Skip to content

Commit

Permalink
Merge pull request #1276 from freelawproject/recap_email_bankruptcy_s…
Browse files Browse the repository at this point in the history 
…hort_description
  • Loading branch information
@mlissner
mlissner authored Jan 10, 2025
2 parents bb6a4e0 + 93c4d4b commit 5981261
Show file tree
Hide file tree
Showing 62 changed files with 4,234 additions and 115 deletions.
192 changes: 86 additions & 106 deletions juriscraper/pacer/email.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -48,6 +48,31 @@ class NotificationEmail(BaseDocketReport, BaseReport):
"Modified Date Filed from",
"Added Correct PDF to document",
]
bankruptcy_courts_with_tests = [
"arwb",
"cacb",
"casb",
"cob",
"ctb",
"dcb",
"deb",
"flsb",
"ianb",
"mdb",
"nceb",
"ndb",
"nhb",
"njb",
"nyeb",
"nysb",
"okeb",
"paeb",
"pamb",
"pawb",
"tnmb",
"txnb",
"vaeb",
]

def __init__(self, court_id):
self.court_id = court_id
Expand All @@ -57,6 +82,7 @@ def __init__(self, court_id):
self.docket_numbers = []
self.subject = None
self.case_names = []
self.raw_docket_numbers = set()
if self.court_id.endswith("b"):
self.is_bankruptcy = True
else:
Expand Down Expand Up @@ -172,16 +198,16 @@ def _parse_docket_number(
return case_number, self._return_default_dn_components()

path = self._sibling_path("Case Number")
docket_numbers_str = current_node.xpath(f"{path}/a/text()")
self.raw_docket_numbers.update(set(docket_numbers_str))
docket_number, docket_number_components = (
self._parse_docket_number_strs(
current_node.xpath(f"{path}/a/text()")
)
self._parse_docket_number_strs(docket_numbers_str)
)
if not docket_number:
docket_numbers_str = current_node.xpath(f"{path}/p/a/text()")
self.raw_docket_numbers.update(set(docket_numbers_str))
docket_number, docket_number_components = (
self._parse_docket_number_strs(
current_node.xpath(f"{path}/p/a/text()")
)
self._parse_docket_number_strs(docket_numbers_str)
)
return docket_number, docket_number_components

Expand All @@ -199,6 +225,7 @@ def _parse_docket_number_plain(
email_body = self.tree.text_content()
regex = r"Case Number:(.*)"
docket_number = re.findall(regex, email_body)
self.raw_docket_numbers.update(set(docket_number))
return self._parse_docket_number_strs(docket_number)

def _get_date_filed(self) -> date:
Expand Down Expand Up @@ -396,7 +423,6 @@ def _get_dockets(self) -> DocketType:
docket_number, docket_number_components = (
self._parse_docket_number_plain()
)
docket_number = docket_number
# Cache the docket number for its later use.
self.docket_numbers.append(docket_number)

Expand Down Expand Up @@ -534,115 +560,69 @@ def _get_docket_entries(
return []

def _parse_bankruptcy_short_description(self, subject: str) -> str:
"""Parse the short description of a bankruptcy case from the email
subject. Subjects for bankruptcy varies a lot from court to court.
This function supports parsing the short description for courts with
known examples.
"""Parse the short description of a bankruptcy case from the email subject
The subject contains: the docket number, the case name, the short description
and special characters. The presence and order of these elements varies from
court to court. The short description that we parse out should match the
"Description" on the "History/Document" report on PACER
A special case are multi-docket NEFs, of which we have only seen
2-docket NEFs. These are labelled as "_multi_" on the example files, and we
have only seen "Adversary Cases", so far. The subject will use one
of the case names and one of the docket numbers, and then follow
the general rules
:param subject: The email subject string.
:return: The parsed short description.
"""
# Some courts have subjects like
# `Multiple Cases "{docket} {case name} Close [Aa]dversary [Cc]ase" - {initials}`
if close_adv_match := re.search(
r"close adversary case", subject, flags=re.IGNORECASE
):
return close_adv_match.group(0)

# raw dockets from "plain/text" email may have a URL
raw_docket_numbers = [
i.strip().split(" ")[0] for i in self.raw_docket_numbers
]
raw_docket_numbers.sort(key=len, reverse=True) # use longest first
for part in raw_docket_numbers + self.docket_numbers + self.case_names:
subject = subject.replace(part, " ")

# See paeb_3.txt for a test of multi docket NEF
# So far, we have only seen 2-docket NEF
is_multidocket = len(self.docket_numbers) == 2
if len(self.docket_numbers) > 1 and self.court_id != "njb":
# Sometimes the full case name is not used in the `subject`
# Some courts use a 18 character limit
# See deb_2.txt, pamb_1 and pamb_3 for examples
for case_name in self.case_names:
subject = subject.replace(case_name[:18].strip(), " ")
subject = subject.replace(case_name.split(" and ")[0], " ")

# Deletes:
# - "NEF: " placeholder
# - "Ch \d{1,2}" abbreviations
cleanup_regex = r"(NEF:? )|(C[Hh][- ]?(7|9|11|13))|(C[hH][\s-]*$)"
subject = re.sub(cleanup_regex, " ", subject)

# Courts like `nhb` do not use the "Ch \d{1,2}" abbreviation
# and we must delete the "Chapter..." string; but only once
# See nhb_2 for an example with 2 "Chapter..." strings
chapter_regex = r"Chapter[- ]?(7|9|11|13)"
if self.court_id in ["nhb"]:
subject = re.sub(chapter_regex, " ", subject, 1)
elif len(re.findall(chapter_regex, subject)) > 1:
logger.error(
"Not parsing description for Bankruptcy Multi Docket NEF for court '%s'",
self.court_id,
extra={
"fingerprint": [
f"{self.court_id}-not-parsing-multi-docket-short-description"
]
},
"Trying to parse a RECAP email with more than 1 'Chapter...' string"
)
return ""

short_description = ""
docket_number = self.docket_numbers[0]
case_name = self.case_names[0]

if is_multidocket:
# Docket number / case name from one or both of the 2 cases
# may be used in the subject string
if self.docket_numbers[1] in subject:
docket_number = self.docket_numbers[1]
if self.case_names[1] in subject:
case_name = self.case_names[1]

if self.court_id in [
"cacb",
"ctb",
"cob",
"ianb",
"nyeb",
"txnb",
"okeb",
]:
# In: 6:22-bk-13643-SY Request for courtesy Notice of Electronic Filing (NEF)
# Out: Request for courtesy Notice of Electronic Filing (NEF)
short_description = subject.split(docket_number)[-1]

# Remove docket number traces "-AAA"
regex = r"^-.*?\s"
short_description = re.sub(regex, "", short_description)
elif self.court_id in ["njb", "dcb", "vaeb", "paeb", "mdb", "arwb"]:
# In: Ch-11 19-27439-MBK Determination of Adjournment Request - Hollister Construc
# Out: Determination of Adjournment Request
# In Ch-13 1:24-bk-70534 Meeting of Creditors - Second Non-Appearance; Michael Clayton Lowry
# Out: Meeting of Creditors - Second Non-Appearance
short_description = subject.split(docket_number)[-1]
# Remove docket number traces "-AAA"
# Remove CH after docket and BK after short description for dcb
regex = r"^-.*?\s|C[Hh][\s\d]+|[ (]?B[Kk]( Other)?[) ]?"
short_description = re.sub(regex, "", short_description)
separator = ";" if self.court_id == "arwb" else "-"
short_description = short_description.rsplit(separator, 1)[0]
elif self.court_id == "nysb":
# In: 22-22507-cgm Ch13 Affidavit Re: Gerasimos Stefanitsis
# Out: Affidavit
short_description = subject.split(case_name)[0]
short_description = short_description.replace("Re:", "")
short_description = short_description.split(docket_number)[-1]

# Remove strings starting with "Ch" followed by a number
regex = r"\bCh\d+\b"
short_description = re.sub(regex, "", short_description)

# Remove docket number traces "-AAA"
regex = r"^-.*?\s"
short_description = re.sub(regex, "", short_description)
elif self.court_id in ["pawb", "ndb", "deb", "pamb", "nhb"]:
# In: Ch-7 22-20823-GLT U LOCK INC Reply
# Out: Reply
if case_name in subject:
short_description = subject.split(case_name)[-1]
elif case_name[:18] in subject:
# See deb_2.txt, pamb_1 and pamb_3 for examples
short_description = subject.split(case_name[:18])[-1]
elif (
" and " in case_name and case_name.split(" and ")[0] in subject
):
# See pamb_2.txt
short_description = subject.split(case_name.split(" and ")[0])[
-1
]
elif self.court_id in [
"tnmb",
]:
# In: Docket Order - Continue Hearing (Auto) Ch 13 Jeffery Wayne Lovell and Tiffany Nicole Lovell 1:24-bk-01377
# Out: Docket Order - Continue Hearing (Auto) Ch 13
if case_name in subject:
short_description = subject.split(case_name)[0]
else:
subject = subject.strip(" ;:,- ")
# some courts use "Re: {case name}"
short_description = re.sub("( Re$)|(^Re:? )", "", subject)

if self.court_id not in self.bankruptcy_courts_with_tests:
logger.error(
"Short description has no parsing for bankruptcy court '%s'",
"Parsing a RECAP email from court %s without tests",
self.court_id,
extra={
"fingerprint": [
f"{self.court_id}-not-parsing-short-description"
]
},
)

return short_description
Expand Down
46 changes: 46 additions & 0 deletions tests/examples/pacer/nef/s3/cacb_7.json
View file
Original file line number Diff line number Diff line change
@@ -0,0 +1,46 @@
{
"appellate": false,
"contains_attachments": false,
"court_id": "cacb",
"dockets": [
{
"case_name": "Lee v. Newman",
"date_filed": null,
"docket_entries": [
{
"date_filed": "2023-02-15",
"description": "Order granting stipulation to continue pre-trial conference from 2/28/23 to 4/25/23 @ 10:00 a.m. (BNC-PDF) (Related Doc [75]) Signed on 2/15/2023 (JC6)",
"document_number": "76",
"document_url": "https://ecf.cacb.uscourts.gov/doc1/9730106502427?pdf_header=&magic_num=94356268&de_seq_num=252&caseid=1927388",
"pacer_case_id": "1927388",
"pacer_doc_id": "9730106502427",
"pacer_magic_num": "94356268",
"pacer_seq_no": "252",
"short_description": "ORDER to continue/reschedule hearing (BNC-PDF)"
}
],
"docket_number": "6:21-ap-01071",
"federal_defendant_number": null,
"federal_dn_case_type": "ap",
"federal_dn_judge_initials_assigned": "SC",
"federal_dn_judge_initials_referred": null,
"federal_dn_office_code": "6"
}
],
"email_recipients": [
{
"email_addresses": [
"[email protected]",
"[email protected]",
"[email protected]",
"[email protected]",
"[email protected]",
"[email protected]@y.net",
"[email protected]",
"[email protected]",
"[email protected]"
],
"name": ""
}
]
}
Loading

0 comments on commit 5981261

Please sign in to comment.