Charges[count] = Charge(count, statute, description, level, degree, disposition, disposition_date, offense_date,
citation_number, None, None)
# Pleas are not in the 'plea' field, but instead in the dockets.
for plea_element in docket_pleas:
plea_text = plea_element.text.strip()
plea = ScraperUtils.parse_plea_type(plea_text)
plea_date = plea_element.find_element_by_xpath('./../td[2]').text.strip()
plea_number = ScraperUtils.parse_plea_case_numbers(plea_text, list(Charges.keys()))
# If no case number is specified in the plea, then we assume it applies to all charges in the trial.
if len(plea_number) == 0:
for charge in Charges.values():
charge.plea = plea
charge.plea_date = plea_date
else:
# Apply plea to relevant charge count(s).
for count in plea_number:
Charges[count].plea = plea
Charges[count].plea_date = plea_date
ArrestingOfficer = None # Can't be found on this portal
ArrestingOfficerBadgeNumber = None # Can't be found on this portal
profile_link = driver.find_element_by_xpath("//table[@id='gridParties']/tbody/tr/*[contains(text(), 'DEFENDANT')]/../td[2]/div/a").get_attribute(
'href')
# profile_link = driver.find_element_by_xpath('//*[@id="gridParties"]/tbody/tr[1]/td[2]/div[1]/a').get_attribute(
# 'href')
load_page(profile_link, 'Party Details:', settings['verbose'])
Suffix = None
DOB = None # This portal has DOB as N/A for every defendent
Race = driver.find_element_by_xpath(
'//*[@id="fd-table-2"]/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[7]/td[2]').text.strip()
Sex = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[6]/td[2]').text.strip()
FirstName = None
MiddleName = None
LastName = None
PartyID = None
# Only collect PII if configured
if settings['collect-pii']:
# Navigate to party profile
full_name = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[1]/td[2]').text.strip()
MiddleName = None
LastName = None
if ',' in full_name:
name_split = full_name.split(',')[1].lstrip().split()
FirstName = name_split[0]
MiddleName = " ".join(name_split[1:])
LastName = full_name.split(',')[0]
else:
# If there's no comma, it's a corporation name.
FirstName = full_name
PartyID = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[8]/td[2]').text.strip() # PartyID is a field within the portal system to uniquely identify defendants
record = Record(_id, _state, _county, case_number, CaseNum, AgencyReportNum, PartyID, FirstName, MiddleName,
LastName, Suffix, DOB, Race, Sex, ArrestDate, FilingDate, OffenseDate, DivisionName, CaseStatus,
DefenseAttorney, PublicDefender, Judge, list(Charges.values()), ArrestingOfficer,
ArrestingOfficerBadgeNumber)
ScraperUtils.write_csv(output_file, record, settings['verbose'])
def test_parse_plea_case_numbers_messy(self):
# Test a really ugly plea docket I found in one case
plea = "PLEA OF NOT GUILTY/DENIAL, WAIVER OF ARRAIGNMENT, DEMAND FOR NOTICE OF EXPERT TESTIMONY, DEMAND FOR DISCOVERY, DEMAND FOR STATEMENT OF PARTICULARS, DEMAND FOR JURY TRIAL, DESIGNATION OF E-MAIL ADDRESSES PURSUANT TO RULE 2.516 1/28/2020"
assert ScraperUtils.parse_plea_case_numbers(plea, [1, 2, 3]) == []
def test_parse_plea_type_blank(self):
assert ScraperUtils.parse_plea_type('') is None
def scrape_record(case_number):
"""
Scrapes a record once the case has been opened.
:param case_number: The current case's case number.
"""
# Wait for court summary to load
for i in range(FLAGS.connect_thresh):
try:
WebDriverWait(driver, 5).until(
EC.presence_of_element_located((By.ID, 'summaryAccordion')))
except TimeoutException:
if i == FLAGS.connect_thresh - 1:
raise RuntimeError(
'Summary details did not load for case {}.'.format(
case_number))
else:
driver.refresh()
# Get relevant page content
summary_table_col1 = driver.find_elements_by_xpath(
'//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[1]/dl/dd')
summary_table_col2 = driver.find_elements_by_xpath(
'//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[2]/dl/dd')
summary_table_col3 = driver.find_elements_by_xpath(
'//*[@id="summaryAccordionCollapse"]/table/tbody/tr/td[3]/dl/dd')
# Wait for court dockets to load
for i in range(FLAGS.connect_thresh):
try:
WebDriverWait(driver, 5).until(
EC.presence_of_element_located((By.ID, 'gridDocketsView')))
except TimeoutException:
if i == FLAGS.connect_thresh - 1:
raise RuntimeError(
'Dockets did not load for case {}.'.format(case_number))
else:
driver.refresh()
charges_table = driver.find_elements_by_xpath(
'//*[@id="gridCharges"]/tbody/tr')
docket_public_defender = driver.find_elements_by_xpath(
"//*[contains(text(), 'COURT APPOINTED ATTORNEY') and contains(text(), 'ASSIGNED')]"
)
docket_attorney = driver.find_elements_by_xpath(
"//*[contains(text(), 'DEFENSE') and contains(text(), 'ASSIGNED')]")
docket_pleas = driver.find_elements_by_xpath(
"//*[contains(text(), 'PLEA OF')]")
docket_attachments = driver.find_elements_by_class_name('casedocketimage')
r = BenchmarkRecordBuilder()
r.id = str(uuid.uuid4())
r.state = FLAGS.state
r.county = FLAGS.county
r.portal_id = case_number
r.case_num = Pii.String(summary_table_col2[1].text.strip())
r.agency_report_num = summary_table_col1[4].text.strip()
r.arrest_date = None # Can't be found on this portal
r.filing_date = summary_table_col1[2].text.strip()
r.offense_date = None # Can't be found on this portal
r.division_name = summary_table_col3[3].text.strip()
r.case_status = summary_table_col3[1].text.strip()
# Create list of assigned defense attorney(s)
defense_attorney_text = list(map(lambda x: x.text, docket_attorney))
r.defense_attorney = ScraperUtils.parse_attorneys(defense_attorney_text)
# Create list of assigned public defenders / appointed attorneys
public_defender_text = list(map(lambda x: x.text, docket_public_defender))
r.public_defender = ScraperUtils.parse_attorneys(public_defender_text)
# Get Judge
r.judge = Pii.String(summary_table_col1[0].text.strip())
# Download docket attachments.
# Todo(OscarVanL): This could be parallelized to speed up scraping if save-attachments is set to 'all'.
if FLAGS.save_attachments:
for attachment_link in docket_attachments:
attachment_text = attachment_link.find_element_by_xpath(
'./../../td[3]').text.strip()
if FLAGS.save_attachments == 'filing':
if not ('CITATION FILED' in attachment_text
or 'CASE FILED' in attachment_text):
# Attachment is not a filing, don't download it.
continue
ScraperUtils.save_attached_pdf(
driver, output_attachments,
'{}-{}'.format(case_number, attachment_text),
FLAGS.portal_base, attachment_link, 20, FLAGS.verbose)
Charges = {}
for charge in charges_table:
charge_builder = ChargeBuilder()
charge_cols = charge.find_elements_by_tag_name('td')
count = int(charge_cols[0].text.strip())
charge_builder.count = count
charge_desc = charge_cols[1].text
charge_builder.description, charge_builder.statute = (
ScraperUtils.parse_charge_statute(charge_desc))
charge_builder.level = charge_cols[2].text.strip()
charge_builder.degree = charge_cols[3].text.strip()
# plea = charge_cols[4].text.strip() # Plea is not filled out on this portal.
charge_builder.disposition = charge_cols[5].text.strip()
charge_builder.disposition_date = charge_cols[6].text.strip()
Charges[count] = charge_builder.build()
r.charges = list(Charges.values())
# Pleas are not in the 'plea' field, but instead in the dockets.
for plea_element in docket_pleas:
plea_text = plea_element.text.strip()
plea = ScraperUtils.parse_plea_type(plea_text)
plea_date = plea_element.find_element_by_xpath(
'./../td[2]').text.strip()
plea_number = ScraperUtils.parse_plea_case_numbers(
plea_text, list(Charges.keys()))
# If no case number is specified in the plea, then we assume it applies to all charges in the trial.
if len(plea_number) == 0:
for charge in Charges.values():
charge.plea = plea
charge.plea_date = plea_date
else:
# Apply plea to relevant charge count(s).
for count in plea_number:
Charges[count].plea = plea
Charges[count].plea_date = plea_date
r.arresting_officer = None # Can't be found on this portal
r.arresting_officer_badge_number = None # Can't be found on this portal
profile_link = driver.find_element_by_xpath(
"//table[@id='gridParties']/tbody/tr/*[contains(text(), 'DEFENDANT')]/../td[2]/div/a"
).get_attribute('href')
# profile_link = driver.find_element_by_xpath('//*[@id="gridParties"]/tbody/tr[1]/td[2]/div[1]/a').get_attribute(
# 'href')
load_page(profile_link, 'Party Details:', FLAGS.verbose)
r.suffix = None
r.dob = None # This portal has DOB as N/A for every defendent
r.race = driver.find_element_by_xpath(
'//*[@id="fd-table-2"]/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[7]/td[2]'
).text.strip()
r.sex = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[6]/td[2]'
).text.strip()
# Navigate to party profile
full_name = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[1]/td[2]'
).text.strip()
r.middle_name = None
r.last_name = None
if ',' in full_name:
r.first_name, r.middle_name, r.last_name = ScraperUtils.parse_name(
full_name)
else:
# If there's no comma, it's a corporation name.
r.first_name = Pii.String(full_name)
r.party_id = driver.find_element_by_xpath(
'//*[@id="mainTableContent"]/tbody/tr/td/table/tbody/tr[2]/td[2]/table[2]/tbody/tr/td[2]/table/tbody/tr[8]/td[2]'
).text.strip(
) # PartyID is a field within the portal system to uniquely identify defendants
record = r.build()
ScraperUtils.write_csv(FLAGS.output, record, FLAGS.verbose)
