def find_property_sections(
only_page: pdfplumber.pdf.Page,
) -> Optional[Iterator[Tuple[Union[int, Any], int, Union[int, Any]]]]:
"""Find property sections to iterate over
Find Sections with white font and identifiers for each section
:param only_page: PDF page to crop
:return: None or Iterable sections as top bottom and id
"""
rows = only_page.filter(input_white_text_and_left_side).extract_words()
rows = [{
"top": int(row["top"]),
"text": row["text"]
} for row in rows if len(row["text"]) > 2 and row["text"][0] in "P12345"
and row["text"][1] == "."]
if len(rows) == 0:
return None
bottoms = [
int(line["top"]) for line in only_page.lines
if line["top"] > rows[0]["top"] and line["width"] > 530
][:len(rows)]
tops = [r["top"] for r in rows]
keys = [r["text"] for r in rows]
return zip(tops, bottoms, keys)
def get_3_to_8_form_a_b(
page: pdfplumber.pdf.Page,
) -> Tuple[List[Union[Dict[str, Any], Dict[Optional[Any], str]]], list,
Optional[dict]]:
"""Parse sections 3 to 8 of 106 A/B property form
:param page:The pdf page to parse
:return:Organized property data in the document.
"""
part = 0
totals, section, key = None, None, None
results, data, part_eight = [], [], []
rows = page.filter(filter_106_ab_content).extract_text().splitlines()
debtors = get_ab_debtors(rows)
# Remove debtor rows from lines
rows = [r for r in rows if r not in debtors]
for debtor in debtors:
rows = [r for r in rows if debtor not in r]
for row in rows[1:]:
match = re.match(r"Part \d:", row)
if match:
part += 1
continue
# Extract parts 3 to 7
if part in [3, 4, 5, 6, 7]:
match = re.match(r"^\d{1,2}\. ?|^5", row)
if not match:
data.append(row)
continue
if section == row:
continue
if "54. " in row:
results.append({"54.": row.split(" ")[1]})
if key:
data = [d for d in data if "[" not in d]
if data:
if key == "24." and data == ["2"]:
data = []
continue
results.append({key: clean_ab_data(data)})
data = []
section = row
key = row
if part == 8:
# Part 8 is the section containing grand totals.
part_eight.append(row)
if "63. " in row: # this is the final row of Part 8
totals = make_ab_totals(part_eight)
return results, debtors, totals
def get_checkboxes(crop: pdfplumber.pdf.Page) -> Dict:
"""Find and identify checked checkboxes
Using multiple tolerances, find checkboxes and identify them by the content
to the right of the checkbox.
:param crop: Section of pdf to extract checkboxes from
:return: Dictionary of selected checkboxes
"""
results = {}
# Use multiple tolerances to line up checkboxes on weird PDFs
for tolerance in [3, 4, 5]:
filtered_data = crop.filter(filter_boxes).extract_text(
y_tolerance=tolerance)
filtered_data = filtered_data.replace(
"Type of NONPRIORITY unsecured claim:", "")
if "[]" not in filtered_data:
# Checkboxes unreadable
return {}
filtered_lines = filtered_data.splitlines()
checkboxes = [x.replace(" ", " ") for x in filtered_lines if "[" in x]
query1 = ["debtor"]
query2 = ["community", "see instructions", "claim relates"]
query3 = ["No", "Yes"]
query4 = ["contingent", "unliquidated", "disputed"]
query5 = [
"domestic",
"taxes",
"death",
"specify",
"loans",
"obligations",
"pension",
"including",
"judgment",
"statutory",
"agreement",
]
debtor = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box.lower() for s in query1)
]
community = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box.lower() for s in query2)
]
offset = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box for s in query3)
]
offset = [ans for ans in offset if re.match(r"^(Yes|No)$", ans)]
info = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box.lower() for s in query4)
]
claim_type = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box.lower() for s in query5)
]
property_values = [
box.split(" ", 1)[1].strip() for box in checkboxes
if "√" in box and any(s in box for s in property_options)
]
property_values = [
s for s in property_options
if any(s in box for box in property_values)
]
if claim_type:
if "Specify" in claim_type[0]:
claim_type = ["Other. Specify"]
data = {
"debtor": debtor,
"community": community,
"offset": offset,
"info": info,
"claim_type": claim_type,
"property": property_values,
}
if not results:
results = data
else:
datum = [{k: v} for k, v in data.items() if v != []]
data = {}
for item in datum:
data = {**data, **item}
results = {**results, **data}
return results