import os
import re
from bs4 import BeautifulSoup
from funcs import read_from_json, save_to_file_by_address
base_address = os.getcwd()
html_address = base_address + "/main_qa_data/data/htmlner.json"
data = read_from_json(html_address)
html_file = ''
data = data[300:500]
for i, line in enumerate(data):
id = line["id"]
if id == 'qs211576':
pass
content = line["html"]
start_part = ""
end_part = ""
ners = content.lstrip(start_part)
ners = ners.rstrip(end_part)
if ners == "":
continue
html_ner = start_part + ners + end_part
parsed_html = BeautifulSoup(html_ner, 'html.parser')
# ner_span = parsed_html.find_all('div', {'class': 'off'})
# ner_span = parsed_html.find_all('span', {'style': 'background-color: #e8e'})
span_tags = parsed_html.find_all('span', style=re.compile(r'background-color'))
for span in span_tags:
text = span.text
raw_html = str(span)
raw_html = raw_html.replace('"', "'")
html_parts = raw_html.split(text)
# ORG
if html_parts[0].__contains__("background-color: #8ee"):
new_html = f"{text}ORG"
ners = ners.replace(raw_html,new_html)
# DATE
elif html_parts[0].__contains__("background-color: #ee6"):
new_html = f"{text}DATE"
ners = ners.replace(raw_html,new_html)
# NUM
elif html_parts[0].__contains__("background-color: #e8e"):
new_html = f"{text}NUM"
ners = ners.replace(raw_html,new_html)
# LOC
elif html_parts[0].__contains__("background-color: #ee8"):
new_html = f"{text}LOC"
ners = ners.replace(raw_html,new_html)
# PERCENT
elif html_parts[0].__contains__("background-color: #cc8"):
new_html = f"{text}PERCENT"
ners = ners.replace(raw_html,new_html)
#ners = ners.replace("", "ORG")
html_file += "" +str(id)+"
"+ "" + ners + "
"
print(str(i) + "/" + str(len(data)))
style = """"""
html_s = ""
body_s = ""
html_file = html_s + style + body_s + html_file + end_part
result_address = base_address + "/main_qa_data/data/htmlner.htm"
save_to_file_by_address(result_address, html_file)