def _get_kindcode_and_startpage(country, patnum, kindcode, target_folder):
"""
Get kindcode and startpage of a given patent.
Iterates through a list of possible kindcodes; for every kindocde, the
first page of the patent document is tried to be retrieved. A successful download
gives us the correct kindcode, for which the start page of the description is then retrieved.
- country: Country code
- patnum: Patent number
- kindocde: Kindcode (e.g. "A1", "B", ...)
- target_folder: Folder that the first page is downloaded to
"""
startpage = None
possible_kindcodes = []
if kindcode: possible_kindcodes = [kindcode]
possible_kindcodes += ['A1', 'A2', 'A3', 'A4', 'B1']
print possible_kindcodes
for code in possible_kindcodes:
try:
kindcode = download.get_pdf_page(country, patnum, code, 1, target_folder)
if kindcode:
meta = download.get_meta_data(country, patnum, 'DESCRIPTION', kindcode, skip=True)
print meta
if meta:
startpage = meta['DESCRIPTION']
break
except:
e = "Unexpected error:", sys.exc_info()[0]
continue
print kindcode, startpage
return kindcode, startpage
def _parse_pages_to_download(locations, country, patnum, kindcode, folder):
"""
Parses input and tries to retrieve pages of the description part of a patent.
"""
pages, columns = parser.get_pages(locations)
target_folder = _create_target_folder(country, patnum, folder)
kindcode, startpage = _get_kindcode_and_startpage(country, patnum, kindcode, target_folder)
if startpage and kindcode:
offset = 0
if country != "EP": offset += startpage - 1
for page in pages:
download.get_pdf_page(country, patnum, kindcode, page + offset, target_folder, "page-" + str(page))
for column in columns:
page = int(math.ceil(column / 2.))
offset = startpage - 1
print country, patnum, kindcode
download.get_pdf_page(country, patnum, kindcode, page + offset, target_folder, "column-" + str(column))
return kindcode, startpage