def crop_and_extract(
page: pdfplumber.pdf.Page,
line: Dict,
adjust=False,
left_shift: int = 0,
up_shift: int = 20,
) -> str:
"""Extract text content for pdf line if any
Given a line of a pdf - extracted the text around it. If adjust
is True, reduce the cropped area to within the first line (usually above)
:param page: Page to crop
:param line: Line to crop around
:param adjust: Whether to check if another line is inside our crop
:param left_shift: Leftward crop adjustment
:param up_shift: Upward crop adjustment
:return: Content of the section
"""
bbox = (
int(line["x0"]) - left_shift,
int(line["top"]) - up_shift,
line["x1"],
line["top"],
)
crop = page.crop(bbox)
if adjust:
tops = [row["top"] for row in crop.lines if row["top"] != line["top"]]
if len(tops) > 0:
crop = page.crop(bbox=(*bbox[:1], tops[-1], *bbox[2:]))
return crop.filter(keys_and_input_text).extract_text()
def extract_other_creditors_ef(page: pdfplumber.pdf.Page, start: Dict,
stop: Dict, creditors: List) -> Dict:
"""Process other creditors to be notified if any
:param page:Page to crop
:param start:Y coordinate of the top of the creditor section
:param stop:Y coordinate of the bottom of the creditor section
:return: The key, address and acct information
"""
key_bbox = (start["x0"], start["top"] - 20, start["x1"], start["top"])
addy_bbox = (0, start["top"] - 20, start["x0"] - 20, stop["top"])
acct_bbox = (start["x1"] + 150, start["top"] + 20, page.width, stop["top"])
key = page.crop(key_bbox).filter(just_text_filter).extract_text()
address = page.crop(addy_bbox).filter(keys_and_input_text).extract_text()
acct = page.crop(acct_bbox).filter(keys_and_input_text).extract_text()
for creditor in creditors:
if creditor["key"] == str(key):
other_creditors = creditor["other_creditors"]
other_creditors.append({
"address": address,
"acct": acct,
"key": key
})
creditor["other_creditors"] = other_creditors
return creditors
def parse_unsecured_creditors(page: pdfplumber.pdf.Page, top: int,
bottom: int) -> Dict:
"""Extract the information on the unsecured creditor section
:param page: PDF page
:param top: Y coordinate of the top of section
:param bottom: Y coordinate of the bottom of section
:return: Organized creditor data
"""
data = []
crop_one = page.crop((0, max(100, top - 500), page.width, bottom))
crop = crop_one.crop((0, top, page.width, bottom))
key = crop.filter(key_filter).extract_text().replace("\n", "")
boxes = get_checkboxes(crop)
lines = crop.filter(remove_margin_lines).lines
for line in sorted(lines, key=lambda x: x["top"]):
if not data and line["width"] > 20:
continue
output = crop_and_extract(crop_one, line, adjust=True, up_shift=100)
if data or (output is not None and key == output.replace("\n", "")):
if len(data) == 10 and "2." in key or len(
data) == 8 and "4." in key:
continue
data.append(output)
if data:
return make_creditor_dict(data, boxes, key)
return {}
def extract_other_creditors_d(page: pdfplumber.pdf.Page, markers: List[Dict],
creditors: List) -> None:
"""Crop and extract address, key and acct # from the PDf
:param page: PDF page
:param markers: The top and bottom markers
:return: Address, key and account information
"""
adjust = 0 if len(markers) == 5 else 12
addy_bbox = (
0,
markers[0]["top"],
int(markers[-1]["x1"]) * 0.35,
markers[-1]["top"],
)
key_bbox = (
markers[-3]["x0"],
markers[0]["top"] - adjust,
markers[-3]["x1"],
markers[-3]["top"],
)
acct_bbox = (
markers[1]["x0"],
markers[1]["top"] - 12,
markers[1]["x1"],
markers[1]["top"],
)
address = page.crop(addy_bbox).filter(keys_and_input_text).extract_text()
key = page.crop(key_bbox).filter(
keys_and_input_text).extract_text().strip()
acct = page.crop(acct_bbox).filter(keys_and_input_text).extract_text()
for creditor in creditors:
if creditor["key"] == key:
other_creditors = creditor["other_creditors"]
other_creditors.append({
"key": key,
"address": address,
"acct": acct
})
creditor["other_creditors"] = other_creditors
return creditors
def get_1_to_2_from_a_b(only_page: pdfplumber.pdf.Page) -> List[Dict]:
"""Extract real estate, automobile, jet ski, boats etc, from A/B.
:param only_page:The PDF page to extract from
:return: Extracted content
"""
property_content = []
sections = find_property_sections(only_page)
if not sections:
return property_content
for top, bottom, key in sections:
bbox = (0, top, only_page.width, bottom)
crop = only_page.crop(bbox)
data = get_all_values_from_crop(crop.lines, only_page)
if "1." in key:
section = make_property_dict(key, data)
checkboxes = get_checkboxes(crop)
if not checkboxes:
section["property_interest"] = "Checkbox unreadable"
section["debtor"] = "Checkbox unreadable"
else:
section["property_interest"] = checkboxes["property"]
section["debtor"] = checkboxes["debtor"]
property_content.append(section)
if "3." in key or "4." in key:
if "3." in key:
section = make_car_dict(key, data)
else:
section = make_other_dict(key, data)
checkboxes = get_checkboxes(crop)
if not checkboxes:
section["debtor"] = "Checkbox unreadable"
else:
section["debtor"] = checkboxes["debtor"]
property_content.append(section)
return property_content
def parse_secured_creditors(only_page: pdfplumber.pdf.Page, top: int,
bottom: int) -> Dict:
"""Find and extract content from secured creditor portion of 106D
:param only_page:PDF page
:param top: Y coordinate for top of section
:param bottom: Y coordinate of bottom of section
:return: Organized data of the section
"""
page = only_page.crop((0, max(100, top - 500), only_page.width, bottom))
section = page.crop((0, top, only_page.width, bottom))
key = section.filter(key_filter).extract_text()
checkboxes = get_checkboxes(section)
data = []
for line in sorted(section.filter(remove_margin_lines).lines,
key=lambda x: x["top"]):
top = int(line["top"])
if not data and line["width"] > 20:
continue
page_crop = page.crop((line["x0"], top - 200, line["x1"], top))
tops = [
row["top"] for row in page_crop.lines if int(row["top"]) != top
]
if len(tops) > 0:
if len(data) == 6:
page_crop = page.crop(
(line["x0"], tops[-1] - 20, line["x1"], top))
elif len(data) == 8:
page_crop = page.crop((line["x0"], top - 50, line["x1"], top))
else:
page_crop = page.crop(
(line["x0"], tops[-1], line["x1"], line["top"]))
output = page_crop.filter(keys_and_input_text).extract_text()
if data or key == output:
data.append(output)
if data and len(data) > 10:
return make_secured_creditor_dict(data, checkboxes)
return {}