def contests():
"""
Show the upcoming contests
"""
today = datetime.datetime.today()
today = datetime.datetime.strptime(str(today)[:-7], "%Y-%m-%d %H:%M:%S")
start_date = today.date()
end_date = start_date + datetime.timedelta(90)
url = "https://contesttrackerapi.herokuapp.com/"
response = requests.get(url)
if response.status_code == 200:
response = response.json()["result"]
else:
# @todo: something better
return dict()
ongoing = response["ongoing"]
upcoming = response["upcoming"]
contests = []
cal = pdt.Calendar()
table = TABLE(_class="centered striped")
thead = THEAD(
TR(TH("Contest Name"), TH("Site"), TH("Start"), TH("Duration/Ending"),
TH("Link"), TH("Add Reminder")))
table.append(thead)
tbody = TBODY()
for i in ongoing:
if i["Platform"] in ("TOPCODER", "OTHER"):
continue
try:
endtime = datetime.datetime.strptime(i["EndTime"],
"%a, %d %b %Y %H:%M")
except ValueError:
continue
tr = TR()
span = SPAN(_class="green tooltipped",
data={"position": "right",
"delay": "50",
"tooltip": "Live Contest"},
_style="cursor: pointer; " + \
"float:right; " + \
"height:10px; " + \
"width:10px; " + \
"border-radius: 50%;")
tr.append(TD(i["Name"], span))
tr.append(TD(i["Platform"].capitalize()))
tr.append(TD("-"))
tr.append(TD(str(endtime).replace("-", "/"),
_class="contest-end-time"))
tr.append(
TD(
A(I(_class="fa fa-external-link-square fa-lg"),
_class="btn-floating btn-small green accent-4 tooltipped",
_href=i["url"],
data={
"position": "left",
"tooltip": "Contest Link",
"delay": "50"
},
_target="_blank")))
tr.append(
TD(
BUTTON(
I(_class="fa fa-calendar-plus-o"),
_class=
"btn-floating btn-small orange accent-4 tooltipped disabled",
data={
"position": "left",
"tooltip": "Already started!",
"delay": "50"
})))
tbody.append(tr)
# This id is used for uniquely identifying
# a particular contest in js
button_id = 1
for i in upcoming:
if i["Platform"] in ("TOPCODER", "OTHER"):
continue
start_time = datetime.datetime.strptime(i["StartTime"],
"%a, %d %b %Y %H:%M")
tr = TR(_id="contest-" + str(button_id))
tr.append(TD(i["Name"]))
tr.append(TD(i["Platform"].capitalize()))
tr.append(TD(str(start_time)))
duration = i["Duration"]
duration = duration.replace(" days", "d")
duration = duration.replace(" day", "d")
tr.append(TD(duration))
tr.append(
TD(
A(I(_class="fa fa-external-link-square fa-lg"),
_class="btn-floating btn-small green accent-4 tooltipped",
_href=i["url"],
data={
"position": "left",
"tooltip": "Contest Link",
"delay": "50"
},
_target="_blank")))
tr.append(
TD(
BUTTON(
I(_class="fa fa-calendar-plus-o"),
_class="btn-floating btn-small orange accent-4 tooltipped",
data={
"position": "left",
"tooltip": "Set Reminder to Google Calendar",
"delay": "50"
},
_id="set-reminder-" + str(button_id))))
tbody.append(tr)
button_id += 1
table.append(tbody)
return dict(table=table, upcoming=upcoming)
def __init__(self):
self._db = DatabaseManager()
self._pdt = parsedatetime.Calendar(parsedatetime.Constants(
self.LOCALE))
def conv_dt(dt_string):
cal = parsedatetime.Calendar()
dt_obj, _ = cal.parseDT(datetimeString=dt_string)
return dt_obj
def process_message(config, data, event, context):
logger = logging.getLogger('pubsub2inbox')
# Ignore messages submitted before our retry period
retry_period = '2 days ago'
if 'retryPeriod' in config:
retry_period = config['retryPeriod']
retry_period_parsed = parsedatetime.Calendar().parse(retry_period)
if len(retry_period_parsed) > 1:
retry_earliest = datetime.fromtimestamp(mktime(retry_period_parsed[0]),
timezone.utc)
else:
retry_earliest = datetime.fromtimestamp(mktime(retry_period_parsed),
timezone.utc)
message_time = parser.parse(context.timestamp)
if (message_time - retry_earliest) < timedelta(0, 0):
logger.warning('Ignoring message because it\'s past the retry period.',
extra={
'event_id': context.event_id,
'retry_period': retry_period,
'retry_earliest': retry_earliest.strftime('%c'),
'event_timestamp': message_time
})
raise MessageTooOldException(
'Ignoring message because it\'s past the retry period.')
template_variables = {
'data': data,
'event': event,
'context': context,
}
jinja_environment = get_jinja_environment()
if 'processors' in config:
for processor in config['processors']:
logger.debug('Processing message using input processor: %s' %
processor)
mod = __import__('processors.%s' % processor)
processor_module = getattr(mod, processor)
processor_class = getattr(processor_module,
'%sProcessor' % processor.capitalize())
processor_instance = processor_class(config, jinja_environment,
data, event, context)
processor_variables = processor_instance.process()
template_variables.update(processor_variables)
jinja_environment.globals = {
**jinja_environment.globals,
**template_variables
}
if 'processIf' in config:
processif_template = jinja_environment.from_string(config['processIf'])
processif_template.name = 'processif'
processif_contents = processif_template.render()
if processif_contents.strip() == '':
logger.info(
'Will not send message because processIf evaluated to empty.')
return
if 'resendBucket' in config:
if 'resendPeriod' not in config:
raise NoResendConfigException(
'No resendPeriod configured, even though resendBucket is set!')
resend_key_hash = hashlib.sha256()
if 'resendKey' not in config:
default_resend_key = template_variables.copy()
default_resend_key.pop('context')
resend_key_hash.update(
json.dumps(default_resend_key).encode('utf-8'))
else:
key_template = jinja_environment.from_string(config['resendKey'])
key_template.name = 'resend'
key_contents = key_template.render()
resend_key_hash.update(key_contents.encode('utf-8'))
resend_file = resend_key_hash.hexdigest()
logger.debug('Checking for resend object in bucket...',
extra={
'bucket': config['resendBucket'],
'blob': resend_file
})
client_info = grpc_client_info.ClientInfo(
user_agent='google-pso-tool/pubsub2inbox/1.1.0')
storage_client = storage.Client(client_info=client_info)
bucket = storage_client.bucket(config['resendBucket'])
resend_blob = bucket.blob(resend_file)
if resend_blob.exists():
resend_blob.reload()
resend_period = config['resendPeriod']
resend_period_parsed = parsedatetime.Calendar().parse(
resend_period, sourceTime=resend_blob.time_created)
if len(resend_period_parsed) > 1:
resend_earliest = datetime.fromtimestamp(
mktime(resend_period_parsed[0]))
else:
resend_earliest = datetime.fromtimestamp(
mktime(resend_period_parsed))
if datetime.now() >= resend_earliest:
logger.debug('Resending the message now.',
extra={
'resend_earliest': resend_earliest,
'blob_time_created': resend_blob.time_created
})
resend_blob.upload_from_string('')
else:
logger.info(
'Can\'t resend the message now, resend period not elapsed.',
extra={
'resend_earliest': resend_earliest,
'blob_time_created': resend_blob.time_created
})
return
else:
try:
resend_blob.upload_from_string('', if_generation_match=0)
except Exception as exc:
# Handle TOCTOU condition
if 'conditionNotMet' in str(exc):
logger.warning(
'Message (re)sending already in progress (resend key already exist).',
extra={'exception': exc})
return
else:
raise exc
return
if 'outputs' in config:
for output_config in config['outputs']:
if 'type' not in output_config:
raise NoTypeConfiguredException(
'No type configured for output!')
if 'processIf' in output_config:
processif_template = jinja_environment.from_string(
output_config['processIf'])
processif_template.name = 'processif'
processif_contents = processif_template.render()
if processif_contents.strip() == '':
logger.info(
'Will not use output processor %s because processIf evaluated to empty.'
% output_config['type'])
continue
logger.debug('Processing message using output processor: %s' %
output_config['type'])
output_type = output_config['type']
mod = __import__('output.%s' % output_type)
output_module = getattr(mod, output_type)
output_class = getattr(output_module,
'%sOutput' % output_type.capitalize())
output_instance = output_class(config, output_config,
jinja_environment, data, event,
context)
try:
output_instance.output()
except Exception as exc:
logger.error('Output processor %s failed, trying next...' %
(output_type),
extra={'exception': traceback.format_exc()})
if 'allOutputsMustSucceed' in config and config[
'allOutputsMustSucceed']:
raise exc
else:
raise NoOutputsConfiguredException('No outputs configured!')
def __init__(self, date_format=None, **kwargs):
super(Date, self).__init__(**kwargs)
self.date_format = date_format
self.parser = parsedatetime.Calendar(
version=parsedatetime.VERSION_CONTEXT_STYLE)
def __call__(self, args):
try:
os.makedirs(args.directory)
except:
pass
since = None
if args.since:
since = parsedatetime.Calendar().parse(args.since)
since = time.mktime(since[0])
since = datetime.datetime.fromtimestamp(since)
if args.labels:
font = ImageFont.truetype("arial.ttf", 14)
else:
font = None
workers = session.query(turkic.models.Worker)
for worker in workers:
print "Sampling worker {0}".format(worker.id)
jobs = session.query(Job)
jobs = jobs.filter(Job.worker == worker)
jobs = jobs.join(Segment)
jobs = jobs.join(Video)
jobs = jobs.filter(Video.isfortraining == False)
if since:
jobs = jobs.filter(turkic.models.HIT.timeonserver >= since)
jobs = jobs.order_by(sqlalchemy.func.rand())
jobs = jobs.limit(args.number)
for job in jobs:
print "Visualizing HIT {0}".format(job.hitid)
paths = [x.getboxes(interpolate = True,
bind = True,
label = True) for x in job.paths]
if args.frames > job.segment.stop - job.segment.start:
frames = range(job.segment.start, job.segment.stop + 1)
else:
frames = random.sample(xrange(job.segment.start,
job.segment.stop + 1),
args.frames)
size = math.sqrt(len(frames))
video = job.segment.video
bannersize = (video.width * int(math.floor(size)),
video.height * int(math.ceil(size)))
image = Image.new(video[0].mode, bannersize)
size = int(math.floor(size))
offset = (0, 0)
horcount = 0
paths = vision.visualize.highlight_paths(video, paths,
font = font)
for frame, framenum in paths:
if framenum in frames:
image.paste(frame, offset)
horcount += 1
if horcount >= size:
offset = (0, offset[1] + video.height)
horcount = 0
else:
offset = (offset[0] + video.width, offset[1])
image.save("{0}/{1}-{2}.jpg".format(args.directory,
worker.id,
job.hitid))
# limitations under the License.
"""For processing BigQuery timestamp dates."""
import datetime
import logging
import re
import parsedatetime
# Turn off the chatty parsedatetime module's logging
logging.getLogger('parsedatetime').setLevel(logging.ERROR)
# Initialize parsedatetime
pdt_constants = parsedatetime.Constants()
pdt_constants.BirthdayEpoch = 50 # TODO(user) provide a way to set this.
pdt = parsedatetime.Calendar(pdt_constants)
# e.g. 1989-10-02 05:23:48 1958-06-24T12:18:35.5803 1988-08-15T19:06:56.235
TIMESTAMP_RE = re.compile(r'^\d{4}-(0[1-9]|1[0-2])-(0[1-9]|[12]\d|3[01])'
r'[ T]([01]\d|2[0-3]):[0-5]\d'
r':([0-5]\d|60)(\.\d{3,4})?' # Leap seconds!
r'( [+-][012]\d:[0-5]\d)?$')
# YYYY-MM-DD HH:MM:SS.micro +08:00
OUTPUT_TIMESTAMP_FORMAT = '%Y-%m-%d %H:%M:%S.%f %z'
INPUT_TIMESTAMP_FORMATS = (
'%Y-%m-%d %H:%M:%S.%f %z',
'%Y/%m/%d %H:%M:%S.%f %z',
'%m/%d/%Y %H:%M:%S.%f %z',
'%m/%d/%y %H:%M:%S.%f %z',
async def mute(self, ctx, *, member = None, cooldown = None):
"""Mencegah member untuk mengirim pesan dalam chat atau berbicara dalam voice channel(bot-admin/admin only)."""
# isOwner = self.settings.isOwner(ctx.author)
# if isOwner == False:
# await ctx.send("Fitur ini sedang dalam tahap ujicoba")
# return
if not await Utils.is_bot_admin_reply(ctx): return
if member == None:
em = discord.Embed(color = 0XFF8C00, description = "Mencegah member untuk mengirim pesan dalam chat atau berbicara dalam voice channel\n\n"
"**Panduan**\n"
"`{}mute [member] [cooldown]`"
.format(ctx.prefix))
em.set_footer(text = "{}#{}".format(ctx.author.name, ctx.author.discriminator), icon_url = f"{ctx.author.avatar_url}")
return await ctx.send(embed=em)
# Let's search for a name at the beginning - and a time at the end
parts = member.split()
for j in range(len(parts)):
# Reverse search direction
i = len(parts)-1-j
memFromName = None
endTime = None
# Name = 0 up to i joined by space
nameStr = ' '.join(parts[0:i+1])
# Time = end of name -> end of parts joined by space
timeStr = ' '.join(parts[i+1:])
memFromName = DisplayName.memberForName(nameStr, ctx.guild)
if memFromName:
# We got a member - let's check for time
# Get current time - and end time
try:
# Get current time - and end time
currentTime = int(time.time())
cal = parsedatetime.Calendar()
time_struct, parse_status = cal.parse(timeStr)
start = datetime(*time_struct[:6])
end = time.mktime(start.timetuple())
# Get the time from now to end time
endTime = end-currentTime
except:
pass
if not endTime == None:
# We got a member and a time - break
break
if memFromName == None:
# We couldn't find one or the other
em = discord.Embed(color = 0XFF8C00, description = "> Mencegah member untuk mengirim pesan dalam chat atau berbicara dalam voice channel\n> \n"
"> **Panduan**\n"
"> `{}mute [member] [cooldown]`"
.format(ctx.prefix))
em.set_author(name = "Oops!", url = "https://acinonyxesports.com/", icon_url = "https://cdn.discordapp.com/attachments/518118753226063887/725569194304733435/photo.jpg")
em.set_footer(text = f"Request By : {ctx.author.name}", icon_url = f"{ctx.author.avatar_url}")
msg = 'Usage: `{}mute [member] [cooldown]`'.format(ctx.prefix)
return await ctx.send(msg)
cooldown = None if endTime == 0 else endTime
member = memFromName
# Check if we're muting ourself
if member is ctx.author:
msg = 'Akan lebih mudah untuk ku kalo kamu sendiri yang diam!'
em = discord.Embed(color = 0XFF8C00, description = msg)
em.set_footer(text = "{}".format(ctx.author),
icon_url = f"{ctx.author.avatar_url}")
return await ctx.send(embed = em)
# Check if we're muting the bot
if member.id == self.bot.user.id:
msg = '┐(￣ヘ￣;)┌\nAku nggk mau mute diri sendiri.'
em = discord.Embed(color = 0XFF8C00, description = msg)
em.set_footer(text = "{}".format(ctx.author),
icon_url = f"{ctx.author.avatar_url}")
return await ctx.send(embed = em)
# Check if member is admin or bot admin
if await Utils.is_bot_admin_reply(ctx,member=member,message="┐(￣ヘ￣;)┌\nKamu tidak dapat melakukan mute pada admin lain.",message_when=True): return
# Set cooldown - or clear it
if type(cooldown) is int or type(cooldown) is float:
if cooldown < 0:
msg = '┐(￣ヘ￣;)┌\nCooldown tidak dapat dilakukan dengan angka negatif!'
em = discord.Embed(color = 0XFF8C00, description = msg)
em.set_footer(text = "{}#{}".format(ctx.author),
icon_url = f"{ctx.author.avatar_url}")
return await ctx.send(embed = em)
currentTime = int(time.time())
cooldownFinal = currentTime+cooldown
else:
cooldownFinal = None
# Check if we're using the old mute and suggest the quicker version
try: role = ctx.guild.get_role(int(self.settings.getServerStat(ctx.guild,"MuteRole")))
except: role = None
msg = "Muting...{}".format(
"" if role else "\n\nKamu dapat membuat memilih role dengan cara mengetik `{}setmuterole [role]`\nAtau `{}createmuterole [role_name]` untuk mute member ***lebih cepat***.".format(ctx.prefix,ctx.prefix)
)
em = discord.Embed(color = 0XFF8C00, description = msg)
em.set_footer(text = "{}".format(ctx.author), icon_url = "{}".format(ctx.author.avatar_url))
mess = await ctx.send(embed = em)
# Do the actual muting
await self._mute(member, ctx.guild, cooldownFinal, ctx.author)
if cooldown:
mins = "menit"
checkRead = ReadableTime.getReadableTimeBetween(currentTime, cooldownFinal)
msg = '*{}* telah di **Mute** hingga *{}*.'.format(DisplayName.name(member), checkRead)
# pm = 'You have been **Muted** by *{}* for *{}*.\n\nYou will not be able to send messages on *{}* until either that time has passed, or you have been **Unmuted**.'.format(DisplayName.name(ctx.author), checkRead, Utils.suppressed(ctx, ctx.guild.name))
else:
msg = '*{}* telah di **Mute** *hingga pemberitahuan lebih lanjut*.'.format(DisplayName.name(member))
# pm = 'You have been **Muted** by *{}* *until further notice*.\n\nYou will not be able to send messages on *{}* until you have been **Unmuted**.'.format(DisplayName.name(ctx.author), Utils.suppressed(ctx, ctx.guild.name))
await mess.edit(content=Utils.suppressed(ctx,msg))
'''try:
def run():
# print("Test")
from bs4 import BeautifulSoup
import requests
from main_site.models import NewsItem
import parsedatetime
from datetime import datetime
print("-" * 100)
print("Hackernews parser")
print("-" * 100)
# url="https://news.ycombinator.com/"
urls = [
"https://news.ycombinator.com/news?p=1",
"https://news.ycombinator.com/news?p=2",
"https://news.ycombinator.com/news?p=3"
]
# res=requests.get(url,stream=True)
tr1s = []
tr2s = []
for u in urls:
res = requests.get(u)
html = res.content
soup = BeautifulSoup(html)
tr1s_tmp = []
tr1s_tmp = soup.findAll('tr', {'class': 'athing'})
tr1s += tr1s_tmp
for tr1 in tr1s_tmp:
tr2s.append(tr1.findNextSibling())
url = []
title = []
hacker_news_url = []
upvote_count = []
comment_count = []
posted_on = []
for i in range(len(tr1s)):
url.append(tr1s[i].select_one('td:nth-of-type(3) > a').get('href'))
title.append(tr1s[i].select_one('td:nth-of-type(3) > a').get_text())
# hacker_news_url.append(tr1s[i].select_one('td:nth-of-type(3) > span > a').get('href'))
# hacker_news_url.append("https://news.ycombinator.com/"+tr2s[i].select_one('td:nth-of-type(2) > a:nth-of-type(3)').get('href'))
if tr2s[i].select_one('td:nth-of-type(2) > span.age > a') == None:
hacker_news_url.append('')
else:
hacker_news_url.append(
"https://news.ycombinator.com/" + tr2s[i].select_one(
'td:nth-of-type(2) > span.age > a').get('href'))
if tr2s[i].select_one('td:nth-of-type(2) > span.score') == None:
upvote_count.append('0')
else:
upvote_count.append(tr2s[i].select_one(
'td:nth-of-type(2) > span.score').get_text().split()[0])
#exception handling
if tr2s[i].select_one('td:nth-of-type(2) > a:nth-of-type(3)') == None:
comment_count.append('0')
elif tr2s[i].select_one('td:nth-of-type(2) > a:nth-of-type(3)'
).get_text() == "discuss":
comment_count.append('0')
else:
comment_count.append(tr2s[i].select_one(
'td:nth-of-type(2) > a:nth-of-type(3)').get_text().split()[0])
if tr2s[i].select_one('td:nth-of-type(2) > span.age > a') == None:
posted_on.append('')
else:
posted_on.append(tr2s[i].select_one(
'td:nth-of-type(2) > span.age > a').get_text())
# test=['7 hours ago', '58 minutes ago', '13 hours ago']
cal = parsedatetime.Calendar()
date_list = []
date_db = []
for date_str in posted_on:
time_struct, parse_status = cal.parse(date_str)
res = datetime(*time_struct[:6])
# date_list.append(res)
date_db.append(res.strftime('%Y-%m-%d %H:%M:%S'))
# result = list(reversed([x for _,x in sorted(zip(date_list, posted_on))]))
print(posted_on)
# print(date_list)
print(date_db)
# date_list[0].strftime('%Y-%m-%d %H:%M:%S')
# print(result)
for i in range(len(url)):
item, created = NewsItem.objects.get_or_create(url=url[i])
if created:
print('New item was created')
item.title = title[i]
item.hacker_news_url = hacker_news_url[i]
item.posted_on = date_db[i]
item.comment_count = comment_count[i]
item.upvote_count = upvote_count[i]
else:
print('updating current item')
item.title = title[i]
item.hacker_news_url = hacker_news_url[i]
item.posted_on = date_db[i]
item.comment_count = comment_count[i]
item.upvote_count = upvote_count[i]
item.save()
# obj=NewsItem(url=url[i],title=title[i],hacker_news_url=hacker_news_url[i],posted_on=posted_on[i],comment_count=comment_count[i],upvote_count=upvote_count[0])
# obj.save()
print(url)
print(title)
print(hacker_news_url)
print(upvote_count)
print(comment_count)
# print(posted_on)
print(date_db)
def parse_date(dtstring):
if dtstring.strip():
time_struct, parse_status = dtparser.Calendar().parse(dtstring)
return datetime.fromtimestamp(mktime(time_struct))
else:
return ""
def validate_content(file, content_dir):
"""
Validate content for a file based on the rules listed in __doc__
:file: (str) the name of the content file to validate
:content_dir: (str) the path to the content directory
"""
with open(file, 'r') as f:
file_contents = [
content.replace('\n', '') for content in f.readlines()
]
contents = {}
for line in file_contents:
items = line.split(':')
field, value = cleanse(items[0]), cleanse(':'.join(items[1:]))
contents[field] = value
status_code = 0
filename = file
field = ""
short_message = ""
long_message = "n/a"
loop = True
while (loop):
loop = False
# validate filetype
field = ''
filetype = os.path.splitext(file)[-1]
if not filetype == '.md':
status_code = 1
short_message = "invalid filetype"
long_message = f"File type `{filetype}` is not supported. Please ensure that your contribution is written in a Markdown file (`.md`)."
break
# validate name
field = 'name'
if not 'name' in contents:
status_code = 1
short_message = "missing required field `name`"
break
if contents['name'] == '':
status_code = 1
short_message = "empty required field `name`"
break
if not str(contents['name']):
status_code = 1
short_message = "invalid field: `name`"
break
if not len(contents['name']) <= 100:
status_code = 1
short_message = "length of field `name` exceeds 100 characters"
break
# validate author
field = 'author'
if not 'author' in contents:
status_code = 1
short_message = "missing required field `author`"
break
if contents['author'] == '':
status_code = 1
short_message = "empty required field `author`"
break
if not str(contents['author']):
status_code = 1
short_message = "invalid field `author`"
break
if not len(contents['author']) <= 100:
status_code = 1
short_message = "length of field `author` exceeds 100 characters"
break
# validate author_github
field = 'author_github'
author_github = contents.get('author_github', '')
if not author_github == '':
if not str(author_github):
status_code = 1
short_message = "invalid field `author_github`"
break
if not 'github.com' in author_github:
status_code = 1
short_message = "field `author_github` must be a GitHub URL"
break
try:
response = requests.get(author_github, timeout=15)
if not response:
status_code = 1
short_message = "URL provided for field `author_github` returned an error HTTP code when accessed"
long_message = f"The result of accessing \"{contents['author_github']}\" resulted in an HTTP response code of `{response.status_code}`, which is an error."
break
except Exception as e:
status_code = 1
short_message = "URL provided for field `author_github` is inaccessible"
long_message = f"Trying to access \"{contents['author_github']}\" resulted in an unknown exception, likely indicating that the URL is invalid."
break
# validate blurb
field = 'blurb'
if not 'blurb' in contents:
status_code = 1
short_message = "missing required field `blurb`"
break
if contents['blurb'] == '':
status_code = 1
short_message = "empty required field `blurb`"
break
if not str(contents['blurb']):
status_code = 1
short_message = "invalid field `blurb`"
break
if not len(contents['blurb']) <= 100:
status_code = 1
short_message = "length of field `blurb` exceeds 100 characters"
break
# validate description
field = 'description'
if not 'description' in contents:
status_code = 1
short_message = "missing required field `description`"
break
if contents['description'] == '':
status_code = 1
short_message = "empty required field `description`"
break
if not str(contents['description']):
status_code = 1
short_message = "invalid field `description`"
break
if not len(contents['description']) <= 1000:
status_code = 1
short_message = "length of field `description` exceeds 1000 characters"
break
# validate url
field = 'url'
if not 'url' in contents:
status_code = 1
short_message = "missing required field `url`"
break
if contents['url'] == '':
status_code = 1
short_message = "empty required field `url`"
break
if not str(contents['url']):
status_code = 1
short_message = "invalid field `url`"
break
if 'github.com' in contents['url']:
status_code = 1
short_message = "URL provided for field `url` is a GitHub repository"
long_message = f"It appears the URL for this contribution ({contents['url']}) is a GitHub repository, not a published web app. Unfortunately, Soliloquy is not a portfolio for source code. Please refer to the FAQs for more information: https://www.soliloquy.dev/about/"
break
try:
response = requests.get(contents['url'], timeout=15)
if not response:
status_code = 1
short_message = "URL provided for field `url` returned an error"
long_message = f"The result of accessing \"{contents['url']}\" resulted in an HTTP response code of `{response.status_code}`, which is an error."
break
except Exception as e:
status_code = 1
short_message = "URL provided for field `url` is inaccessible"
long_message = f"Trying to access \"{contents['url']}\" resulted in an unknown exception, likely indicating that the URL is invalid."
break
# validate img
field = 'img'
img = contents.get('img', '')
if not img == '':
reserved_file_names = ['about', 'default', 'willcarhartportfolio']
supported_filetypes = ['.png', '.jpg', '.jpeg', '.gif']
image_filename, filetype = os.path.splitext(img)
if not filetype in supported_filetypes:
status_code = 1
short_message = f"Value of field `img` file type not supported (`{filetype}`)"
long_message = "Supported image file types are `.png`, `.jpg`, `.jpeg`, and `.gif`."
break
if image_filename in reserved_file_names:
status_code = 1
short_message = f"filename provided for field `img` is prohibited (`{image_filename}`)"
long_message = "Some filenames are reserved for the system, as they are used elsewhere in Soliloquy's assets and thus are prohibited for use in names of contribution images. These filenames are `about`, `default`, and `willcarhartportfolio`, extension agnostic."
break
if not os.path.isfile(f'{content_dir}/app_img/{img}'):
status_code = 1
short_message = f"filename provided for field `img` not found, no such file `{os.path.basename(content_dir)}/app_img/{img}`"
long_message = f"Make sure to add your image file to `{os.path.basename(content_dir)}/app_img/`, as this is where Soliloquy will look for it."
break
# validate timestamp
field = 'timestamp'
if not 'timestamp' in contents:
status_code = 1
short_message = "missing required field `timestamp`"
break
if contents['timestamp'] == '':
status_code = 1
short_message = "empty required field `timestamp`"
break
if any(
re.match(regex, contents['timestamp'])
for regex in ['../../.*', '.*/../..', '..-..-.*', '.*-..-..']):
status_code = 1
short_message = "value provided for field `timestamp` is ambiguous"
long_message = f"The timestamp you provided (`{contents['timestamp']}`) is ambiguous. This means that its value is not deterministic. For example, **\"05/06/2018\"** could be interpreted as **May 6th, 2018** or **June 5th, 2018**."
break
try:
parse = parsedatetime.Calendar()
time_struct, parse_status = parse.parse(value)
if not parse_status == 1:
status_code = 1
short_message = f"invalid field `timestamp`"
long_message = f"The provided timestamp `{contents['timestamp']}` could not be parsed. Try using the format **Month Day, Year**, like August 7th, 2019."
break
dt = datetime.datetime(*time_struct[:6])
timestamp = dt.timestamp()
except ValueError:
status_code = 1
short_message = "invalid field `timestamp`"
long_message = f"The provided timestamp `{contents['timestamp']}` could not be parsed. Try using the format **Month Day, Year**, like August 7th, 2019."
break
return ContentError(status_code=status_code,
filename=os.path.basename(filename),
field=field,
short_message=short_message,
long_message=long_message)
def onLoad(self):
self.cal = parsedatetime.Calendar()
def __init__(self, name, emitter=None):
super(ScheduledSkill, self).__init__(name, emitter)
self.timer = None
self.calendar = pdt.Calendar()
self.time_rules = time_rules.create(self.lang)
self.init_format()
#!/usr/bin/env python
import time, calendar, parsedatetime as pdt, pytz, re
from datetime import datetime
from time import mktime
from pytz import timezone
from dateutil import tz
c = pdt.Constants()
c.BirthdayEpoch = 80
p = pdt.Calendar(c)
dateFormats = [
"%d/%m/%y-%H:%M:%S", "%d/%m/%y", "%d/%m/%Y", "%d/%m/%y-%H:%M",
"%d/%m/%Y-%H:%M:%S", "%d/%m/%y-%H:%M", "%H:%M:%S"
]
#will try to match a time given in a form of hh:mm:ss from current time
def regexmatch(timestring):
pattern = re.compile("^(?:(?:([0-9])*:)?([0-9]*?\d):)?([0-9]*\d)$")
hours = mins = secs = 0
if pattern.match(timestring):
args = timestring.split(":")
if len(args) == 3:
hours = int(args[0])
mins = int(args[1])
secs = int(args[2])
elif len(args) == 2:
mins = int(args[0])
secs = int(args[1])
else:
def __init__(self, date_format=None, **kwargs):
super(Date, self).__init__(**kwargs)
self.date_format = date_format
self.parser = parsedatetime.Calendar()
def setUp(self):
self.cal = pdt.Calendar()
self.yr, self.mth, self.dy, self.hr, self.mn, self.sec, self.wd, self.yd, self.isdst = time.localtime()
def parse(*args, **kwargs):
cal = parsedatetime.Calendar()
dt, flag = cal.parse(*args, **kwargs)
return dt
def __init__(self):
self.date_constants = parsedatetime.Constants()
self.date_calendar = parsedatetime.Calendar(self.date_constants)
def parse_other_answer_page(self, response):
c = pdt.Constants()
p = pdt.Calendar(c)
f = '%Y-%m-%d %H:%M:%S'
hxs = HtmlXPathSelector(response)
all_answer = hxs.xpath('//ul[contains(@id,"ya-qn-answers")]/li')
current_ans_id = response.meta['ult_ans_id']
for single_answer in all_answer:
item = YahooItem()
ans_data = single_answer.xpath(
'.//div[contains(@class,"Pt-15")]/span[contains(@class, "Clr-88")]'
).extract()
data_string = html2text.html2text(ans_data[0])
data_format = p.parseDT(
str(
data_string.encode("utf8").replace("\xc2\xb7",
"").strip()))
item['date_time'] = data_format[0].strftime(f)
item['uid'] = str(
str(response.meta['quest_id']) + "." + str(current_ans_id))
item['type'] = "answer"
item['tags'] = "N/A"
item['title'] = ""
item['resolve'] = ""
item['answers'] = 0
item['views'] = 0
text_to_gain = single_answer.xpath(
'.//a[contains(@class,"uname Clr-b")]').extract()
if text_to_gain:
h = html2text.HTML2Text()
h.ignore_links = True
author_string = h.handle(text_to_gain[0])
item['author'] = str(
author_string.encode('utf-8', 'ignore').strip())
else:
item['author'] = "anonymous"
item['url'] = response.url
text_to_gain = single_answer.xpath(
'.//span[contains(@class,"ya-q-full-text")][@itemprop="text"]'
).extract()
if text_to_gain:
item['text'] = html2text.html2text(text_to_gain[0]).encode(
'utf-8', 'ignore')
else:
item['text'] = ""
text_to_gain = single_answer.xpath(
'.//div[contains(@class,"D-ib Mend-10 Clr-93")]/div[1]/div[1]'
).extract()
if text_to_gain:
item['upvotes'] = int(html2text.html2text(text_to_gain[0]))
else:
item['upvotes'] = 0
current_ans_id = current_ans_id + 1
yield item
try:
if (hxs.xpath('//div[contains(@id, "ya-qn-pagination")]' +
'/a[contains(@class,"Clr-bl")][last()]/@href')):
url_of_the_next_page = hxs.xpath(
'//div[contains(@id, "ya-qn-pagination")]' +
'/a[contains(@class,"Clr-bl")][last()]/@href').extract()
next_page_composed = "https://hk.answers.yahoo.com" + \
url_of_the_next_page[0]
request = scrapy.Request(next_page_composed,
callback=self.parse_other_answer_page)
request.meta['quest_id'] = response.meta['quest_id']
request.meta['ult_ans_id'] = current_ans_id
yield request
except NoSuchElementException:
pass
def process_url(self, url):
"""
For the given url find or create an entry for each source in the database.
If no sources found raise an exception.
"""
h = httplib2.Http( settings.HTTPLIB2_CACHE_DIR )
response, content = h.request(url)
# print content
# parse content
soup = BeautifulSoup(
content,
convertEntities=BeautifulStoneSoup.HTML_ENTITIES
)
spans = soup.findAll( 'span', 'contenttype-repositoryitem summary')
links = [ span.a for span in spans ]
# Check that we found some links. This is to detect when the page changes or our
# scraper breaks (see issue #905 for example). Checking that the most recent
# source is not more that X weeks old might also be a good idea, but could lead
# to lots of false positives as there is often a long hiatus.
if not len(links):
raise NoSourcesFoundError()
for link in links:
# print '==============='
# print link
href = link['href'].strip()
# print "href: " + href
name = ' '.join(link.contents).strip()
# print "name: " + name
if not Source.objects.filter(name=name).exists():
cal = pdt.Calendar()
# Sometimes the space is missing between before the
# month, so insert that if it appears to be missing:
tidied_name = re.sub(r'(\d+(st|nd|rd|th))(?=[^ ])', '\\1 ', name)
# Commas in the name confuse parsedatetime, so strip
# them out too:
tidied_name = re.sub(r',', '', tidied_name)
result = cal.parseDateText(tidied_name)
source_date = datetime.date(*result[:3])
# print "source_date: " + str(source_date)
# I don't trust that we can accurately create the download link url with the
# details that we have. Instead fetche the page and extract the url.
download_response, download_content = h.request(href)
download_soup = BeautifulSoup(
download_content,
convertEntities=BeautifulStoneSoup.HTML_ENTITIES
)
download_div = download_soup.find( id="archetypes-fieldname-item_files" )
if not download_div:
warn("Failed to find the download div on {0}".format(href))
continue
download_url = download_div.a['href']
# print download_url
# create the source entry
Source.objects.create(
name = name,
url = download_url,
date = source_date,
)
def parse_page(self, response):
# Time tools
c = pdt.Constants()
p = pdt.Calendar(c)
f = '%Y-%m-%d %H:%M:%S'
now = datetime.datetime.now()
# Start to scraping a single question
# Checking question category
try:
hxs = HtmlXPathSelector(response)
category = hxs.xpath(
'(//a[contains(@class,"Clr-b")])[2]').extract()
h = html2text.HTML2Text()
h.ignore_links = True
category_text = h.handle(category[0])
url_category = str(category_text).encode('utf8').strip()
except IndexError:
print(str(self.uid) + "Warning: this Url is not more available...")
url_category = "Error"
# If the question is related to programming and design
# start item creation process
# if "程式編寫" and "設計" in url_category:
if (True):
# increment id
# copy id and use uid_copy in order to preserve from concurrent request
self.uid = self.uid + 1
uid_copy = self.uid
# Print current uid any 100 times
if self.uid % 100 == 0:
print(str(self.uid))
# Initialize scrapy item
item = YahooItem()
# Read in the date field associated to URL if info data are present
for istance in self.url_to_scrape:
if response.url == istance.url:
if istance.date == "not available":
item['date_time'] = "not available"
break
else:
data_format = p.parseDT(
str(
str(istance.date).replace("\xc2\xb7",
"").strip()))
item['date_time'] = data_format[0].strftime(f)
break
item['type'] = "question"
item['uid'] = uid_copy
item['url'] = response.url
item['tags'] = "N/A"
item['views'] = 0
item['upvotes'] = 0
text_to_gain = hxs.xpath('//h1').extract()
# Take title of the question
item['title'] = (html2text.html2text(
text_to_gain[0]).encode("utf8").strip())
# Take text from the question
full_text_answer = hxs.xpath(
'//span[contains(@class,"ya-q-full-text Ol-n")]').extract()
if full_text_answer:
item['text'] = html2text.html2text(full_text_answer[0]).encode(
'utf-8', 'ignore')
else:
text_to_gain = hxs.xpath(
'//span[contains(@class,"ya-q-text")]').extract()
if text_to_gain:
item['text'] = html2text.html2text(text_to_gain[0]).encode(
'utf-8', 'ignore')
# Take username of the questioner
text_to_gain = hxs.xpath(
'//div[contains(@id,"yq-question-detail-profile-img")]' +
'/a/img/@alt').extract()
if text_to_gain:
try:
h = html2text.HTML2Text()
h.ignore_links = True
author_string = h.handle(text_to_gain[0])
item['author'] = author_string.encode('utf-8',
'ignore').strip()
# Handle HTMLtoText except
except:
item['author'] = "anonymous"
else:
item['author'] = "anonymous"
text_to_gain = hxs.xpath(
'(//div[contains(@class,"Mend-10 Fz-13 Fw-n D-ib")])' +
'[2]/span[2]').extract()
# Read number of answers
if text_to_gain:
if " answers" in (str(html2text.html2text(
text_to_gain[0])).strip()):
item['answers'] = int(
str(html2text.html2text(text_to_gain[0])).replace(
" answers", "").strip())
else:
if " answer" in (str(html2text.html2text(
text_to_gain[0])).strip()):
item['answers'] = int(
str(html2text.html2text(text_to_gain[0])).replace(
" answer", "").strip())
else:
item['answers'] = 0
# Check if question is closed (resolve with a best answer)
text_to_gain = hxs.xpath(
'//span[contains(@class,"ya-ba-title Fw-b")]/text()').extract(
)
if text_to_gain:
item['resolve'] = "True"
else:
item['resolve'] = "False"
# yield item for the question istance
yield item
# Taking the best answer if present
if hxs.xpath('//div[contains(@id,"ya-best-answer")]'):
ans_uid = 1
item = YahooItem()
ans_data = hxs.xpath(
'(//div[contains(@class,"Pt-15")]/' +
'span[contains(@class, "Clr-88")])[1]').extract()
data_string = html2text.html2text(ans_data[0]).strip()
data_format = p.parseDT(
str(
data_string.encode("utf8").replace("\xc2\xb7",
"").strip()))
item['date_time'] = data_format[0].strftime(f)
item['uid'] = str(str(uid_copy) + ("." + str(ans_uid)))
item['type'] = "answer"
item['resolve'] = "solution"
item['tags'] = "N/A"
item['title'] = ""
item['answers'] = 0
item['views'] = 0
best_text = hxs.xpath(
'(//span[contains(@class,"ya-q-full-text")])[1]').extract(
)
item['text'] = html2text.html2text(best_text[0]).encode(
'utf-8', 'ignore')
text_to_gain = hxs.xpath(
'(//a[contains(@class,"uname Clr-b")])[1]').extract()
if text_to_gain:
h = html2text.HTML2Text()
h.ignore_links = True
author_string = h.handle(text_to_gain[0])
item['author'] = str(
author_string.encode('utf-8', 'ignore').strip())
else:
item['author'] = "anonymous"
upvote_text = hxs.xpath(
'(//div[contains(@class,"D-ib Mstart-23 count")])[1]/text()'
).extract()
item['upvotes'] = int(
str(html2text.html2text(upvote_text[0])).strip())
item['url'] = response.url
ans_uid = ans_uid + 1
yield item
else:
ans_uid = 1
# Taking all the other answers
all_answer = hxs.xpath('//ul[contains(@id,"ya-qn-answers")]/li')
for single_answer in all_answer:
item = YahooItem()
# In this case data is always present
ans_data = single_answer.xpath(
'.//div[contains(@class,"Pt-15")]/span[contains(@class, "Clr-88")]'
).extract()
data_string = html2text.html2text(ans_data[0])
data_format = p.parseDT(
str(
data_string.encode("utf8").replace("\xc2\xb7",
"").strip()))
item['date_time'] = data_format[0].strftime(f)
item['uid'] = str(str(uid_copy) + ("." + str(ans_uid)))
item['tags'] = "N/A"
item['title'] = ""
item['answers'] = 0
item['views'] = 0
item['type'] = "answer"
item['resolve'] = ""
text_to_gain = single_answer.xpath(
'.//a[contains(@class,"uname Clr-b")]').extract()
if text_to_gain:
h = html2text.HTML2Text()
h.ignore_links = True
author_string = h.handle(text_to_gain[0])
item['author'] = str(
author_string.encode('utf-8', 'ignore'))
else:
item['author'] = "anonymous"
# Take url of the question becouse answer don't have URL ref
item['url'] = response.url
# Check if is present long text version of the answer
text_to_gain = single_answer.xpath(
'.//span[contains(@class,"ya-q-full-text")][@itemprop="text"]'
).extract()
if text_to_gain:
item['text'] = html2text.html2text(text_to_gain[0]).encode(
'utf-8', 'ignore')
else:
item['text'] = ""
text_to_gain = single_answer.xpath(
'.//div[contains(@class,"D-ib Mend-10 Clr-93")]/div[1]/div[1]'
).extract()
if text_to_gain:
item['upvotes'] = int(
str(html2text.html2text(text_to_gain[0])).strip())
else:
item['upvotes'] = 0
ans_uid = ans_uid + 1
yield item
# Checking if there are more then 10 answers
# in this case there are other answers in other page
try:
if (hxs.xpath('//div[contains(@id, "ya-qn-pagination")]' +
'/a[contains(@class,"Clr-bl")][last()]/@href')):
url_of_the_next_page = hxs.xpath(
'//div[contains(@id, "ya-qn-pagination")]' +
'/a[contains(@class,"Clr-bl")][last()]/@href').extract(
)
next_page_composed = "https://hk.answers.yahoo.com" + \
url_of_the_next_page[0]
# Go to the next page and take more urls
# passing uid as parameter
request = scrapy.Request(
next_page_composed,
meta={'ans_id': uid_copy},
callback=self.parse_other_answer_page)
request.meta['quest_id'] = uid_copy
request.meta['ult_ans_id'] = ans_uid
yield request
except NoSuchElementException:
pass
else:
print(str(self.uid) + " question not available or not related")
print(str(response.url))
def setUp(self):
self.ptc = pdt.Constants('en_AU', usePyICU=False)
self.cal = pdt.Calendar(self.ptc)
self.yr, self.mth, self.dy, self.hr, self.mn, self.sec, self.wd, self.yd, self.isdst = time.localtime(
)
def parse_time(time_str):
cal = parsedatetime.Calendar()
time_struct, _ = cal.parse(time_str)
return datetime(*time_struct[:6])
