def parse_profile(self, response):
item = {}
item['election_year'] = self.election_year
item['county'] = self.county
item['in_office'] = True
item['term_start'] = '%s-12-25' % item['election_year']
item['term_end'] = {'date': '2018-12-24'}
print response.xpath(
u'//p/span[re:test(., "\s+副?議(員|長)")]/text()').extract_first()
item['name'], item['title'] = response.xpath(
u'//p/span[re:test(., "\s+副?議(員|長)")]/text()').extract_first(
).split()
item['gender'] = self.ref[item['name']]['sex']
item['constituency'] = response.xpath('//td/text()').re(
u'選區：\s*(.+)')[0].strip()
item['district'] = self.constituency[item['constituency']]
item['image'] = urljoin(
response.url,
response.xpath(u'//p/img/@src').extract_first())
item['links'] = [{'url': response.url, 'note': u'議會個人官網'}]
item['party'] = response.xpath('//td/text()').re(
u'政黨：\s*(.+)')[0].strip()
item['birth'] = common.ROC2AD(
response.xpath('//td/text()').re(u'出生日期：\s*(.+)')[0])
website = response.xpath('//td/text()').re(u'網站連結：\s*(.+)')
if website:
item['links'].append({'url': website[0].strip(), 'note': u'個人網站'})
item['contact_details'] = []
contact_mappings = {
u'連絡電話': 'voice',
u'傳真號碼': 'fax',
u'服務處': 'address',
u'電子郵件': 'email'
}
for label, name in contact_mappings.items():
values = [
x.strip() for x in
response.xpath(u'//td[re:test(., "%s：")]/text()' %
'\s*'.join(label)).re(u'%s：\s*(.+)\s*' % label)
if x.strip()
]
for value in values:
item['contact_details'].append({
'label': label,
'type': name,
'value': value
})
item['experience'] = [
x.strip()
for x in response.xpath(u'//img[contains(@src, "speaker0")]')
[1].xpath('ancestor::tr/following-sibling::tr[1]//tr/td[1]/text()'
).extract() if x.strip()
]
item['platform'] = [
x.strip()
for x in response.xpath(u'//img[contains(@src, "speaker0")]')
[2].xpath('ancestor::tr/following-sibling::tr[1]//tr/td[1]/text()'
).extract() if x.strip()
]
yield item
def parse_meeting(self, response):
try:
sitting = response.xpath('//text()').re(u'(.+)日程表')[0]
trs = [
tr for tr in response.xpath('//table/descendant::tr')
if tr.xpath('td[3]/text()').re('\d+')
]
for tr in trs:
item = {}
item['election_year'] = self.election_year
item['date'] = common.ROC2AD(
tr.xpath('td[1]/text()').extract_first())
item['sitting'] = sitting
item['meeting'] = tr.xpath('td[3]/text()').extract_first()
item['download_url'] = tr.xpath(
'td[6]/descendant::a[1]/@href').extract_first()
ext = item['download_url'].split('.')[-1]
file_name = '%s_%s.%s' % (item['sitting'], item['meeting'],
ext)
if ext == 'pdf':
yield response.follow(item['download_url'],
callback=self.download_pdf,
meta={
'item': item,
'file_name': file_name
})
elif ext == 'htm':
yield response.follow(item['download_url'],
callback=self.parse_html,
meta={
'item': item,
'file_name': file_name
})
except scrapy.exceptions.NotSupported:
pass
def parse_list(self, response):
for tr in response.css('#table2 tr'):
link = tr.xpath(u'descendant::a[re:test(., "^第%d屆")]/@href' % self.ad).extract_first()
if link:
item = {}
item['election_year'] = self.election_year
item['date'] = common.ROC2AD(tr.xpath('td[1]/text()').extract_first())
item['meeting'] = tr.xpath('td[3]/descendant::a/text()').extract_first()
item['meeting'] = item['meeting'].replace('.', u'、')
item['download_url'] = urljoin(response.url, link)
ext = item['download_url'].split('.')[-1]
file_name = '%s.%s' % (item['meeting'], ext)
cmd = 'mkdir -p %s && wget --no-check-certificate -c -O %s%s "%s"' % (self.output_path, self.output_path, file_name, item['download_url'])
retcode = subprocess.call(cmd, shell=True)
yield item
def parse_profile(self, response):
item = {}
item['election_year'] = self.election_year
nodes = response.xpath('//table/tbody/tr')
ref = {
u'屆別': {
'key': 'sitting',
'path': 'td/span/text()'
},
u'類別': {
'key': 'category',
'path': 'td/span/text()'
},
u'日期': {
'key': 'date',
'path': 'td/span/text()'
},
u'資料名稱': {
'key': 'meeting',
'path': 'td/span/text()'
},
u'檔案': {
'key': 'download_url',
'path': 'td/a/@href',
'extra': 'http://obas_front.tcc.gov.tw:8080/Agenda/'
},
}
for node in nodes:
value = ref.get(node.xpath('th/text()').extract_first().strip())
if value:
item[value['key']] = '%s%s' % (value.get(
'extra', ''), node.xpath(value['path']).extract_first())
item['date'] = common.ROC2AD(item['date'])
ext = re.search(u'FileName=[\w\d]+\.(\w+)&',
item['download_url']).group(1)
file_name = '%s_%s.%s' % (item['sitting'], item['meeting'], ext)
cmd = 'mkdir -p %s && wget -c -O %s%s "%s"' % (
self.output_path, self.output_path, file_name,
item['download_url'])
retcode = subprocess.call(cmd, shell=True)
return item
def parse_profile(self, response):
item = response.meta['item']
item['id'] = '-'.join(re.findall(u'=([^&]*)', response.url))
for key, label in [('category', u'類別'), ('abstract', u'案由'), ('description', u'說明'), ('methods', u'辦法'), ('remark', u'備註'), ]:
content = response.xpath(u'string((//td[re:test(., "%s")]/following-sibling::td)[1])' % label).extract_first()
if content:
item[key] = content.strip()
item['proposed_by'] = re.split(u'\s|、', re.sub(u'(副?議長|議員)', '', u'、'.join([x.strip() for x in response.xpath(u'(//td[re:test(., "提案(人|單位)")]/following-sibling::td)[1]/text()').extract()])))
item['petitioned_by'] = re.split(u'\s|、', re.sub(u'(副?議長|議員)', '', u'、'.join([x.strip() for x in (response.xpath(u'(//td[re:test(., "連署人")]/following-sibling::td)[1]/text()').extract() or [])])))
item['motions'] = []
for motion in [u'一讀', u'委員會審查意見', u'二讀決議', u'三讀決議', ]:
date = common.ROC2AD(''.join(response.xpath(u'(//td[re:test(., "%s")]/following-sibling::td)[1]/span/text()' % motion).extract()))
resolution = ''.join([x.strip() for x in response.xpath(u'(//td[re:test(., "%s")]/following-sibling::td)[1]/text()' % motion).extract()])
if date or resolution:
item['motions'].append(dict(zip(['motion', 'resolution', 'date'], [motion, resolution, date])))
item['links'] = [
{
'url': response.url,
'note': 'original'
}
]
return item
def parse_profile(self, response):
trs = response.xpath('//table[@id="ContentPlaceHolder1_gvIndex"]/tr')
for tr in trs:
item = {}
tds = tr.xpath('td')
if tds:
item['election_year'] = self.election_year
item['date'] = common.ROC2AD(
tds[1].xpath('text()').extract_first())
meeting = tds[2].xpath('text()').extract_first()
item['meeting'] = tds[2].xpath('text()').re(
u'(.+?)[紀記][錄錄]')[0]
item['download_url'] = urljoin(
response.url,
tds[3].xpath('a/@href').extract_first().strip())
ext = item['download_url'].split('.')[-1]
file_name = '%s.%s' % (item['meeting'], ext)
cmd = 'mkdir -p %s && wget -c -O %s%s "%s"' % (
self.output_path, self.output_path, file_name,
item['download_url'])
retcode = subprocess.call(cmd, shell=True)
time.sleep(1)
yield item
if match:
if match.group('type') == u'定期':
uid = '%s-%s-%02d-CS-%02d' % (county_abbr3, election_years[int(
match.group('ad'))], int(
match.group('session')), int(match.group('times')))
elif match.group('type') == u'臨時':
uid = '%s-%s-T%02d-CS-%02d' % (county_abbr3, election_years[int(
match.group('ad'))], int(
match.group('session')), int(match.group('times')))
sittings.append({
"uid": uid,
"name": re.sub('\s', '', match.group('name')),
"county": county,
"election_year": election_year,
"session": match.group('session'),
"date": common.ROC2AD(total_text[match.end():]),
"start": match.start(),
"end": match.end()
})
for i in range(0, len(sittings)):
# --> sittings, attendance, filelog
if i != len(sittings) - 1:
one_sitting_text = total_text[
sittings[i]['start']:sittings[i + 1]['start']]
else:
one_sitting_text = total_text[sittings[i]['start']:]
logging.info(sittings[i]['uid'])
common.InsertSitting(c, sittings[i])
common.FileLog(c, sittings[i]['name'])
present_match = Present_Token.search(one_sitting_text)
if present_match:
Namelist_Token = re.compile(u'''
^.*?
具名表決[，,]
(贊成|反對).*?者有(?P<dicision_a>.*)[，,]?共計\s*\d+位[；;]
(贊成|反對).*?者有(?P<dicision_b>.*)[，,]?共計\s*\d+位[；;]
表決結果.*?$
''', re.X | re.M)
sittings = []
for match in Session_Token.finditer(total_text):
if match:
if match.group('type') == u'定期':
uid = '%s-%s-%02d-CS-%02d' % (county_abbreviation, election_years[int(match.group('ad'))], int(match.group('session')), int(match.group('times')))
elif match.group('type') == u'臨時':
uid = '%s-%s-T%02d-CS-%02d' % (county_abbreviation, election_years[int(match.group('ad'))], int(match.group('session')), int(match.group('times')))
sittings.append({"uid":uid, "name": re.sub('\s', '', match.group('name')), "county": county, "election_year": election_years[int(match.group('ad'))], "session": match.group('session'), "date": common.ROC2AD(total_text[match.end():]), "start": match.start(), "end": match.end()})
for i in range(0, len(sittings)):
break
# --> sittings, attendance, filelog
if i != len(sittings)-1:
one_sitting_text = total_text[sittings[i]['start']:sittings[i+1]['start']]
else:
one_sitting_text = total_text[sittings[i]['start']:]
print sittings[i]
common.InsertSitting(c, sittings[i])
common.FileLog(c, sittings[i]['name'])
# absent
absent_match = Absent_Token.search(one_sitting_text)
exclude = []
if absent_match:
names = re.sub(u'(副?議長|議員)', '', absent_match.group('names'))
worksheets = sh.worksheets()
for wks in worksheets:
rows = wks.get_all_records()
position_type = 'mayors'
county = wks.title.replace(u'台', u'臺')
print county
for row in rows:
if not row[u'姓名']:
continue
candidate = {}
candidate['type'] = position_type
candidate['county'] = county
candidate['constituency'] = 0
candidate['name'] = common.normalize_person_name(row[u'姓名'])
candidate['election_year'] = election_year
candidate['birth'] = common.ROC2AD(row[u'出生年月日'])
candidate['party'] = common.normalize_party(row[u'政黨'])
candidate['gender'] = row[u'性別']
candidate['number'] = row[u'號次']
candidate['education'] = row[u'學歷']
candidate['experience'] = row[u'經歷']
candidate['platform'] = row[u'政見']
candidate['image'] = "%s/mayors/%s/%s_%04d.jpg" % (
common.storage_domain(), candidate['election_year'],
candidate['county'], candidate['number'])
if position_type == 'mayors':
candidate[
'candidate_uid'], created = common.get_or_create_moyor_candidate_uid(
c, candidate)
else:
candidate[
def parse_profile(self, response):
item = response.meta['item']
nodes = response.xpath('//div[@id="detail"]/table/tr')
motions, committee_motion, council_motion = [], {}, {}
for node in nodes:
if node.xpath('td/text()')[0].re(u'目前處理程序'):
item['last_action'] = node.xpath('td/text()').extract()[1]
elif node.xpath('td/text()')[0].re(u'案由'):
item['abstract'] = node.xpath('td/text()').extract()[1]
elif node.xpath('td/text()')[0].re(u'提案人'):
item['proposed_by'] = node.xpath(
'td/div/text()').extract()[0].strip().split(u'、')
elif node.xpath('td/text()')[0].re(u'召集人/委員'):
item['proposed_by'] = node.xpath(
'td/text()').extract()[1].strip().split(u'、')
elif node.xpath('td/text()')[0].re(u'議決會次'):
council_motion['motion'] = u'大會議決'
council_motion['date'] = common.ROC2AD(
node.xpath('td/text()').extract()[1].split()[0])
council_motion['sitting'] = ''.join(
node.xpath('td/text()').extract()[1].split()[1:])
elif node.xpath('td/text()')[0].re(u'議決文'):
council_motion['resolution'] = node.xpath(
'td/text()').extract()[1]
elif node.xpath('td/text()')[0].re(u'案(\s| )+?號'):
item['bill_no'] = node.xpath('td/text()').extract()[1].strip()
elif node.xpath('td/text()')[0].re(u'來文文號'):
td = node.xpath('td/text()').extract()[1].split()
d = dict(
zip(['motion', 'resolution', 'date'],
[u'來文', None, common.ROC2AD(td[0])]))
if len(td) > 1:
d['no'] = td[1]
motions.append(d)
elif node.xpath('td/text()')[0].re(u'收文日期'):
motions.append(
dict(
zip(['motion', 'resolution', 'date'], [
u'收文', None,
common.ROC2AD(
node.xpath('td/text()').extract()[1])
])))
elif node.xpath('td/text()')[0].re(u'審查日期'):
committee_motion['motion'] = u'委員會審查意見'
committee_motion['date'] = common.ROC2AD(
node.xpath('td/text()').extract()[1])
elif node.xpath('td/text()')[0].re(u'審查意見'):
committee_motion['resolution'] = '\n'.join(
node.xpath('td/text()').extract()[1:])
elif node.xpath('td/text()')[0].re(u'發文文號'):
td = node.xpath('td/text()').extract()[1].split()
d = dict(
zip(['motion', 'resolution', 'date'],
[u'發文', None, common.ROC2AD(td[0])]))
if len(td) > 1:
d['no'] = td[1]
motions.append(d)
elif node.xpath('td/text()')[0].re(u'執行情形'):
item['execution'] = node.xpath('td/text()').extract()[1]
elif node.xpath('td/text()')[0].re(u'備[\s]*?註'):
item['remark'] = '\n'.join(
node.xpath('td/text()').extract()[1:])
for motion in [committee_motion, council_motion]:
if motion:
motions.append(motion)
item['motions'] = sorted(motions,
key=lambda x: x.get('date'),
reverse=True)
item['links'] = [{'url': response.url, 'note': 'original'}]
return item
meetings = json.load(
open('../../../data/kmc/meeting_minutes-%s.json' % election_year))
for meeting in meetings:
total_text = unicodedata.normalize(
'NFC',
codecs.open(
'../../../data/kmc/meeting_minutes/%s/%s.txt' %
(election_year, meeting['sitting']), "r", "utf-8").read())
total_text = re.sub(u'．', u'‧', total_text)
total_text = re.sub(u' ', ' ', total_text)
match = Session_Token.search(meeting['sitting'])
if match:
for i, session in enumerate(Present_Token.finditer(total_text), 1):
meeting['date'] = common.ROC2AD(
re.search(u'時\s*間[:：](.*)',
total_text[:session.start()].strip().split(
'\n')[-1]).group(1))
if match.group('type') == u'臨時':
uid = '%s-%s-T%s-CS-%02d' % (county_abbr3, election_year,
match.group('session'), i)
else:
uid = '%s-%s-%s-CS-%02d' % (county_abbr3, election_year,
match.group('session'), i)
sitting = {
"uid": uid,
"name": u'%s議會%s第%d次會議' % (county, meeting['sitting'], i),
"county": county,
"election_year": election_year,
"session": match.group('session'),
"date": meeting['date']
}
def parse_profile(self, response):
item = response.meta['item']
item_ad = response.css(u'#lbFmotion_expireb::text').extract_first()
for election_year, ad in self.ads.items():
if int(item_ad) == ad:
item['election_year'] = election_year
break
if item['election_year'] != self.election_year:
return
for key, label in [('bill_id', u'lbFmotion_No'), ('type', u'lbFmotion_Category'), ('category', u'lbFmotion_Class'), ('abstract', u'lbFmotion_From'), ('description', u'lbFmotion_Reason'), ('methods', u'lbFmotion_Way')]:
content = response.css(u'#%s::text' % label).extract_first()
if content:
item[key] = content.strip()
item['proposed_by'] = re.split(u'\s|、', re.sub(u'(副?議長|議員)', '', response.css(u'#lbFmotion_People::text').extract_first()).strip())
item['petitioned_by'] = re.split(u'\s|、', re.sub(u'(副?議長|議員)', '', (response.css(u'#lbFmotion_AddTo::text').extract_first() or '')).strip())
item['motions'] = []
for motion, label in [(u'大會審議', 'lbFmotion_0'), (u'程序會審定', 'lbFmotion_v'), (u'大會決定', 'lbFmotion_1'), (u'分組審查', 'lbFmotion_g'), (u'大會決議', 'lbFmotion_2')]:
date = response.css(u'#%sdate::text' % label).extract_first()
resolution = response.css(u'#%sopinion::text' % label).extract_first()
if date and resolution:
item['motions'].append(dict(zip(['motion', 'resolution', 'date'], [motion, resolution.strip(), common.ROC2AD(date)])))
item['links'] = [
{
'url': response.url,
'note': 'original'
}
]
return item
