def cwb_crawler():
import Lily.crawler.url_string as curlstr
db = cdb.database('data_crawler_cwb_earthquake_list.sqlite')
db_target_tab = 'data_crawler_cwb_sensible_earthquake_download'
eq_dir_pa = '''https://scweb.cwb.gov.tw/earthquake/Page.aspx?ItemId=20&Date={0}'''
sub_url = '''https://scweb.cwb.gov.tw/earthquake/Page.aspx{0}'''
days = pandas.date_range(
datetime.datetime.today() - datetime.timedelta(days=90),
datetime.datetime.today() + datetime.timedelta(days=31),
freq='M')
for mon in days:
eq_list = pandas.DataFrame(
columns=['id', 'lzma_html', 'download_time'])
time.sleep(90 / 1000.0)
day_url = eq_dir_pa.format(mon.strftime('%Y%m'))
cur1 = curlstr.url_string(day_url)
arg1 = LH.fromstring(cur1.to_str())
arg2 = arg1.xpath('//tr/td/a/@href')
arg3 = {}
for elem in arg2:
if elem not in arg3 and elem[1:7] == 'ItemId':
print('download html', elem)
cur2 = curlstr.url_string(sub_url.format(elem))
arg3[elem] = True
eq_list.loc[len(eq_list)] = [
elem[22:],
cur2.to_lzma_xz(),
datetime.datetime.now()
]
eq_list.to_sql(db_target_tab,
db.connect,
if_exists='append',
index=False)
db.connect.execute('''delete from {0} where rowid not in
(select max (rowid) from {0} group by id, lzma_html)'''
.format(db_target_tab))
db.connect.commit()
return
import Lily.ctao.database as cdb
import Lily.crawler.url_string as url
for a in range(1,10):
target='''http://163.29.251.188/botedata/%E4%BA%A4%E9%80%9A%E6%B5%81%E9%87%8F/107%E5%B9%B4%E5%BA%A6/HTM/S00{0}.files/sheet001.htm'''
target = target.format(a)
file = '''D:/107_s00{0}_0001.html'''
file = file.format(a)
arg1 = url.url_string(target)
arg1.to_file(file)
print (target, file)
for a in range(1,10):
target='''http://163.29.251.188/botedata/%E4%BA%A4%E9%80%9A%E6%B5%81%E9%87%8F/107%E5%B9%B4%E5%BA%A6/HTM/S0{0}0.files/sheet001.htm'''
target = target.format(a)
file = '''D:/107_s0{0}0_0001.html'''
file = file.format(a)
arg1 = url.url_string(target)
arg1.to_file(file)
print (target, file)
