def find_patent_by_id(patent_id):
"""
@param patent_id: String with patent id
@return patent dictionary
"""
scraper = scraper_class()
status, soup, url = scraper.request_single_patent(patent_id)
patent_parsed = scraper.process_patent_html(soup)
return patent_parsed
def single_process_scraper(patent,path_to_data_file,data_column_order):
"""Scrapes a single google patent using the google scraper class
Function does not return any values, instead it writes the output
of the data into a csv file specified in the path_to_data_file
parameter
Inputs:
patent (str) : patent number including country prefix
lock (obj) : to prevent collisions, function uses a lock. You can pass whichever
lock you want to this parameter
path_to_data_file : absolute path to csv file to write data to
data_column_order : name of columns in order they will be saved in csv file
"""
# ~ Initialize scraper class ~ #
scraper=scraper_class()
# ~ Scrape single patent ~ #
err, soup, url = scraper.request_single_patent(patent)
# Checks if the scrape is successful.
# If successful -> parse text and deposit into csv file
# Else -> print error statement
if err=='Success':
patent_parsed = scraper.get_scraped_data(soup,url,patent)
# print("## Patent Parsed:")
out_row = ""
for key in data_column_order:
out_row = out_row + patent_parsed.get(key) + ","
out_row = out_row[:-1] + "\n"
with open(path_to_data_file, 'a') as file:
file.write(out_row)
else:
print('Patent {0} has error code {1}'.format(patent,err))
def single_process_scraper(patent, path_to_data_file, data_column_order):
"""Scrapes a single google patent using the google scraper class
Function does not return any values, instead it writes the output
of the data into a csv file specified in the path_to_data_file
parameter
Inputs:
patent (str) : patent number including country prefix
lock (obj) : to prevent collisions, function uses a lock. You can pass whichever
lock you want to this parameter
path_to_data_file : absolute path to csv file to write data to
data_column_order : name of columns in order they will be saved in csv file
"""
# ~ Initialize scraper class ~ #
scraper = scraper_class()
# ~ Scrape single patent ~ #
err, soup, url = scraper.request_single_patent(patent)
# Checks if the scrape is successful.
# If successful -> parse text and deposit into csv file
# Else -> print error statement
if err == 'Success':
patent_parsed = scraper.get_scraped_data(soup, url, patent)
# Save the parsed data to a csv file
# using multiprocessing lock function
# to prevent collisions
with lock:
with open(path_to_data_file, 'a', newline='') as ofile:
writer = csv.DictWriter(ofile, fieldnames=data_column_order)
writer.writerow(patent_parsed)
else:
print('Patent {0} has error code {1}'.format(patent, err))
# ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # # Example 1: Scrape a single patent # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # # ~ Import packages ~ # from google_patent_scraper import scraper_class # ~ Initialize scraper class ~ # scraper = scraper_class() # ~~ Scrape patents individually ~~ # patent_1 = 'US2668287A' patent_2 = 'US266827A' err_1, soup_1, url_1 = scraper.request_single_patent(patent_1) err_2, soup_2, url_2 = scraper.request_single_patent(patent_2) # ~ Parse results of scrape ~ # patent_1_parsed = scraper.get_scraped_data(soup_1, patent_1, url_1) patent_2_parsed = scraper.get_scraped_data(soup_2, patent_2, url_2) # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # # Example 2: Scrape a list of patents # ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ # # ~ Import packages ~ # from google_patent_scraper import scraper_class import json # ~ Initialize scraper class ~ # scraper = scraper_class()
import csv
import sys
import re
import datetime
import google_patent_scraper
import json
csv.field_size_limit(sys.maxsize)
with open('combined.txt', encoding='utf-16-le') as tsvfile:
reader = csv.DictReader(tsvfile, dialect='excel-tab')
scraper = google_patent_scraper.scraper_class()
i = 0
rows = []
for row in reader:
rows.append(row)
patent_numbers = re.findall("(US[0-9]+)", row['\ufeffPN '])
patent_number = patent_numbers[0]
for s in patent_numbers[1:]:
if len(s) < len(patent_number):
patent_number = s
if len(patent_number) > 10:
continue
scraper.add_patents(patent_number)
i += 1
#if i == 5:
# break
scraper.scrape_all_patents()
# ~ Import packages ~ #
from google_patent_scraper import scraper_class
import json
# ~ Initialize scraper class ~ #
scraper = scraper_class(return_abstract=True) #<- TURN ON ABSTRACT TEXT
# ~~ Scrape patents individually ~~ #
patent_1 = 'US20160180719A1'
err_1, soup_1, url_1 = scraper.request_single_patent(patent_1)
# ~ Parse results of scrape ~ #
patent_1_parsed = scraper.get_scraped_data(soup_1, patent_1, url_1)
for inventor in json.loads(patent_1_parsed['inventor_name']):
print(inventor['inventor_name'])
print(patent_1_parsed["grant_date"])
print(patent_1_parsed["patent"])
print(patent_1_parsed["abstract_text"])
# print(patent_1_parsed["image_urls"])
for link in patent_1_parsed['image_urls']:
print(link)
#print(patent_1_parsed)
#
#https://patentimages.storage.googleapis.com/ec/41/7b/76047542d0f945/US20160180719A1-20160623-D00003.png. <full size