def get_meeting(self, meeting_url=None, meeting_id=None):
"""
Load meeting details (e.g. agendaitems) for the given detail page URL or numeric ID
"""
meeting_url = "%sto010.asp?selfaction=ws&template=xyz&SILFDNR=%s" % (self.config.BASE_URL, meeting_id)
logging.info("Getting meeting %d from %s", meeting_id, meeting_url)
r = self.get_url(meeting_url)
if not r:
return
# If r.history has an item we have a problem
if len(r.history):
if r.history[0].status_code == 302:
logging.info("Meeting %d from %s seems to be private", meeting_id, meeting_id)
else:
logging.error("Strange redirect %d from %s with status code %s", meeting_id, meeting_url, r.history[0].status_code)
return
xml = r.text.encode('ascii','xmlcharrefreplace')
parser = etree.XMLParser(recover=True)
root = etree.fromstring(xml, parser=parser)
meeting = Meeting(numeric_id=meeting_id)
# special area
special = {}
for item in root[0].iterchildren():
special[item.tag] = item.text
# Woher kriegen wir das Datum? Nur über die Übersicht?
#if 'sisb' in special:
#if 'sise' in special:
if 'saname' in special:
if special['saname'] in self.config.MEETING_TYPE:
meeting.type = self.config.MEETING_TYPE[special['saname']]
else:
logging.warn("String '%s' not found in MEETING_TYPE", special['saname'])
# head area
head = {}
for item in root[0].iterchildren():
head[item.tag] = item.text
if 'raname' in head:
meeting.room = head['raname']
if 'raort' in head:
meeting.address = head['raort']
agendaitems = []
for item in root[2].iterchildren():
elem = {}
for e in item.iterchildren():
elem[e.tag] = e.text
section = [elem['tofnum'], elem['tofunum'], elem['tofuunum']]
section = [x for x in section if x!="0"]
elem['section'] = ".".join(section)
agendaitem = Agendaitem()
#agendaitem = elem['topnr']
agendaitem.numeric_id = int(elem['tolfdnr'])
if elem['toostLang'] == u'öffentlich':
agendaitem.public = True
else:
agendaitem.public = False
agendaitem.title = elem['totext1']
# get agenda detail page
# TODO: Own Queue
time.sleep(self.config.WAIT_TIME)
agendaitem_url = '%sto020.asp?selfaction=ws&template=xyz&TOLFDNR=%s' % (self.config.BASE_URL, agendaitem.numeric_id)
logging.info("Getting agendaitem %d from %s", agendaitem.numeric_id, agendaitem_url)
agendaitem_r = self.get_url(agendaitem_url)
if not agendaitem_r:
return
if len(agendaitem_r.history):
logging.info("Agenda item %d from %s seems to be private", meeting_id, meeting_url)
else:
agendaitem_xml = agendaitem_r.text.encode('ascii','xmlcharrefreplace')
agendaitem_parser = etree.XMLParser(recover=True)
agendaitem_root = etree.fromstring(agendaitem_xml, parser=parser)
add_agenda_item = {}
for add_item in agendaitem_root[0].iterchildren():
if add_item.tag == "rtfWP" and len(add_item) > 0:
try:
agendaitem.resolution_text = etree.tostring(add_item[0][1][0])
except:
logging.warn("Unable to parse resolution text at %s", agendaitem_url)
else:
add_agenda_item[add_item.tag] = add_item.text
if 'voname' in add_agenda_item:
# create paper with identifier
agendaitem.paper = [Paper(numeric_id = int(elem['volfdnr']), title=add_agenda_item['voname'])]
if add_agenda_item['vobetr'] != agendaitem.title:
logging.warn("different values for title: %s and %s", agendaitem.title, add_agenda_item['vobetr'])
if hasattr(self, 'paper_queue'):
self.paper_queue.add(int(elem['volfdnr']))
elif int(elem['volfdnr']) is not 0:
# create paper without identifier
agendaitem.paper = [Paper(numeric_id = int(elem['volfdnr']))]
if hasattr(self, 'paper_queue'):
self.paper_queue.add(int(elem['volfdnr']))
if "nowDate" not in add_agenda_item:
# something is broken with this so we don't store it
logging.warn("Skipping broken agenda at ", agendaitem_url)
else:
# dereference result
if add_agenda_item['totyp'] in self.config.RESULT_STRINGS:
agendaitem.result = self.config.RESULT_STRINGS[add_agenda_item['totyp']]
else:
logging.warn("String '%s' not found in configured RESULT_STRINGS", add_agenda_item['totyp'])
agendaitems.append(agendaitem)
meeting.agendaitem = agendaitems
oid = self.db.save_meeting(meeting)
logging.info("Meeting %d stored with _id %s", meeting_id, oid)
def get_meeting(self, meeting_url=None, meeting_id=None):
"""
Load meeting details for the given detail page URL or numeric ID
"""
# Read either meeting_id or meeting_url from the opposite
if meeting_id is not None:
meeting_url = self.urls['SESSION_DETAIL_PRINT_PATTERN'] % meeting_id
elif meeting_url is not None:
parsed = parse.search(self.urls['SESSION_DETAIL_PARSE_PATTERN'], meeting_url)
meeting_id = parsed['meeting_id']
logging.info("Getting meeting (session) %d from %s", meeting_id, meeting_url)
meeting = Meeting(numeric_id=meeting_id)
time.sleep(self.config.WAIT_TIME)
response = self.get_url(meeting_url)
if not response:
return
# forms for later document download
mechanize_forms = mechanize.ParseResponse(response, backwards_compat=False)
# seek(0) is necessary to reset response pointer.
response.seek(0)
html = response.read()
html = html.replace(' ', ' ')
parser = etree.HTMLParser()
dom = etree.parse(StringIO(html), parser)
# check for page errors
try:
page_title = dom.xpath('//h1')[0].text
if 'Fehlermeldung' in page_title:
logging.info("Page %s cannot be accessed due to server error", meeting_url)
return
if 'Berechtigungsfehler' in page_title:
logging.info("Page %s cannot be accessed due to permissions", meeting_url)
return
except:
pass
try:
error_h3 = dom.xpath('//h3[@class="smc_h3"]')[0].text.strip()
if 'Keine Daten gefunden' in error_h3:
logging.info("Page %s does not contain any agenda items", meeting_url)
return
if 'Fehlercode: 1104' in error_h3:
logging.info("Page %s cannot be accessed due to permissions", meeting_url)
return
except:
pass
meeting.original_url = meeting_url
# Session title
try:
meeting.title = dom.xpath(self.xpath['SESSION_DETAIL_TITLE'])[0].text
except:
logging.critical('Cannot find session title element using XPath SESSION_DETAIL_TITLE')
raise TemplateError('Cannot find session title element using XPath SESSION_DETAIL_TITLE')
# Committe link
#try:
# links = dom.xpath(self.xpath['SESSION_DETAIL_COMMITTEE_LINK'])
# for link in links:
# href = link.get('href')
# parsed = parse.search(self.urls['COMMITTEE_DETAIL_PARSE_PATTERN'], href)
# if parsed is not None:
# meeting.committees = [Commitee(numeric_id=int(parsed['committee_id']))]
# if hasattr(self, 'committee_queue'):
# self.committee_queue.add(int(parsed['committee_id']))
#except:
# logging.critical('Cannot find link to committee detail page using SESSION_DETAIL_COMMITTEE_LINK_XPATH')
# raise TemplateError('Cannot find link to committee detail page using SESSION_DETAIL_COMMITTEE_LINK_XPATH')
# Meeting identifier, date, address etc
tds = dom.xpath(self.xpath['SESSION_DETAIL_IDENTIFIER_TD'])
if len(tds) == 0:
logging.critical('Cannot find table fields using SESSION_DETAIL_IDENTIFIER_TD_XPATH at session ' + meeting_url)
raise TemplateError('Cannot find table fields using SESSION_DETAIL_IDENTIFIER_TD_XPATH at session ' + meeting_url)
else:
for n in range(0, len(tds)):
try:
tdcontent = tds[n].text.strip()
nextcontent = tds[n + 1].text.strip()
except:
continue
if tdcontent == 'Sitzung:':
meeting.identifier = nextcontent
# We don't need this any more because it's scraped in committee detail page(?)
#elif tdcontent == 'Gremium:':
# meeting.committee_name = nextcontent
elif tdcontent == 'Datum:':
start = nextcontent
end = nextcontent
if tds[n + 2].text == 'Zeit:':
if tds[n + 3].text is not None:
times = tds[n + 3].text.replace(' Uhr', '').split('-')
start = start + ' ' + times[0]
if len(times) > 1:
end = end + ' ' + times[1]
else:
end = start
meeting.start = start
meeting.end = end
elif tdcontent == 'Raum:':
meeting.address = " ".join(tds[n + 1].xpath('./text()'))
elif tdcontent == 'Bezeichnung:':
meeting.description = nextcontent
if not hasattr(meeting, 'identifier'):
logging.critical('Cannot find session identifier using XPath SESSION_DETAIL_IDENTIFIER_TD')
raise TemplateError('Cannot find session identifier using XPath SESSION_DETAIL_IDENTIFIER_TD')
# Agendaitems
found_documents = []
rows = dom.xpath(self.xpath['SESSION_DETAIL_AGENDA_ROWS'])
if len(rows) == 0:
logging.critical('Cannot find agenda using XPath SESSION_DETAIL_AGENDA_ROWS')
raise TemplateError('Cannot find agenda using XPath SESSION_DETAIL_AGENDA_ROWS')
meeting.agendaitem = []
else:
agendaitems = []
agendaitem_id = None
public = True
agendaitem = None
for row in rows:
row_id = row.get('id')
row_classes = row.get('class').split(' ')
fields = row.xpath('td')
number = fields[0].xpath('./text()')
if len(number) > 0:
number = number[0]
else:
# when theres a updated notice theres an additional spam
number = fields[0].xpath('.//span/text()')
if len(number) > 0:
number = number[0]
if number == []:
number = None
if row_id is not None:
# Agendaitem main row
# first: save agendaitem from before
if agendaitem:
agendaitems.append(agendaitem)
# create new agendaitem
agendaitem = Agendaitem(numeric_id=int(row_id.rsplit('_', 1)[1]))
if number is not None:
agendaitem.sequence_number = number
# in some ris this is a link, sometimes not. test both.
if len(fields[1].xpath('./a/text()')):
agendaitem.title = "; ".join(fields[1].xpath('./a/text()'))
elif len(fields[1].xpath('./text()')):
agendaitem.title = "; ".join(fields[1].xpath('./text()'))
# ignore no agendaitem information
if agendaitem.title == 'keine Tagesordnungspunkte':
agendaitem = None
continue
agendaitem.public = public
# paper links
links = row.xpath(self.xpath['SESSION_DETAIL_AGENDA_ROWS_SUBMISSION_LINK'])
papers = []
for link in links:
href = link.get('href')
if href is None:
continue
parsed = parse.search(self.urls['SUBMISSION_DETAIL_PARSE_PATTERN'], href)
if parsed is not None:
paper = Paper(numeric_id=int(parsed['paper_id']), identifier=link.text)
papers.append(paper)
# Add paper to paper queue
if hasattr(self, 'paper_queue'):
self.paper_queue.add(int(parsed['paper_id']))
if len(papers):
agendaitem.paper = papers
"""
Note: we don't scrape agendaitem-related documents for now,
based on the assumption that they are all found via paper
detail pages. All we do here is get a list of document IDs
in found_documents
"""
# find links
links = row.xpath('.//a[contains(@href,"getfile.")]')
for link in links:
if not link.xpath('.//img'):
file_link = self.config.BASE_URL + link.get('href')
document_id = file_link.split('id=')[1].split('&')[0]
found_documents.append(document_id)
# find forms
forms = row.xpath('.//form')
for form in forms:
for hidden_field in form.xpath('input'):
if hidden_field.get('name') != 'DT':
continue
document_id = hidden_field.get('value')
found_documents.append(document_id)
# Alternative für smc_tophz wegen Version 4.3.5 bi (Layout 3)
elif ('smc_tophz' in row_classes) or (row.get('valign') == 'top' and row.get('debug') == '3'):
# additional (optional row for agendaitem)
label = fields[1].text
value = fields[2].text
if label is not None and value is not None:
label = label.strip()
value = value.strip()
if label in ['Ergebnis:', 'Beschluss:', 'Beratungsergebnis:']:
if value in self.config.RESULT_STRINGS:
agendaitem.result = self.config.RESULT_STRINGS[value]
else:
logging.warn("String '%s' not found in configured RESULT_STRINGS", value)
agendaitem.result = value
elif label in ['Bemerkung:', 'Abstimmung:']:
agendaitem.result_details = value
# What's this?
#elif label == 'Abstimmung:':
# agendaitems[agendaitem_id]['voting'] = value
else:
logging.critical("Agendaitem info label '%s' is unknown", label)
raise ValueError('Agendaitem info label "%s" is unknown' % label)
elif 'smcrowh' in row_classes:
# Subheading (public / nonpublic part)
if fields[0].text is not None and "Nicht öffentlich" in fields[0].text.encode('utf-8'):
public = False
meeting.agendaitem = agendaitems
# meeting-related documents
containers = dom.xpath(self.xpath['SESSION_DETAIL_ATTACHMENTS'])
for container in containers:
classes = container.get('class')
if classes is None:
continue
classes = classes.split(' ')
if self.xpath['SESSION_DETAIL_ATTACHMENTS_CONTAINER_CLASSNAME'] not in classes:
continue
documents = []
rows = container.xpath('.//tr')
for row in rows:
if not row.xpath('.//form'):
links = row.xpath('.//a')
for link in links:
# ignore additional pdf icon links
if not link.xpath('.//img'):
title = ' '.join(link.xpath('./text()')).strip()
file_link = self.config.BASE_URL + link.get('href')
document_id = file_link.split('id=')[1].split('&')[0]
if document_id in found_documents:
continue
document = Document(
identifier=document_id,
numeric_id=document_id,
title=title,
original_url=file_link)
document = self.get_document_file(document=document, link=file_link)
if 'Einladung' in title:
document_type = 'invitation'
elif 'Niederschrift' in title:
document_type = 'results_protocol'
else:
document_type = 'misc'
documents.append({'relation': document_type, 'document': document})
found_documents.append(document_id)
else:
forms = row.xpath('.//form')
for form in forms:
title = " ".join(row.xpath('./td/text()')).strip()
for hidden_field in form.xpath('input'):
if hidden_field.get('name') != 'DT':
continue
document_id = hidden_field.get('value')
# make sure to add only those which aren't agendaitem-related
if document_id not in found_documents:
document = Document(
identifier=document_id,
numeric_id=document_id,
title=title
)
# Traversing the whole mechanize response to submit this form
for mform in mechanize_forms:
for control in mform.controls:
if control.name == 'DT' and control.value == document_id:
document = self.get_document_file(document, mform)
if 'Einladung' in title:
document_type = 'invitation'
elif 'Niederschrift' in title:
document_type = 'results_protocol'
else:
document_type = 'misc'
documents.append({'relation': document_type, 'document': document})
found_documents.append(document_id)
if len(documents):
meeting.document = documents
oid = self.db.save_meeting(meeting)
logging.info("Meeting %d stored with _id %s", meeting_id, oid)