def get_post_ids(self, board_id, page_no):
"""
Returns the post IDs in order they appear in the markup, which are
assummed to be in decending order, newest first.
"""
response = self._browser.get_page('board', board_id=board_id, page_no=page_no)
doc = parse_html(response.content)
post_ids = []
for trow in doc.cssselect('tr.mytr'):
tcells = trow.findall('td')
if not tcells:
self.logger.warning('no td elements in tr.mytr')
continue
views = tcells[-1].text_content()
if views == '-':
# Deleted post
continue
post_id_str = tcells[0].text_content()
try:
post_id = int(post_id_str)
except ValueError:
self.logger.warning('Post ID is not an integer: %s', post_id_str)
continue
post_ids.append(post_id)
return post_ids
def get_times(self, mid, fid, tid, date):
resp = self._browser.get_page("time_list", mid=mid, fid=fid, tid=tid, date=date)
doc = parse_html(resp.content)
items = []
for a in doc.cssselect("ul.time-list > li.time-list > a"):
try:
href = a.attrib["href"]
id = re.search(r"ScreenCd=(\d+)", href, re.I).group(1)
play_no = int(re.search(r"PlayNum=(\d+)", href, re.I).group(1))
m = re.search(
ur"""
(?P<time>\d{2}:\d{2})
\s*-\s*
(?P<screen_name>[^/]+)/(?P<play_no>\d+)회
\s*
\(좌석:(?P<current_seats>\d+)/(?P<total_seats>\d+)\)
""",
a.text_content(),
re.VERBOSE | re.I,
)
screen_name = m.group("screen_name")
time = m.group("time")
current_seats = int(m.group("current_seats"))
total_seats = int(m.group("total_seats"))
except (IndexError, KeyError, AttributeError, ValueError):
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(PlayState(Play(id, play_no, screen_name, time), current_seats, total_seats))
return items
def get_post(self, board_id, post_id):
response = self._browser.get_page('post', board_id=board_id, post_id=post_id)
doc = parse_html(response.content)
for script in doc.xpath('//script/text()'):
if u'잘못된접근입니다' in re.sub(r'\s', '', script) and 'history.go(-1)' in script:
# post is deleted
return
try:
title = doc.cssselect('.board_main > .view_title h4')[0].text_content()
content_dom = doc.cssselect('.board_main > .view_content #writeContents')[0]
content = content_dom.text_content()
content_html = lxml.html.tostring(content_dom, encoding='utf-8').decode(
'utf-8', 'replace')
# author_info is only shown when logged in
author, author_id = (None, None)
author_link = doc.cssselect('.board_main > .view_head > p.user_info > a[onclick]')
if author_link:
author, author_id = self._get_author_info(author_link[0].attrib['onclick'])
published_str = doc.cssselect('.board_main > .view_head > p.post_info')[0] \
.text_content()
published = self._get_published_time(published_str)
except (IndexError, AttributeError):
self.logger.warning('Markup is not as expected', exc_info=True)
else:
return Post(response.url, title, content, content_html, author, author_id, published)
def _login(self, username, password):
data = dict(mb_id=username, mb_password=password, url='/')
url = ClienBrowser.BASE_URL + ClienBrowser.LOGIN_PATH
self._request(url, method='post', data=data)
response = self._request(ClienBrowser.BASE_URL + '/')
doc = parse_html(response.content)
if not doc.cssselect('#account .uid'):
raise LoginError()
def _login(self, username, password):
login_url = CGVBrowser.BASE_URL + CGVBrowser.LOGIN_PATH
login_page = self._request(login_url)
doc = parse_html(login_page.content)
try:
viewstate = doc.cssselect('[name="__VIEWSTATE"]')[0].attrib["value"]
eventvalidation = doc.cssselect('[name="__EVENTVALIDATION"]')[0].attrib["value"]
except (IndexError, AttributeError):
raise LoginError("Cannot find necessary tokens")
data = {
"__VIEWSTATE": viewstate,
"__EVENTVALIDATION": eventvalidation,
"Login$tbUserID": username,
"Login$tbPassword": password,
"Login$ibLogin.x": 45,
"Login$ibLogin.y": 37,
}
resp = self._request(login_url, method="post", data=data)
doc = parse_html(resp.content)
if not doc.cssselect("body"):
raise LoginError()
def get_seats(self, tid, date, sid, play_no):
resp = self._browser.get_page("seat_list", tid=tid, date=date, sid=sid, play_no=play_no)
doc = parse_html(resp.content)
items = []
for td in doc.cssselect("#seat_table td.pointer"):
try:
seat = td.attrib["seatname"]
except KeyError:
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(seat)
return items
def get_dates(self, mid, fid, tid):
resp = self._browser.get_page("date_list", mid=mid, fid=fid, tid=tid)
doc = parse_html(resp.content)
items = []
for a in doc.cssselect("table.month td:not(.disabled) a"):
try:
date = re.search(r"PlayYmd=(\d{8})", a.attrib["href"], re.I).group(1)
except (IndexError, KeyError, AttributeError):
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(date)
return items
def get_formats(self, mid):
resp = self._browser.get_page("format_list", mid=mid)
doc = parse_html(resp.content)
items = []
for a in doc.cssselect("ul.movielist2 > li.movielist2 > a"):
try:
id = re.search(r"CgvCode=(\d+)", a.attrib["href"], re.I).group(1)
name = re.search(r"\(([^)]+)\)", a.text_content().strip(), re.I).group(1)
except (IndexError, KeyError, AttributeError):
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(Format(id=id, name=name))
return items
def get_movies(self):
resp = self._browser.get_page("movie_list")
doc = parse_html(resp.content)
items = []
for a in doc.cssselect("table.list tbody tr .subject a"):
try:
id = re.search(r"MovieIdx=(\d+)", a.attrib["href"], re.I).group(1)
name = a.text_content().strip()
except (IndexError, KeyError, AttributeError):
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(Movie(id=id, name=name))
return items
def get_theaters(self, mid, fid):
resp = self._browser.get_page("theater_list", mid=mid, fid=fid)
doc = parse_html(resp.content)
items = []
for a in doc.cssselect("#areaTheater ul.theaterlist > li.theaterlist > a"):
try:
regex = r"'(\d{{4}})'\s*,\s*'{0}'\s*,\s*'{1}'".format(fid, mid)
id = re.search(regex, a.attrib["href"], re.I).group(1)
name = a.text_content()
except (IndexError, KeyError, AttributeError):
self.logger.warning("Markup is not as expected", exc_info=True)
else:
items.append(Theater(id=id, name=name))
return items
def _login_required(self, response):
doc = parse_html(response.content)
scripts = doc.xpath('//script/text()')
return (len(scripts) >= 2 and
re.search(ur'alert.*로그인', scripts[0]) and
re.search(ur'location\.replace', scripts[1]))
def _login_required(self, resp):
doc = parse_html(resp.content)
return doc.cssselect(".member_login")
