def raw_auditor(self):
with _by_pdfplumber(self.pdf_obj) as pdf:
auditor_report_last_pages = [
pdf.pages[page_nums[-1]]
for page_nums in self.get_outline_pageRange()
]
search_results = [
search_pattern_from_page_or_cols(
page=page, pattern=AnnualReport.auditor_regex)
for page in auditor_report_last_pages
]
if not any(search_results) and any(
map(abnormal_page, auditor_report_last_pages)):
auditor_report_last_pages = [
turn_n_page(pdf, page, 2) if is_landscape(page) else
turn_n_page(pdf, page, -1) if is_full_cn(page) else page
for page in auditor_report_last_pages
]
search_results = [
search_pattern_from_page_or_cols(
page=page, pattern=AnnualReport.auditor_regex)
for page in auditor_report_last_pages
]
auditors = set()
for result in filter(None, search_results):
if type(result) is tuple:
for r in filter(None, result):
auditors.add(r.group('auditor'))
else:
auditors.add(result.group('auditor'))
return tuple(auditors)
def search_outline_in_pages(self, pattern, page_range=None, size='fontname', verbose=False, show_matched=False) -> list:
'''
return a list of pages number in tuples that contains pattern
'''
print('search by page!')
# print(f'pattern: {pattern}')
pages = set()
matched_pattern = []
with _by_pdfplumber(self.pdf_obj) as pdf:
if not page_range:
page_range = pdf.pages
else:
page_range = [pdf.pages[p] for p in page_range]
for page in page_range:
p = page.page_number - 1
# if verbose: print(f'searching p.{p}')
try:
title_alike_txts = get_title_liked_txt(page, size=size)
except KeyError:
logging.warning('Non textual page')
continue
for txt in title_alike_txts:
if search_pattern_from_txt(txt, pattern):
pages.add(p)
matched_pattern.append(txt)
if verbose: print(f'with pattern: found {txt} on p.{p}!')
consecutive_pages = [tuple(li) for li in consecutive_int_list(unique(pages))]
if show_matched:
return consecutive_pages, matched_pattern
return consecutive_pages
def get_outline_pageRange(self, outline_pattern=None):
if outline_pattern is None:
outline_pattern = AnnualReport.audit_report_regex
with _by_pdfplumber(self.pdf_obj) as pdf:
return _get_page_by_outline(
self.toc, outline_pattern) or _get_page_by_page_title_search(
pdf, outline_pattern)
def target_section(self, p):
with _by_pdfplumber(self.pdf_obj) as pdf:
page = pdf.pages[p]
df = pd.DataFrame(page.chars)
df = df[~df.text.str.contains(r'[^\x00-\x7F]+')]
target_x0, target_x1 = float(self.target_x0(df)), float(self.target_x1(df))
target_top = float(self.target_top(df))
target_bottom = self.target_bottom(df) or page.height
# print(tagert_x0, tagert_x1)
# x0, x1 = 0, float(page.width)
# section = page.crop((x0, target_top , x1, float(target_bottom)), relative=True)
section = page.crop((target_x0, target_top , target_x1, float(target_bottom)), relative=True)
return section