def fetch_production(country_code='CR', session=None):
# Do not use existing session as some amount of cache is taking place
r = requests.session()
url = 'https://appcenter.grupoice.com/CenceWeb/CencePosdespachoNacional.jsf'
response = r.get(url)
df_yesterday = pd.read_html(response.text, skiprows=1, index_col=0, header=0)[0]
soup = BeautifulSoup(response.text, 'html.parser')
yesterday_date = soup.select('#formPosdespacho:pickFechaInputDate')[0]['value']
jsf_view_state = soup.select('#javax.faces.ViewState')[0]['value']
yesterday = arrow.get(yesterday_date, 'DD/MM/YYYY', tzinfo=TIMEZONE)
today = yesterday.shift(days=+1)
data = [
('formPosdespacho', 'formPosdespacho'),
('formPosdespacho:pickFechaInputDate', today.format(DATE_FORMAT)),
('formPosdespacho:pickFechaInputCurrentDate', today.format(MONTH_FORMAT)),
('formPosdespacho:j_id35.x', ''),
('formPosdespacho:j_id35.y', ''),
('javax.faces.ViewState', jsf_view_state),
]
response = r.post(url, cookies={}, data=data)
df_today = pd.read_html(response.text, skiprows=1, index_col=0)[0]
ydata = df_to_data(country_code, yesterday, df_yesterday)
tdata = df_to_data(country_code, today, df_today)
production = ydata + tdata
unknown_plants()
return production
def get_stats(year, level='pro'): #TODO Switch to regex patterns
'''Scrapes draftexpress.com/stats for of a given level, year'''
front = 'http://www.draftexpress.com/stats.php?sort=8&q='
pages = 2
frontb = '&league=NBA&year=20'
if level == 'col':
frontb = '&league=NCAA&year=20'
pages = 13
midA = '&per=per40pace&qual=prospects&sort2=DESC&pos=all&stage=all&min=10&conference=All&pageno='
back = '&sort=8'
url = front + frontb + year + midA+ '0' + back
reg = pd.DataFrame()
eff = pd.DataFrame()
for n in xrange(pages):
url = front + frontb + year + midA+ str(n) + back
eff_url = front + 'eff'+ frontb + year + midA+ str(n) + back
reg_temps = pd.read_html(url, header=0)
reg_temp = reg_temps[5]
eff_temps = pd.read_html(eff_url)
eff_temp = eff_temps[5]
eff_temp.to_csv('temp.csv')
eff_temp = pd.read_csv('temp.csv', header=3) #im ashamed
reg = reg.append(reg_temp)
eff = eff.append(eff_temp)
reg['year'] = 2000 + float(year)
eff['year'] = 2000 + float(year)
df = reg.merge(eff, how='inner', on='Name', suffixes=('', '_y'))
df = df.drop(['Cmp', 'Team_y', 'year_y', 'Min_y', 'Cmp_y', 'GP_y'], 1)
print df.shape
return df
def get_html_dfs(stryear, strmonth):
year = int(stryear)
month = int(strmonth)
monthly_file = "./" + stryear + "_" + strmonth + ".html"
try:
with open (monthly_file, 'r') as mf:
dfs = pd.read_html(monthly_file, encoding='utf-8')
print ("read html file successfully")
return dfs
except Exception as e:
print(e)
if year > 1990:
year -= 1911
url = 'http://mops.twse.com.tw/nas/t21/sii/t21sc03_'+str(year)+'_'+str(month)+'_0.html'
if year <= 98:
url = 'http://mops.twse.com.tw/nas/t21/sii/t21sc03_'+str(year)+'_'+str(month)+'.html'
headers = {'User-Agent': 'Mozilla/5.0 (Macintosh; Intel Mac OS X 10_10_1) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/39.0.2171.95 Safari/537.36'}
r = requests.get(url, headers=headers)
r.encoding = 'big5'
print ("fetch html file successfully")
with codecs.open( monthly_file, mode='wb') as writefile:
writefile.write(r.text.encode('utf8'))
dfs = pd.read_html(StringIO(r.text), encoding='big-5')
return dfs
def get_box_stats(self, url):
"""
INPUT: NCAAScraper, STRING
OUTPUT: DATAFRAME, DATAFRAME, DATAFRAME
Extract html from box stats page and convert to dataframes
url is a string linking to the box stats page
"""
soup = self.page_opener.open_and_soup(url)
tables = soup.findAll('table', {'class': 'mytable'})
if len(tables) != 3:
print 'Incorrect number of tables'
return None
htable = pd.read_html(str(tables[0]), header=0)[0]
table1 = pd.read_html(str(tables[1]), skiprows=1, header=0, infer_types=False)[0]
table2 = pd.read_html(str(tables[2]), skiprows=1, header=0, infer_types=False)[0]
team1 = htable.iloc[0, 0]
team2 = htable.iloc[1, 0]
table1['Team'] = [team1] * table1.shape[0]
table2['Team'] = [team2] * table2.shape[0]
table1['game_id'] = [self.game_id(url)] * table1.shape[0]
table2['game_id'] = [self.game_id(url)] * table2.shape[0]
# older box stat page versions use different column names so
# we must map them all to common column names (e.g. MIN vs. Min)
table1 = self.rename_box_table(table1)
table2 = self.rename_box_table(table2)
table1 = self.format_box_table(table1)
table2 = self.format_box_table(table2)
return htable, table1, table2
def make_USrepresentative_df():
representative_df = pd.DataFrame()
df = pd.read_html(URLS['dem_USrepresentative'])[0]
df.columns = ['county', 'candidate1', 'candidate2',
'candidate3', 'candidate4', 'candidate5', 'candidate6']
df['county'] = df['county'].fillna('')
splits = df[df.county.str.startswith('DISTRICT')].index.tolist()
splits.append(df.shape[0])
for split in range(len(splits) - 1):
df_ = df.iloc[splits[split]:splits[split+1]]
df_ = df_.drop(df_.index[0])
df_.columns = df_.iloc[0]
df_ = df_.drop(df_.index[0])
df_.columns = ['county'] + list(df_.columns[1:])
df_ = df_.dropna(subset=[df_.columns.values[1]])
df_ = df_.dropna(axis=1)
df_ = pd.melt(df_, id_vars=['county'], value_vars=list(df_.columns[1:]))
df_.columns = ['county', 'candidate', 'votes']
df_ = df_[df_['county'] != '']
df_['party'] = 'Democratic'
df_['candidate'] = df_['candidate'].str.lstrip('*')
df_['candidate'] = df_['candidate'].str.replace('\((.*?)\)', '')
df_['candidate'] = df_['candidate'].str.rstrip('()')
df_['office'] = 'US Representative'
representative_df = representative_df.append(df_)
df = pd.read_html(URLS['rep_USrepresentative'])[0]
df.columns = ['county', 'candidate1', 'candidate2',
'candidate3', 'candidate4', 'candidate5']
df['county'] = df['county'].fillna('')
splits = df[df.county.str.startswith('DISTRICT')].index.tolist()
splits.append(df.shape[0])
for split in range(len(splits) - 1):
df_ = df.iloc[splits[split]:splits[split+1]]
df_ = df_.drop(df_.index[0])
df_.columns = df_.iloc[0]
df_ = df_.drop(df_.index[0])
df_.columns = ['county'] + list(df_.columns[1:])
df_ = df_.dropna(subset=[df_.columns.values[1]])
df_ = df_.dropna(axis=1)
df_ = pd.melt(df_, id_vars=['county'], value_vars=list(df_.columns[1:]))
df_.columns = ['county', 'candidate', 'votes']
df_ = df_[df_['county'] != '']
df_['party'] = 'Republican'
df_['candidate'] = df_['candidate'].str.lstrip('*')
df_['candidate'] = df_['candidate'].str.replace('\((.*?)\)', '')
df_['candidate'] = df_['candidate'].str.rstrip('()')
df_['office'] = 'US Representative'
representative_df = representative_df.append(df_)
return representative_df
def read_ema_ings():
with open('all_ingredients.html', 'rb') as f_in:
text_ings = f_in.read()
df_sort = pd.read_html(text_ings)[0]
df_sort.columns = ['name', 'sort_name', 'form']
df_sort = df_sort.sort_values('name')
df_sort.reset_index(drop=True, inplace=True)
cols_ings = ['ingredient', 'function', 'form', 'id_test', 'assay']
df_type = ['high_suscept', 'low_suscept', 'high_suscept_id', 'pending']
with open('ema_ingredients.html', 'rb') as f_in:
text_ings = f_in.read()
dfs = pd.read_html(text_ings)
for i, df_ in enumerate(dfs):
df_.columns = cols_ings
df_['type'] = df_type[i]
all_dfs = pd.concat(dfs)
all_dfs = all_dfs.sort_values('ingredient')
all_dfs = all_dfs.drop_duplicates('ingredient')
all_dfs.reset_index(drop=True, inplace=True)
all_dfs['sort_name'] = df_sort['sort_name']
all_dfs = all_dfs.sort_values('sort_name')
all_dfs.reset_index(drop=True, inplace=True)
return all_dfs
def pulldata():
# Schedule, odds and TV listing DataFrame
da = pd.read_html('https://www.teamrankings.com/ncaa-basketball/team/florida-gators')[1]
db = pd.read_html('http://stats.gatorsports.com/cbk/teamstats.asp?team=210&report=schedule',
header=0, parse_dates=False, attrs={'class': 'shsTable shsBorderTable'})[0]['TV']
df = pd.concat([da, db], axis=1)
df = df.set_index('Date')
return df
def test_processed_table(self):
pd.testing.assert_series_equal(
pd.read_html(test_two_d.get_result().table)[0]["X-Coordinate"],
two_d_result.table["X-Coordinate"]
)
pd.testing.assert_series_equal(
pd.read_html(test_two_d.get_result().table)[0]["Y-Coordinate"],
two_d_result.table["Y-Coordinate"]
)
def get_fangraph_pitchers():
# get al pitchers
al = pd.read_html('http://www.fangraphs.com/dailyprojections.aspx?pos=all&stats=pit&type=sabersim&team=0&lg=al&players=0')
fgpal = al[15]
sleep(2)
# get nl pitchers
nl = pd.read_html('http://www.fangraphs.com/dailyprojections.aspx?pos=all&stats=pit&type=sabersim&team=0&lg=nl&players=0')
fgpnl = nl[15]
# merge and return
fgp = fgpal.append(fgpnl)
return(fgp)
def get_cruz():
cruzh = pd.read_html('https://rotogrinders.com/pages/c-r-u-z-mlb-model-792518')
czh = cruzh[1]
cruzp = pd.read_html('https://rotogrinders.com/pages/c-r-u-z-mlb-model-792521')
czp = cruzp[1]
czh['cruz'] = czh['\tRating\t']
czp['cruz'] = czp['\tRating\t']
return(czh, czp)
def get_household_income_from_orlando_sentinel():
name = 'city_median_income'
url = 'http://databases.sun-sentinel.com/Orlando/ftlaudOS2011income/income2011_list.php'
df = pd.read_html(url)[6][:-1]
for i in range(2,28):
print i
df = df.append(pd.read_html(url + '?goto=%d'%i)[6][:-1])
df.columns = ['0', 'City', 'State', 'Median_Income', '4']
df = df[['City', 'State', 'Median_Income']]
df.Median_Income = df.Median_Income.str.strip('$').apply(locale.atof)
return df
def get_fangraph_batters():
poslist = ['c','1b','2b','ss','3b','rf','cf','lf','dh']
df = pd.DataFrame()
for pos in poslist:
tmp = pd.read_html('http://www.fangraphs.com/dailyprojections.aspx?pos='+pos+'&stats=bat&type=sabersim&team=0&lg=al&players=0')
df = df.append(tmp[15])
sleep(2)
tmp2 = pd.read_html('http://www.fangraphs.com/dailyprojections.aspx?pos='+pos+'&stats=bat&type=sabersim&team=0&lg=nl&players=0')
df = df.append(tmp2[15])
sleep(2)
return(df)
def main():
"""Main execution."""
# Determine command line arguments.
try:
rawopts, _ = getopt.getopt(sys.argv[1:], 'i:o:')
except getopt.GetoptError:
usage()
sys.exit(2)
opts = {}
# Process each command line argument.
for o, a in rawopts:
opts[o[1]] = a
# The following arguments are required in all cases.
for opt in ['i', 'o']:
if not opt in opts:
usage()
sys.exit(2)
# Make sure the output directory exists.
if not os.path.exists(opts['o']):
os.makedirs(opts['o'])
# Traverse the root folder that contains sub folders
# that represent each pitcher.
for root, dirs, _ in os.walk(opts['i']):
# Traverse each folder in the root.
for pid in dirs:
outfile = os.path.join(opts['o'], pid + ".csv")
# Check if this pitcher was already processed.
if os.path.isfile(outfile):
continue
for proot, _, files in os.walk(os.path.join(root, pid)):
try:
# Read in the first game for this pitcher.
with open(os.path.join(proot, files[0]), 'r') as f:
df = pd.read_html(f.read(), header=0)[0]
# Read in the subsequent games and append to the
# running DataFrame.
for file in files[1:]:
with open(os.path.join(proot, file), 'r') as f:
df = df.append(pd.read_html(f.read(), header=0)[0])
# Save to disk as a csv file.
df.to_csv(outfile)
except ValueError:
print("Error processing " + pid)
continue
def predict_by_stats(games=[]):
scores = get_team_scores(team_scores_url)
num_scores = len(scores)
team_stats = pd.read_html(team_misc_stats_url, header=1)[0].iloc[:-1, :]
team_stats['Team'] = [t.strip('*') for t in team_stats['Team'].values]
scores['home-away'] = scores['PTS.1'] - scores['PTS'] - home_court_advantage # home court adv = 2 pts
param_columns = team_stats.columns[13:21].tolist() # starts from column eFG%
param_columns.remove('FT/FGA')
param_columns.remove('FT/FGA.1')
num_params = len(param_columns)
x = np.zeros([num_scores, num_params])
for idx, row in scores.iterrows():
home = row['Home/Neutral']
away = row['Visitor/Neutral']
x[idx] = team_stats.loc[team_stats['Team'] == home][param_columns].values - \
team_stats.loc[team_stats['Team'] == away][param_columns].values
x = pd.DataFrame(x, columns=param_columns)
y = scores['home-away']
model = sm.OLS(y, x)
result = model.fit()
print(result.summary())
print()
team_ranking = pd.read_html(team_ranking_url, header=1)[0]
game_spreads = {} #get_game_spreads()
print('{:22s} - {:22s} = {:7s} | {:7s} | {:6s} | {:6s} | {:6s}'.format('home', 'away', 'fit mov', 'ref mov',
'spread', 'vs fit', 'vs mov'))
for [home, away] in games:
fit_mov = sum(result.params * (
team_stats.loc[team_stats['Team'] == home][param_columns].values -
team_stats.loc[team_stats['Team'] == away][
param_columns].values)[0]) + home_court_advantage
mov = team_ranking.loc[team_stats['Team'] == home]['MOV/A'].values - \
team_ranking.loc[team_stats['Team'] == away]['MOV/A'].values + 2
home_spread = -999
for k, v in game_spreads.items():
if home.find(k) > -1:
home_spread = v * -1
print('{:22s} - {:22s} = {:7.1f} | {:7.1f} | {:6.1f} | {:>6s} | {:>6s}'.format(home, away, fit_mov, mov[0],
home_spread,
'home' if fit_mov > home_spread else 'away',
'home' if mov > home_spread else 'away'
))
def getpe_google(stocks):
pe_list = dict()
pe_list['ticker'] = []
pe_list['value'] = []
for ticker in stocks:
try:
key_statistics = pd.read_html('https://www.google.com/finance?q=' + str(ticker) + '&ei')
except:
key_statistics = pd.read_html('https://www.google.com/finance?q=NYSEARCA%3A' + str(ticker) + '&ei')
convert = key_statistics[0][1][5:6]
input = convert.tolist()
pe_list['ticker'].append(ticker)
pe_list['value'].extend(input)
return(pe_list)
def scrape_mock(year):
'''Scrapes a mock_draft off the web in a wierd format'''
url = MOCK_URL + str(year) + '/list/'
crap = pd.read_html(url, header=0, match='First Round')
first_round = crap[-1]
crap = pd.read_html(url, header=0, match='Second Round')
second_round = crap[-1]
first_round.columns = ['pick', 'year', 'details']
second_round.columns = ['pick', 'year', 'details']
second_round['pick'] = second_round['pick'] + 30
mock_draft = first_round.append(second_round)
mock_draft['year'] = year
mock_draft = mock_draft.set_index('pick')
mock_draft['pick'] = mock_draft.index
return mock_draft
def crawler(start, difference, maximum):
try:
result = pd.DataFrame()
origin = "http://fraunhofer-repdose.de/repdose/"
parameter1 = start
parameter2 = parameter1 + difference
if parameter2 > maximum:
parameter2 = maximum
with tqdm(total=math.ceil((maximum-start)/difference)) as pbar:
while parameter1 < maximum:
target = 'http://fraunhofer-repdose.de/repdose/query.php?cas_where=&cas_string=&cas_show=on&species=' \
'&species_show=on&organ=&organ_show=on&name=&name_show=on&s_sex=&ssex_show=on&effect=&effect_show=' \
'on&route=&route_show=on&e_sex=&esex_show=on&boilingpoint_c=&boilingpoint_show=on&duration_from=' \
'&duration_to=&duration_show=on&eloel_mg_from=&eloel_mg_to=&eloel_mg_show=on&watersolubility_c=&watersolubility_show' \
'=on&noel_mg_from=&noel_mg_to=&noel_mg_show=on&logpow_c=&logpow_show=on&loel_mg_from=&loel_mg_to=&loel_mg_show' \
'=on&pressure_c=&pressure_show=on&reliabilityA=on&reliabilityB=on&mol_from='+str(parameter1)+'&mol_to='+str(parameter2)+'&molweight_show=on&reference_show=0'
page = requests.get(target).text
if "Please restrict query conditions." in page:
print(str(parameter1)+":error")
elif "Page" in page:
lists = []
bsObj = BeautifulSoup(page, 'lxml')
found_a = bsObj.find_all('a')
for item in found_a:
found_href = item.get('href')
if "query.php" in found_href:
lists.append(found_href)
for i in lists:
html = origin + i
r_page = requests.get(html).text
table = pd.read_html(r_page)[0]
table.drop([0,1], inplace=True)
result = pd.concat([result,table])
else:
table = pd.read_html(page)[0]
table.drop([0,1], inplace=True)
result = pd.concat([result,table])
parameter1 = parameter2
parameter2 += difference
if parameter2 > maximum:
parameter2 = maximum
time.sleep(0.5)
pbar.update(1)
finally:
get_c_name = pd.read_html(page)[0]
c_name = get_c_name.iloc[1,:]
result.rename(columns=c_name, inplace=True)
result.to_csv("result_" + str(maximum) + ".csv", index=False)
def dfmaker(pagetitle):
# Get page HTML and find tables
wikipage = wikipedia.page(title=pagetitle)
alltables = BeautifulSoup(wikipage.html(), 'html.parser').find_all('table')
# Keep tables that have show results
rtabs = [pd.read_html(str(t), header=0, encoding='utf-8')[0]
for t in alltables if t.th.text == ' No.\n']
yeares = pd.concat(rtabs)
# Clean up dataframe
newnames = {'Original air date[3]': 'airdate',
'Original air date': 'airdate',
'Runner-up': 'Runnerup',
'Last place': 'Lastplace',
'No.': 'shownum'}
yeares.rename(columns=newnames,
inplace=True)
qrem = ['Winner', 'Runnerup', 'Lastplace']
yeares[qrem] = yeares[qrem].replace(regex='["]', value='')
yeares.airdate = yeares.airdate.str[:-13]
results = yeares[pd.notnull(yeares['Winner'])] # removes not results rows
return results
def _today_ticks(symbol, tdate, pageNo, retry_count, pause):
ct._write_console()
for _ in range(retry_count):
time.sleep(pause)
try:
html = lxml.html.parse(ct.TODAY_TICKS_URL % (ct.P_TYPE['http'],
ct.DOMAINS['vsf'], ct.PAGES['t_ticks'],
symbol, tdate, pageNo
))
res = html.xpath('//table[@id=\"datatbl\"]/tbody/tr')
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
sarr = '<table>%s</table>'%sarr
sarr = sarr.replace('--', '0')
df = pd.read_html(StringIO(sarr), parse_dates=False)[0]
df.columns = ct.TODAY_TICK_COLUMNS
df['pchange'] = df['pchange'].map(lambda x : x.replace('%', ''))
except Exception as e:
print(e)
else:
return df
raise IOError(ct.NETWORK_URL_ERROR_MSG)
def _get_report_data(year, quarter, pageNo, dataArr, orderby):
ct._write_console()
try:
request = Request(ct.REPORT_URL % (ct.P_TYPE['http'], ct.DOMAINS['vsf'], ct.PAGES['fd'],
year, quarter, pageNo, ct.PAGE_NUM[1], orderby))
# 默认排序抓取的信息有重复和遗漏,增加排序功能参数orderby
text = urlopen(request, timeout=10).read()
text = text.decode('GBK')
text = text.replace('--', '')
html = lxml.html.parse(StringIO(text))
res = html.xpath("//table[@class=\"list_table\"]/tr")
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
sarr = '<table>%s</table>' % sarr
df = pd.read_html(sarr)[0]
df = df.drop(11, axis=1)
df.columns = ct.REPORT_COLS
dataArr = dataArr.append(df, ignore_index=True)
nextPage = html.xpath('//div[@class=\"pages\"]/a[last()]/@onclick')
if len(nextPage) > 0:
pageNo = re.findall(r'\d+', nextPage[0])[0]
return _get_report_data(year, quarter, pageNo, dataArr,orderby)
else:
return dataArr
except Exception as e:
print(e)
def _parse_fq_data(url, index, retry_count, pause):
for _ in range(retry_count):
time.sleep(pause)
try:
request = Request(url)
text = urlopen(request, timeout=10).read()
text = text.decode('GBK')
html = lxml.html.parse(StringIO(text))
res = html.xpath('//table[@id=\"FundHoldSharesTable\"]')
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
df = pd.read_html(sarr, skiprows = [0, 1])[0]
if len(df) == 0:
return pd.DataFrame()
if index:
df.columns = ct.HIST_FQ_COLS[0:7]
else:
df.columns = ct.HIST_FQ_COLS
if df['date'].dtypes == np.object:
df['date'] = df['date'].astype(np.datetime64)
df = df.drop_duplicates('date')
except Exception as e:
print(e)
else:
return df
raise IOError(ct.NETWORK_URL_ERROR_MSG)
def _get_forecast_data(year, quarter, pageNo, dataArr):
ct._write_console()
try:
gparser = etree.HTMLParser(encoding='GBK')
html = lxml.html.parse(ct.FORECAST_URL%(ct.P_TYPE['http'], ct.DOMAINS['vsf'],
ct.PAGES['fd'], year, quarter, pageNo,
ct.PAGE_NUM[1]),
parser=gparser)
res = html.xpath("//table[@class=\"list_table\"]/tr")
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
sarr = sarr.replace('--', '0')
sarr = '<table>%s</table>'%sarr
df = pd.read_html(sarr)[0]
df = df.drop([4, 5, 8], axis=1)
df.columns = ct.FORECAST_COLS
dataArr = dataArr.append(df, ignore_index=True)
nextPage = html.xpath('//div[@class=\"pages\"]/a[last()]/@onclick')
if len(nextPage)>0:
pageNo = re.findall(r'\d+',nextPage[0])[0]
return _get_forecast_data(year, quarter, pageNo, dataArr)
else:
return dataArr
except Exception as e:
print(e)
def _newstocks(data, pageNo, retry_count, pause):
for _ in range(retry_count):
time.sleep(pause)
ct._write_console()
try:
html = lxml.html.parse(rv.NEW_STOCKS_URL%(ct.P_TYPE['http'],ct.DOMAINS['vsf'],
ct.PAGES['newstock'], pageNo))
res = html.xpath('//table[@id=\"NewStockTable\"]/tr')
if len(res) == 0:
return data
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
sarr = sarr.replace('<font color="red">*</font>', '')
sarr = '<table>%s</table>'%sarr
df = pd.read_html(StringIO(sarr), skiprows=[0, 1])[0]
df = df.drop([df.columns[idx] for idx in [12, 13, 14]], axis=1)
df.columns = rv.NEW_STOCKS_COLS
df['code'] = df['code'].map(lambda x : str(x).zfill(6))
df['xcode'] = df['xcode'].map(lambda x : str(x).zfill(6))
res = html.xpath('//table[@class=\"table2\"]/tr[1]/td[1]/a/text()')
tag = '下一页' if ct.PY3 else unicode('下一页', 'utf-8')
hasNext = True if tag in res else False
data = data.append(df, ignore_index=True)
pageNo += 1
if hasNext:
data = _newstocks(data, pageNo, retry_count, pause)
except Exception as ex:
print(ex)
else:
return data
def extract_coinmarketcap(self, coin, coin_col=False):
"""Retrieve basic historical information for a specific cryptocurrency from coinmarketcap.com
Parameters
----------
coin : the name of the cryptocurrency (e.g. 'bitcoin', 'ethereum', 'dentacoin')
coin_col : whether to include the coin name as a column
(default is False i.e. the column is not included)
Returns
-------
pandas Dataframe
"""
try:
output = pd.read_html("https://coinmarketcap.com/currencies/{}/historical-data/?start={}&end={}".format(
coin, self.from_date.replace("-", ""), self.to_date.replace("-", "")))[0]
except Exception as e:
return pd.DataFrame({"error":e}, index=[0])
output = output.assign(Date=pd.to_datetime(output['Date']))
for col in output.columns:
if output[col].dtype == np.dtype('O'):
output.loc[output[col]=="-",col]=0
output[col] = output[col].astype('int64')
output.columns = [col.lower() for col in output.columns]
if coin_col:
output['coin'] = coin
return output
def get_quote_yahoojp(code, start=None, end=None, interval='d'):
base = 'http://info.finance.yahoo.co.jp/history/?code={0}.T&{1}&{2}&tm={3}&p={4}'
start, end = web._sanitize_dates(start, end)
start = 'sy={0}&sm={1}&sd={2}'.format(start.year, start.month, start.day)
end = 'ey={0}&em={1}&ed={2}'.format(end.year, end.month, end.day)
p = 1
results = []
if interval not in ['d', 'w', 'm', 'v']:
raise ValueError("Invalid interval: valid values are 'd', 'w', 'm' and 'v'")
while True:
url = base.format(code, start, end, interval, p)
tables = pd.read_html(url, header=0)
if len(tables) < 2 or len(tables[1]) == 0:
break
results.append(tables[1])
p += 1
result = pd.concat(results, ignore_index=True)
result.columns = ['Date', 'Open', 'High', 'Low', 'Close', 'Volume', 'Adj Close']
if interval == 'm':
result['Date'] = pd.to_datetime(result['Date'], format='%Y年%m月')
else:
result['Date'] = pd.to_datetime(result['Date'], format='%Y年%m月%d日')
result = result.set_index('Date')
result = result.sort_index()
return result
def scrape_nba_results():
''' Scrape recent NBA results'''
url = 'http://www.betexplorer.com/basketball/usa/nba/results/'
df = pd.read_html(get(url).text)[0]
homeTeam = df[0].apply(lambda r: str.split(r, sep='-')[0].strip())
homeTeam[homeTeam == 'Portland Trail Blazers'] = 'Portland Trailblazers'
homePoints = df[1].apply(lambda r: str.split(r, sep=':')[0].strip())
awayTeam = df[0].apply(lambda r: str.split(r, sep='-')[1].strip())
awayTeam[awayTeam == 'Portland Trail Blazers'] = 'Portland Trailblazers'
awayPoints = df[1].apply(lambda r: str.split(r, sep=':')[1].strip())
awayPoints = awayPoints.apply(lambda r: str.split(r, sep='ET')[0].strip())
dates = df[4].apply(lambda r: datetime.strptime(r, '%d.%m.%Y'))
# The dates on this website are GMT so are one day advanced.
dates = dates.apply(lambda r: datetime.strftime(r - timedelta(days=1),
'%d/%m/%Y'))
df['Date'] = dates
df['HomeTeam'] = homeTeam
df['AwayTeam'] = awayTeam
df['HomePoints'] = homePoints
df['AwayPoints'] = awayPoints
df['HomeWin'] = homePoints > awayPoints
df = df.ix[:, 5:11]
teams = lookup_teams()
df['HomeId'] = df.merge(df.merge(teams, left_on='HomeTeam',
right_on='Franchise', sort=False))['TeamId']
df['AwayId']= df.merge(df.merge(teams, left_on='AwayTeam',
right_on='Franchise', sort=False))['TeamId']
return df
def scrape_best_odds():
''' Scrape best odds offered for next round of matches'''
url = 'http://www.oddschecker.com/basketball/nba'
df = pd.read_html(get(url).text)[0]
df = df[pd.notnull(df.ix[:,1])].ix[:,1:3]
def parse_team(string):
return str.split(string, sep='(')[0].strip()
def parse_odds(string):
s = string.split(sep='(')[1]
s = s.replace(')', '')
f = s.split(sep='/')
d = 1
if(len(f) > 1):
d = float(s.split(sep='/')[1])
return (float(f[0])/d)+1
df['Date'] = datetime.today().strftime('%d/%m/%Y')
df['AwayTeam'] = df['1'].apply(parse_team)
df['HomeTeam'] = df['2'].apply(parse_team)
df['AwayOdds'] = df['1'].apply(parse_odds)
df['HomeOdds'] = df['2'].apply(parse_odds)
df['AwayOddsProb'] = 1/df['AwayOdds']
df['HomeOddsProb'] = 1/df['HomeOdds']
df = df[['Date', 'HomeTeam', 'AwayTeam', 'HomeOdds', 'HomeOddsProb',
'AwayOdds', 'AwayOddsProb']]
return df
def _get_cashflow_data(year, quarter, pageNo, dataArr,
retry_count=3, pause=0.001):
ct._write_console()
for _ in range(retry_count):
time.sleep(pause)
try:
request = Request(ct.CASHFLOW_URL%(ct.P_TYPE['http'], ct.DOMAINS['vsf'],
ct.PAGES['fd'], year,
quarter, pageNo, ct.PAGE_NUM[1]))
text = urlopen(request, timeout=10).read()
text = text.decode('GBK')
text = text.replace('--', '')
html = lxml.html.parse(StringIO(text))
res = html.xpath("//table[@class=\"list_table\"]/tr")
if ct.PY3:
sarr = [etree.tostring(node).decode('utf-8') for node in res]
else:
sarr = [etree.tostring(node) for node in res]
sarr = ''.join(sarr)
sarr = '<table>%s</table>'%sarr
df = pd.read_html(sarr)[0]
df.columns = ct.CASHFLOW_COLS
dataArr = dataArr.append(df, ignore_index=True)
nextPage = html.xpath('//div[@class=\"pages\"]/a[last()]/@onclick')
if len(nextPage)>0:
pageNo = re.findall(r'\d+', nextPage[0])[0]
return _get_cashflow_data(year, quarter, pageNo, dataArr)
else:
return dataArr
except Exception as e:
pass
raise IOError(ct.NETWORK_URL_ERROR_MSG)
def scrape_model_and_odds():
model = scrape_model_probs()
for i in range(len(model)):
home_team = model.HomeTeam[i]
away_team = model.AwayTeam[i]
url = _odds_checker_page(home_team, away_team)
df = pd.read_html(get(url).text)[0]
df0 = df.loc[0:1, bookies.keys()]
df0.columns = bookies.keys()
df0.index = [_adjust_portland(df[2][0]), _adjust_portland(df[2][1])]
odds = pd.DataFrame(data = df0.loc[home_team, :].apply(_parse_odds),
index = bookies.keys())
odds[away_team] = pd.DataFrame(data = df0.loc[away_team, :].apply(_parse_odds),
index = bookies.keys())
model.loc[i, 'HomeOdds'] = odds[home_team].max()
model.loc[i, 'HomeBookie'] = bookies[odds.sort_values(by=home_team).index[-1]]
model['HomeOddsProb'] = 1 / model.HomeOdds
model.loc[i, 'AwayOdds'] = odds[away_team].max()
model.loc[i, 'AwayBookie'] = bookies[odds.sort_values(by=away_team).index[-1]]
model['AwayOddsProb'] = 1 / model.AwayOdds
return model
def get(self, code, start=None, end=None, interval='d'):
if code in {'N225', 'GSPC', 'IXIC', 'DJI'}:
start = datetime.datetime.strptime(start, '%Y-%m-%d')
result = data.DataReader("".join(['^', code]),
'yahoo', start, end)
return result.asfreq('B')
base = self._base_url()
start, end = self._sanitize_dates(start, end)
start = 'sy={0}&sm={1}&sd={2}'.format(
start.year, start.month, start.day)
end = 'ey={0}&em={1}&ed={2}'.format(end.year, end.month, end.day)
p = 1
results = []
if interval not in ['d', 'w', 'm', 'v']:
raise ValueError(
"Invalid interval: valid values are 'd', 'w', 'm' and 'v'")
while True:
url = base.format(int(code), start, end, interval, p)
tables = pd.read_html(url, header=0)
if len(tables) < 2 or len(tables[1]) == 0:
break
results.append(tables[1])
p += 1
result = pd.concat(results, ignore_index=True)
result.columns = [
'Date', 'Open', 'High', 'Low', 'Close', 'Volume', 'Adj Close']
result['Date'] = pd.to_datetime(result['Date'], format='%Y年%m月%d日')
result = result.set_index('Date')
result = result.sort_index()
return result.asfreq('B')
# To avoid unwanted float values
pd.options.display.float_format = '{:,.0f}'.format
url = 'https://www.worldometers.info/coronavirus/'
# To avoid 403 Error
header = {
"User-Agent":
"Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/50.0.2661.75 Safari/537.36",
"X-Requested-With": "XMLHttpRequest"
}
r = requests.get(url, headers=header)
df = pd.read_html(r.text)
df = df[0]
df = df[1:212]
df.columns = [
'Country', 'TotalCases', 'NewCases', 'TotalDeaths', 'NewDeaths',
'TotalRecovered', 'ActiveCases', 'Critical', 'Tot Cases/1M pop',
'Deaths/1M pop', ' TotalTests', 'Tests/ 1M pop'
]
# Replace few countries names
df = df.replace(to_replace="UK", value="United Kingdom")
df = df.replace(to_replace='S. Korea', value="South Korea")
df = df.replace(to_replace='UAE', value="United Arab Emirates")
df = df.replace(to_replace='0', value=0)
# Add player links to player_links list.
for key, letter_soup in soups.items():
# First row is a header row, all the rest are players. Create a players list.
player_rows=letter_soup.table.find_all("tr")[1:]
counter=0
#Choose only centers that began playing after 1980
for idx, player in enumerate(player_rows):
if int(player.td.string) >= 1980 and fnmatch.fnmatch(player.find_all("td")[2].string, "*C*"):
player_links.append(base_link+player.th.a["href"])
counter+=1
"""Collection of individual player URLs is done, now we begin creating the 'player data' dataframe."""
player_data_df= player_df=pd.read_html(str(BeautifulSoup(urlopen(player_links[0])).table))[0].iloc[:-1, 0:30]
dataframe_index=1
countries=np.array(["albania","andorra","armenia","austria","azerbaijan","belarus","belgium",
" bosnia and herzegovina","bulgaria","croatia","cyprus","czech republic",
"denmark","estonia","finland","france","georgia","germany","greece","hungary",
"iceland","ireland","italy","kazakhstan","kosovo","latvia","liechtenstein",
"lithuania","luxembourg","macedonia","malta","moldova","monaco","montenegro",
"netherlands","norway","poland","portugal","romania","russia","san marino",
"serbia","slovakia","slovenia","spain","sweden","switzerland","turkey",
"ukraine","united kingdom","vatican city"])
states=np.empty(shape=len(us_states), dtype=object)
for index, state in enumerate(us_states):
states[index]=str(state).lower()
for player_link in player_links[1:]: # Creating and adding player dataframes together.
for link in soup_level1.find_all(
'a', id=re.compile("^MainContent_uxLevel2_JobTitles_uxJobTitleBtn_")):
#Selenium visits each Job Title page
python_button = driver.find_element_by_id(
'MainContent_uxLevel2_JobTitles_uxJobTitleBtn_' + str(x))
python_button.click() #click link
#Selenium hands of the source of the specific job page to Beautiful Soup
soup_level2 = BeautifulSoup(driver.page_source, 'html.parser')
#Beautiful Soup grabs the HTML table on the page
table = soup_level2.find_all('table')[0]
#Giving the HTML table to pandas to put in a dataframe object
df = pd.read_html(str(table), header=0)
#Store the dataframe in a list
datalist.append(df[0])
#Ask Selenium to click the back button
driver.execute_script("window.history.go(-1)")
#increment the counter variable before starting the loop over
x += 1
#end loop block
#loop has completed
#end the Selenium browser session
def scrape():
# URL of page to be scraped
url = "https://mars.nasa.gov/news/"
response = requests.get(url)
soup = BeautifulSoup(response.text, 'html.parser')
# print(response.text)
# Find latest news title about Mars
news_title = soup.find('div', class_="content_title").text
news_title
# Find latest news blurb
news_p = soup.find('div', class_="rollover_description_inner").text
news_p
# * Use splinter to navigate the site and find the image url for the current Featured Mars Image
executable_path = {'executable_path': 'chromedriver.exe'}
browser = Browser('chrome', **executable_path)
url = 'https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars'
browser.visit(url)
featured_image = browser.find_by_id('full_image')
featured_image.click()
time.sleep(5)
more_info = browser.find_link_by_partial_text('more info')
more_info.click()
# Pull featured image url
html = browser.html
soupsearch = BeautifulSoup(html, 'html.parser')
part_image_url = soupsearch.find('img', class_='main_image').get('src')
featured_image_url = 'https://www.jpl.nasa.gov' + part_image_url
featured_image_url
# Exit browser
browser.quit()
# Visit the Mars Weather twitter account [here](https://twitter.com/marswxreport?lang=en)
# and scrape the latest Mars weather tweet from the page.
url = "https://twitter.com/marswxreport?lang=en"
response = requests.get(url)
soup = BeautifulSoup(response.text, 'html.parser')
mars_weather = soup.find('div', class_='js-tweet-text-container').text
mars_weather
# # Pull Mars facts table from Space-Facts
executable_path = {'executable_path': 'chromedriver.exe'}
browser = Browser('chrome', **executable_path)
url = 'https://space-facts.com/mars/'
marsFacts_df = pd.read_html(url)
marsFacts_df = marsFacts_df[0]
marsFacts_df
# # * Use Pandas to convert the data to a HTML table string.
# marsFacts_df.to_html('mars_facts.html', index=False)
marsHTML = marsFacts_df.to_html()
print(marsHTML)
executable_path = {'executable_path': 'chromedriver.exe'}
browser = Browser('chrome', **executable_path)
url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(url)
cerberus = browser.find_link_by_partial_text('Cerberus')
cerberus.click()
html = browser.html
soupsearch = BeautifulSoup(html, 'html.parser')
astrogeology_url = 'https://astrogeology.usgs.gov/'
#---------------------------------------
cerberus_url = soupsearch.find('img', class_='wide-image').get('src')
cerberus_img_url = astrogeology_url + cerberus_url
print('cerberus image')
print(cerberus_img_url)
back = browser.find_link_by_partial_text('Back')
back.click()
#---------------------------------------
schiaparelli = browser.find_link_by_partial_text('Schiaparelli')
schiaparelli.click()
html = browser.html
soupsearch = BeautifulSoup(html, 'html.parser')
schiaparelli_url = soupsearch.find('img', class_='wide-image').get('src')
schiaparelli_img_url = astrogeology_url + schiaparelli_url
back = browser.find_link_by_partial_text('Back')
back.click()
print('schiaparelli image')
print(schiaparelli_img_url)
#---------------------------------------
syrtis = browser.find_link_by_partial_text('Syrtis')
syrtis.click()
html = browser.html
soupsearch = BeautifulSoup(html, 'html.parser')
syrtis_url = soupsearch.find('img', class_='wide-image').get('src')
syrtis_img_url = astrogeology_url + syrtis_url
back = browser.find_link_by_partial_text('Back')
back.click()
valles = browser.find_link_by_partial_text('Valles')
valles.click()
html = browser.html
soupsearch = BeautifulSoup(html, 'html.parser')
valles_url = soupsearch.find('img', class_='wide-image').get('src')
valles_img_url = astrogeology_url + valles_url
valles_img_url
print(cerberus_img_url, schiaparelli_img_url, syrtis_img_url, valles_img_url)
# # Scrape Hemisphere image urls
# executable_path = {'executable_path': 'chromedriver.exe'}
# browser = Browser('chrome', **executable_path)
# url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
# browser.visit(url)
# cerberus = browser.find_link_by_partial_text('Cerberus')
# cerberus.click()
# html = browser.html
# soupsearch = BeautifulSoup(html, 'html.parser')
# astrogeology_url = 'https://astrogeology.usgs.gov/'
# #---------------------------------------
# cerberus_url = soupsearch.find('img', class_='wide-image').get('src')
# cerberus_img_url = astrogeology_url + cerberus_url
# back = browser.find_link_by_partial_text('Back')
# # back.click()
# #---------------------------------------
# url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
# browser.visit(url)
# schiaparelli = browser.find_link_by_partial_text('Schiaparelli')
# schiaparelli.click()
# time.sleep(2)
# schiaparelli_url = soupsearch.find('img', class_='wide-image').get('src')
# schiaparelli_img_url = astrogeology_url + schiaparelli_url
# back = browser.find_link_by_partial_text('Back')
# back.click()
# #---------------------------------------
# syrtis = browser.find_link_by_partial_text('Syrtis')
# syrtis.click()
# time.sleep(2)
# syrtis_url = soupsearch.find('img', class_='wide-image').get('src')
# syrtis_img_url = astrogeology_url + syrtis_url
# back = browser.find_link_by_partial_text('Back')
# back.click()
# valles = browser.find_link_by_partial_text('Valles')
# valles.click()
# time.sleep(2)
# valles_url = soupsearch.find('img', class_='wide-image').get('src')
# valles_img_url = astrogeology_url + valles_url
# valles_img_url
# # Exit browser
# browser.quit()
# print(cerberus_img_url, schiaparelli_img_url, syrtis_img_url, valles_img_url)
# Save hemisphere image urls in a dictionary.
hemisphere_image_urls = [
{"title": "Valles Marineris Hemisphere", "img_url": valles_img_url},
{"title": "Cerberus Hemisphere", "img_url": cerberus_img_url},
{"title": "Schiaparelli Hemisphere", "img_url": schiaparelli_img_url},
{"title": "Syrtis Major Hemisphere", "img_url": syrtis_img_url},
]
print(hemisphere_image_urls)
# Save all variables in a dictionary
mars_data = {
"hemisphere_image_urls": hemisphere_image_urls,
"news_p" : news_p,
"news_title" : news_title,
"featured_image_url": featured_image_url,
"mars_weather": mars_weather,
"mars_facts": marsHTML
}
return mars_data
html = browser.html
image_soup = BeautifulSoup(html, 'html.parser')
# use CSS selector in BeautifulSoup to extract img_url
image_relative_url = image_soup.select_one('figure.lede a img').get('src')
#combine with base URL to create an absolute img URL
img_url = f'https://www.jpl.nasa.gov{image_relative_url}'
img_url
# check if a valid img_url
#browser.visit(img_url)
# %% [markdown]
# ## Web Scrape TABLE from Mars facts website
# - (use Pandas funtions to parse HTML Table)
# - (No BeautifulSoup used)
# - (No auto browser used)
# %%
fact_url = 'http://space-facts.com/mars/' # only accept http without s
# extract the first one in list of DFs
df = pandas.read_html(fact_url)[0]
df.columns = ['description', 'value']
df.set_index('description', inplace=True)
df
# convert back to HTML string
fact_table_html = df.to_html()
fact_table_html
# %%
browser.quit()
# %%
def scrape_all():
browser = init_browser()
url = 'https://mars.nasa.gov/news'
browser.visit(url)
html = browser.html
soup = bs(html, 'html.parser')
titles = soup.find_all('div', class_='content_title')
texts = soup.find_all('div', class_='article_teaser_body')
title_text = []
text_only = []
#keep only the text
for x in titles:
title_text.append(x.text.strip())
for x in texts:
text_only.append(x.text.strip())
# JPL Mars Space Image
#These lines of code are needed to navigate to the next page
image_url = 'https://data-class-jpl-space.s3.amazonaws.com/JPL_Space/index.html'
browser.visit(image_url)
html = browser.html
soup = bs(html, 'html.parser')
image_url_src = soup.find('img', class_='headerimage fade-in')['src']
url_short = image_url.split('/')
#rearrange and concatenate URL
featured_image_url = url_short[0] + '//' + url_short[1] + url_short[
2] + '/' + url_short[3] + '/' + image_url_src
# Mars Facts
facts_url = 'https://space-facts.com/mars/'
df = pd.read_html(facts_url)[0]
mars_facts = df.to_html()
# Mars Hemispheres
pic_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(pic_url)
hemi_url = []
links = browser.find_by_css('a.product-item h3')
for i in range(len(links)):
hemi = {}
browser.find_by_css('a.product-item h3')[i].click()
sample_image = browser.links.find_by_text('Sample').first
hemi['img_url'] = sample_image['href']
hemi['title'] = browser.find_by_css('h2.title').text
hemi_url.append(hemi)
browser.back()
browser.quit()
# Store data in one dictionary
mars_data = {
"news_title": title_text,
"news_paragraph": text_only,
"featured_image": featured_image_url,
"mars_facts": mars_facts,
"hemispheres": hemi_url
}
return mars_data
import requests
import pandas as pd
wiki = 'https://en.wikipedia.org/wiki/List_of_postal_codes_of_Canada:_M'
wikipedia_page = requests.get(wiki)
df_raw = pd.read_html(wikipedia_page.content, header=0)[0]
df_new = df_raw[df_raw.Borough != 'Not assigned']
df_new.head()
df_new.loc[df_new.Neighborhood == 'Not assigned']
df_new.Neighborhood.replace('Not assigned', df_new.Borough, inplace=True)
df_new.head(8)
df_toronto = df_new.groupby(['Postal Code', 'Borough'
])['Neighborhood'].apply(lambda x: ', '.join(x))
df_toronto = df_toronto.reset_index()
df_toronto.rename(columns={'Postal Code': 'PostCode'}, inplace=True)
df_toronto.rename(columns={'Neighborhood': 'Neighbourhood'}, inplace=True)
df_toronto.head()
df_toronto.shape
@author: Brad
"""
#Importing libraries
import sqlalchemy
import pandas as pd
import time
#Loading login credentials into Python
credentials = pd.read_csv('credentials.csv')
SQL_user = credentials.loc[0, 'SQL_user']
SQL_pass = credentials.loc[0, 'SQL_pass']
API_key = credentials.loc[0, 'API_key']
#Creating list of stocks for table
asx_200 = pd.read_html('https://en.wikipedia.org/wiki/S%26P/ASX_200')
asx_200 = asx_200[0][0]
asx_200 = asx_200[1:]
asx_200.columns = ['Symbol']
asx_200 = asx_200.str.lower()
#Setting up database
engine = sqlalchemy.create_engine('mysql+mysqlconnector://' + str(SQL_user) +
':' + str(SQL_pass) + '@localhost:3306')
con = engine.connect()
con.execute('CREATE database ASX_API')
con.close()
#Creating connection with MySQL server for API upload
engine = sqlalchemy.create_engine('mysql+mysqlconnector://' + str(SQL_user) +
':' + str(SQL_pass) +
def scrape():
browser = init_browser()
# mars_scrapped_data = {}
url= 'https://mars.nasa.gov/news/'
browser.visit(url)
html= browser.html
soup = BeautifulSoup(html, 'html.parser')
news_title = soup.find('div', class_='content_title').text
news_p = soup.find('div', class_='article_teaser_body').text
# just to check the output
print(f"Title: {news_title}")
print(f"Paragraph: {news_p}")
url_image= 'https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars'
browser.visit(url_image)
browser.click_link_by_partial_text('FULL IMAGE')
time.sleep(3)
browser.click_link_by_partial_text('more info')
new_html= browser.html
imgsoup = BeautifulSoup(new_html, 'html.parser')
temp_img = imgsoup.find('img', class_='main_image')['src']
featured_image_url= 'https://www.jpl.nasa.gov/' + temp_img
print(featured_image_url)
mars_twitter_url= 'https://twitter.com/marswxreport?lang=en'
browser.visit(mars_twitter_url)
mars_twitter= browser.html
soup = BeautifulSoup(mars_twitter, 'html.parser')
find_tweet = soup.find('p', class_='TweetTextSize').text
mars_weather= find_tweet
print(f"Latest Tweet: {mars_weather}")
mars_facts_url = 'https://space-facts.com/mars/'
tables = pd.read_html(mars_facts_url)
tables
df = tables[0]
df.columns = ['Profile', 'Details']
df.head()
df.set_index('Profile', inplace=True)
df.head()
html_table = df.to_html()
html_table
html_table.replace('\n', '')
df.to_html('table.html')
url_mars= 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(url_mars)
hemi_dicts = []
for i in range(1,9,2):
hemi_dict = {}
browser.visit(url_mars)
# time.sleep(1)
hemispheres_html = browser.html
hemispheres_soup = BeautifulSoup(hemispheres_html, 'html.parser')
hemi_name_links = hemispheres_soup.find_all('a', class_='product-item')
hemi_name = hemi_name_links[i].text.strip('Enhanced')
detail_links = browser.find_by_css('a.product-item')
detail_links[i].click()
time.sleep(1)
browser.find_link_by_text('Sample').first.click()
time.sleep(1)
browser.windows.current = browser.windows[-1]
hemi_img_html = browser.html
browser.windows.current = browser.windows[0]
browser.windows[-1].close()
hemi_img_soup = BeautifulSoup(hemi_img_html, 'html.parser')
hemi_img_path = hemi_img_soup.find('img')['src']
print(hemi_name)
hemi_dict['title'] = hemi_name.strip()
print(hemi_img_path)
hemi_dict['img_url'] = hemi_img_path
hemi_dicts.append(hemi_dict)
mars_scrapped_data ={"news_title": news_title,
"news_paragraph": news_p,
"featured_image": featured_image_url,
"Latest_Tweet": mars_weather,
"Hemispheres_details": hemi_dicts
# "Table": html_table
}
return mars_scrapped_data
def init_browser():
executable_path = {'executable_path': 'chromedriver.exe'}
return browser('chrome', **executable_path, headless=False)
def scrape_info():
browser = init_browser()
url = 'https://mars.nasa.gov/news/?page=0&per_page=40&order=publish_date+desc%2Ccreated_at+desc&search=&category=19%2C165%2C184%2C204&blank_scope=Latest'
browser.visit(url)
time.sleep(5)
# HTML
html_news = browser.html
# Parse HTML with Beautiful Soup
soup = BeautifulSoup(html_news, 'html.parser')
news_title = soup.find("div", class_="content_title").text
news_p = soup.find("div", class_ ="article_teaser_body").text
news_p
browser.quit()
# Feature Image
featured_img_url = 'https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars'
browser.visit(featured_img_url)
browser.links.find_by_partial_text('FULL IMAGE')
browser.links.find_by_partial_text('more info')
browser.links.find_by_partial_text('jpg')
# HTML Object
html_image = browser.html
# Parse HTML with Beautiful Soup
soup = BeautifulSoup(html_image, 'html.parser')
featured_img_url = soup.find('img')['src']
featured_img_url
# Mars Weather
weather_url = 'https://twitter.com/marswxreport?lang=en'
browser.visit(weather_url)
time.sleep(5)
html_weather = browser.html
weather_soup = BeautifulSoup(html_weather, "html.parser")
mars_weather=weather_soup.find('div', class_ ='css-901oao r-hkyrab r-1qd0xha r-a023e6 r-16dba41 r-ad9z0x r-bcqeeo r-bnwqim r-qvutc0').text
mars_weather
# Mars Facts
mars_df = pd.read_html("https://space-facts.com/mars/")[0]
#Panda DataFrame
#mars_df.columns=["Des", "MARS PLANET PROFILE"]
#mars_df.set_index("Des", inplace=True)
#mars_df
mars_df.columns=["Description", "Value"]
mars_df.set_index("Description", inplace = True)
mars_facts = mars_df.to_html(classes="table")
mars_facts =mars_df.replace("'","")
mars_facts
# MARS HEMISPHERES
# Cerberus Hemisphere
mars_hemi_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(mars_hemi_url)
# Click to URL of page to be scraped and extract data
browser.click_link_by_partial_text('Cerberus Hemisphere Enhanced')
html_mars_hemi = browser.html
soup_mars_hemi = BeautifulSoup(mars_hemi_url, 'html.parser')
#Pull specific data from webpage
hemi_title1_url = soup_mars_hemi.find('h2', class_ ='title').text
img1_url =soup_mars_hemi.find('a', text ='Sample').get("href")
#Put data into a dictionary
one ={'title':hemi_title1_url, "img_url":img1_url}
one
# Schiaparelli Hemisphere
mars_hemi_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(mars_hemi_url)
# Click to URL of page to be scraped and extract data
browser.click_link_by_partial_text('Schiaparelli Hemisphere Enhanced')
html_mars_hemi = browser.html
soup_mars_hemi = BeautifulSoup(html_mars_hemi, 'html.parser')
#Pull specific data from webpage
hemi_title2_url = soup_mars_hemi.find('h2', class_ ='title').text
img2_url =soup_mars_hemi.find('a', text ='Sample').get("href")
#Put data into a dictionary
two ={'title':hemi_title2_url, "img_url":img2_url}
two
# Syrtis Major Hemisphere
mars_hemi_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(mars_hemi_url)
# Click to URL of page to be scraped and extract data
browser.click_link_by_partial_text('Syrtis Major Hemisphere Enhanced')
html_mars_hemi = browser.html
soup_mars_hemi = BeautifulSoup(html_mars_hemi, 'html.parser')
# Pull specific data from webpage
hemi_title3_url = soup_mars_hemi.find('h2', class_ ='title').text
img3_url =soup_mars_hemi.find('a', text ='Sample').get("href")
#Put data into a dictionary
three ={'title':hemi_title3_url, "img_url":img3_url}
three
# Valles Marineris Hemisphere
mars_hemi_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(mars_hemi_url)
browser.click_link_by_partial_text('Syrtis Major Hemisphere Enhanced')
html_mars_hemi = browser.html
soup_mars_hemi = BeautifulSoup(html_mars_hemi, 'html.parser')
#Pull specific data from webpage
hemi_title4_url = soup_mars_hemi.find('h2', class_ ='title').text
img4_url =soup_mars_hemi.find('a', text ='Sample').get("href")
#Put data into a dictionary
four ={'title':hemi_title4_url, "img_url":img4_url}
four
#Summary
hemisphere_url = [one, two, three, four]
hemisphere_url
# Store data in a dictionary
mars_data = {
"News_Title" : news_p,
"Featured_Image" : featured_img_url,
"Mars_Weather" : mars_weather,
"Mars_Facts" : mars_facts,
"Hemisphere_Images" : hemisphere_img_url
}
browser.quit()
# Return results
return mars_data
def scrape_info():
# Making space soup
browser = init_browser()
url = 'https://www.nasa.gov/missions/'
html = browser.visit(url)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
# Stiring the soup
links = [
a for a in soup.find('div', class_="static-landing-page").find_all(
'a', href=True)
]
# Pouring out the list and title
link_list = []
title_list = []
for i in range(len(links)):
if links[i].get('href').find("mission_pages") == 1:
link_list.append(links[i]['href'])
title_list.append(links[i].text)
else:
print('No Mission Page')
mission_dict = {}
mission_dict["Mission"] = title_list
mission_dict["Mission Link"] = link_list
url = 'https://www.cdscc.nasa.gov/Pages/trackingtoday.html'
html = browser.visit(url)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
abv_table = pd.read_html(url)[3]
abv_table.columns = ['ABV', 'Name']
abv_dict = abv_table.to_dict("records")
MADRID = {}
GOLDSTONE = {}
CANBERRA = {}
browser.visit('https://eyes.jpl.nasa.gov/dsn/dsn.html')
time.sleep(.3)
for i in browser.find_by_tag('a'):
if i['class'] == 'inactive' or None:
pass
elif i.text == '':
pass
elif i['id'] == '' or None:
pass
else:
if i['id'][:2] == 'sp':
ABV = i.text
if i['id'][-5] == '0':
MADRID[ABV] = {}
elif i['id'][-5] == '1':
GOLDSTONE[ABV] = {}
elif i['id'][-5] == '2':
CANBERRA[ABV] = {}
# Platting the soup
mission_data = {
"Madrid": MADRID,
"Goldstone": GOLDSTONE,
"Canberra": CANBERRA,
"Mission_titles": mission_dict,
"Mission_Code": abv_dict
}
browser.quit()
return mission_data
def get_family_info_from_mysqldb(save_dataset=True):
# 建立连接
conn = pymysql.connect(host='172.16.201.103',
port=3306,
user='******',
password='******',
db='resource',
charset='utf8')
# 获取游标
cursor = conn.cursor(pymysql.cursors.DictCursor)
# 执行sql语句
# 获取构件库中所有的族（包含各专业）
sql = 'select i.id, i.name, i.resource_desc ' \
'from resource_item i, resource_parameter_value pv ' \
'where i.parameter_code=pv.code and i.resource_lib_id=54 and i.could_be_shown_in_front=1'
rows = cursor.execute(sql)
items = cursor.fetchall()
# 获取所有族的参数
sql = 'select ip.resource_item_id , i.name, ip.resource_parameter_id, p.name as parameter, pv.id as param_id, pv.value ' \
'from resource_item i, resource_item_parameter ip, resource_parameter p, resource_parameter_value pv ' \
'where i.id=ip.resource_item_id and p.id=ip.resource_parameter_id and ip.resource_parameter_value_id=pv.id ' \
'and (pv.resource_parameter_id=1 or pv.resource_parameter_id=51 or pv.resource_parameter_id=52 ' \
'or pv.resource_parameter_id=53 or pv.resource_parameter_id=10004) '
rows = cursor.execute(sql)
item_params = cursor.fetchall()
# 获取结构化参数
sql = 'select pv.id, pv.code, pv.`value` from resource_parameter_value pv ' \
'where (pv.resource_parameter_id=1 or pv.resource_parameter_id=51 or pv.resource_parameter_id=52 ' \
'or pv.resource_parameter_id=53 or pv.resource_parameter_id=10004) ' \
'order by code'
rows = cursor.execute(sql)
params = cursor.fetchall()
# 关闭游标
cursor.close()
# 关闭连接
conn.cursor()
items_dict = {}
for idx in range(len(items)):
# if items[idx]["id"] not in items_dict:
try:
items[idx]["resource_desc"] = pd.read_html(
items[idx]["resource_desc"],
header=0)[0].to_dict(orient="records")
except:
pass
if items[idx]["name"][0] not in ["A", "S"]: # 过滤土建专业
continue
item_id = items[idx]["id"]
items[idx].pop("id", None)
items[idx].pop("resource_desc", None)
items_dict[item_id] = items[idx]
tmp1 = {c["code"]: c["id"] for c in params if c["code"]}
tmp2 = {c["id"]: c["value"] for c in params}
pid2v = {}
for c in params:
if not c["code"]:
pid2v[c["id"]] = [c["value"]]
else:
ks = c["code"].split("_")
pid2v[c["id"]] = []
for i in range(len(ks)):
tmp_k = "_".join(ks[:i + 1])
if tmp1[tmp_k] in pid2v:
pid2v[c["id"]].append(tmp2[tmp1[tmp_k]])
for idx in range(len(item_params)):
p = item_params[idx]
iid = p["resource_item_id"]
if iid in items_dict:
if p["parameter"] not in items_dict[iid]:
items_dict[iid][item_params[idx]["parameter"]] = []
items_dict[iid][item_params[idx]["parameter"]] += pid2v[
p["param_id"]]
# 保存结果
if save_dataset:
with open(
os.path.join(
os.path.dirname(os.path.dirname(
os.path.abspath(__file__))),
"data/standard_vocab.json"), "w") as f:
json.dump(list(items_dict.values()), f)
if save_dataset:
with open(
os.path.join(
os.path.dirname(os.path.dirname(
os.path.abspath(__file__))),
"data/standard_param.json"), "w") as f:
json.dump(params, f)
return items_dict
'Visitors': [43,34,65,56,29,76],
'Bounce Rate': [65,67,78,65,45,52]
}
df = pd.DataFrame(web_stats)
df.set_index('Day', inplace=True)
'''
##read write csv
dfx = pd.read_csv('/Users/zhangsicai/Desktop/Panda/grade.csv')
dfx.set_index('Date', inplace=True)
dfx.rename(columns={'math': 'shuxue'}, inplace=True)
print(dfx['shuxue'])
#df['math'].to_csv('/Users/zhangsicai/Desktop/Panda/grade1.csv')
df = quandl.get('FMAC/HPI_TX', authtoken='HUg-EPXknoSxzbk26DMu')
fiddy_states = pd.read_html(
'https://simple.wikipedia.org/wiki/List_of_U.S._states')
print(df.head())
print(df[0:2])
print(df['NSA Value'])
df.plot()
plt.show()
print(fiddy_states[0]['Name'])
'''
for dd in df['NSA Value'][1:]:
print(dd)
'''
df1 = pd.DataFrame(
{
def scrape():
executable_path = {'executable_path': '/usr/local/bin/chromedriver'}
browser = Browser('chrome', **executable_path, headless=False)
# NASA MARS NEWS
news_url = (
'https://mars.nasa.gov/news/?page=0&per_page=40'
'&order=publish_date+desc%2Ccreated_at+desc&search=&category=19%2C165%2C184%2C204&blank_scope=Latest'
)
browser.visit(news_url)
time.sleep(2)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
news_title = soup.body.find_all('div',
class_="content_title")[1].text.strip()
news_p = soup.body.find_all('div',
class_="article_teaser_body")[0].text.strip()
#JPL MARS SPACE IMAGES - FEATURED IMAGE
image_url = "https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars"
browser.visit(image_url)
time.sleep(2)
my_xpath = '/html/body/div[1]/div/div[3]/section[1]/div/div/article/div[1]/footer/a'
results = browser.find_by_xpath(my_xpath)
img = results[0]
img.click()
browser.click_link_by_partial_text('more info')
html1 = browser.html
soup = BeautifulSoup(html1, 'html.parser')
feat_img = soup.find_all('figure', class_='lede')
feat_img_result = feat_img[0].a['href']
featured_image_url = 'https://www.jpl.nasa.gov' + feat_img_result
# MARS FACTS
facts_url = 'https://space-facts.com/mars/'
facts_table = pd.read_html(facts_url)
table_df = facts_table[0]
# mars_table_df = table_df.rename(columns={0: 'Mars: Measurement', 1: 'Measurement: Value'})
mars_table_df = table_df.to_html(header=False, index=False)
# mars_table_df.to_html(classes="table table-striped")
print(mars_table_df)
# MARS HEMISPHERES
#Note the inconsistent url
hemispheres_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
#alternate site, if previous site is unavailable
# hemispheres_url = 'https://web.archive.org/web/20181114171728/https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(hemispheres_url)
time.sleep(2)
hemisphere_image_urls = []
url_links = browser.find_by_css('a.product-item h3')
for i in range(len(url_links)):
# create an empty dictionary for each hemisphere
hemisphere = {}
browser.find_by_css('a.product-item h3')[i].click()
#get hemisphere title
hemisphere['title'] = browser.find_by_css("h2.title").text
#next find the sample image anchor tag and get href
sample_elem = browser.find_link_by_text('Sample').first
hemisphere['img_url'] = sample_elem['href']
#Append hemisphere object to list
hemisphere_image_urls.append(hemisphere)
#Finally navigate back to start again on loop
browser.back()
#*************** CREATE A DICTOIONARY *********************
mars_info = {}
mars_info['news_title'] = news_title
mars_info['news_detail'] = news_p
mars_info['featured_img_url'] = featured_image_url
mars_info['mars_facts_html'] = mars_table_df
mars_info['hemisphere_image_urls'] = hemisphere_image_urls
# Close the browser
browser.quit()
# Return results
return mars_info
#
# xpaths = [
# '/html/body/div[1]/div[1]/div[2]/section/div/div[2]/div[1]/div/a',
# '/html/body/div[1]/div[1]/div[2]/section/div/div[2]/div[2]/div/a',
# '/html/body/div[1]/div[1]/div[2]/section/div/div[2]/div[3]/div/a',
# '/html/body/div[1]/div[1]/div[2]/section/div/div[2]/div[4]/div/a'
# ]
# hem_title = []
# hem_url = []
# mars_hem_title_url = []
# for path in xpaths :
# results = browser.find_by_xpath(path)
# img = results[0]
# img.click()
# html = browser.html
# soup = BeautifulSoup(html, 'html.parser')
# title = soup.find('h2', class_ = 'title').text
# hem_title.append(title)
# hem = soup.find('div', class_='downloads')
# hem_result = hem
# img_url = hem_result.find('a')['href']
# hem_url.append(img_url)
# mars_hem_title_url.append({'title': title, 'img_url': img_url})
# browser.visit(hemispheres_url)
# browser.quit()
# #Store results in dictionary
# notebook_dict = {}
# notebook_dict = {
# 'article_title': news_title,
# 'article_paragraph': news_p,
# 'mars_image': featured_image_url,
# 'mars_data_table': mars_table_df,
# 'hemisphere_image_urls': mars_hem_title_url}
# print(f"index 0 {notebook_dict['article_title']}")
# print(f"index 1 {notebook_dict['article_paragraph']}")
# print(f"index 2 {notebook_dict['mars_image']}")
# print(f"index 3 {notebook_dict['mars_data_table']}")
# print(f"index 4 {notebook_dict['hemisphere_image_urls']}")
# return notebook_dict
def scrape():
browser = init_browser()
url = 'https://mars.nasa.gov/news/?page=0&per_page=40&order=publish_date+desc%2Ccreated_at+desc&search=&category=19%2C165%2C184%2C204&blank_scope=Latest'
response = requests.get(url)
time.sleep(1)
soup = BeautifulSoup(response.text, 'lxml')
#grabbling the 'slide' class element from the url
results = soup.find_all(class_="slide")
#creating a list to hold scraped data
news_data = []
for result in results:
# Error handling
try: #loop thru and get the text within these classes, replace \n with blank space
news_p = result.find(
class_="rollover_description_inner").text.replace('\n', '')
news_title = result.find(class_="content_title").text.replace(
'\n', '')
post = {"news_title": news_title, "news_p": news_p}
news_data.append(post)
print(post)
except Exception as e:
print(e)
browser = Browser('chrome', headless=False)
url = 'https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars'
browser.visit(url)
time.sleep(1)
#use splinter to click the "Full Image" button
browser.click_link_by_partial_text('FULL IMAGE')
time.sleep(1)
#HTML Object
html = browser.html
# Parse HTML with Beautiful Soup
soup = BeautifulSoup(html, 'html.parser')
#find the class where pic is stored
results = soup.find(class_='fancybox-image')
#retrieve source attribute, i.e. the path
url = results['src']
#attach the path to the main site link, this is the full image link
featured_image_url = 'https://www.jpl.nasa.gov' + url
post_two = {'featured_image': featured_image_url}
news_data.append(post_two)
print(post_two)
#visit the mars twitter page to get the Weather
url = 'https://twitter.com/marswxreport?lang=en'
browser.visit(url)
time.sleep(1)
response = requests.get(url)
#parse HTML with Beautiful soup, get the text
soup = BeautifulSoup(response.text, 'html.parser')
#get the text from the first p tag with appropriate class (from inspecting the site)
mars_weather = soup.find(
'p',
class_="TweetTextSize TweetTextSize--normal js-tweet-text tweet-text"
).text
post_three = {'mars_weather': mars_weather}
print(post_three)
news_data.append(post_three)
browser = Browser('chrome', headless=False)
#visit the mars space facts site
url = 'https://space-facts.com/mars/'
#read the table, put into list variable
tables = pd.read_html(url)
#convert the list to a dataframe
mars_df = tables[0]
#put column headers on
mars_df.columns = ["Characteristic", "Value"]
#convert the datframe to dictionary, using 'records' orientation, this does not neeed to be, nor should be, appended to news_data, as will create a list of a dictionary within the list, and not be able to be inserted to mongodb
mars_dict = mars_df.to_dict('records')
print(mars_dict)
#Visit the site to get images of Mars Hemispheres
url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(url)
time.sleep(1)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
results = soup.find_all(class_='item')
#loop through the item class
for result in results:
#find the first a tag
link = result.find('a')
#assign the href to variable 'links'
links = link['href']
#assign the link h3 title text to variable 'title'
title = result.find('h3').text
#concatenate the path with the main site link, assign to variable 'url'
url = 'https://astrogeology.usgs.gov' + links
#open brower, chromedriver
browser = Browser('chrome', headless=False)
#visit the concatenated url
browser.visit(url)
time.sleep(1)
html = browser.html
#parse the html with beautiful soup
soup = BeautifulSoup(html, 'html.parser')
#find all elemenst with class 'downloads', assign results to variable list 'infos'
infos = soup.find_all(class_='downloads')
#loop thru infos, pull out links to images, assign with title to dictionary post, and then append to list
#mars_images
for info in infos:
link_two = info.find('a')
img_url = link_two['href']
post_four = {'img_url': img_url, 'title': title}
news_data.append(post_four)
print(post_four)
#return your data, so it can be accessed by flask app (where the insertion into mongodb will occur)
return news_data + mars_dict
browser = Browser('chrome', headless=False)
weather_url = 'https://twitter.com/marswxreport?lang=en'
browser.visit(weather_url)
html = browser.html
weather_soup = BeautifulSoup(html, 'html.parser')
weather = weather_soup.find('div', class_='js-tweet-text-container')
mars_weather= weather.p.text.lstrip()
print(mars_weather)
facts_url = 'http://space-facts.com/mars/'
fact_table = pd.read_html(facts_url)
fact_table
df = fact_table[0]
df.columns = ['Mars', 'Value']
df
html_table = df.to_html()
df.to_html('table.html')
mars_facts=df.to_dict('records')
mars_facts
tem=list(mars_facts[0].values())
tem
def _read_change_from_url(self, url: str) -> pd.DataFrame:
"""read change from url
Parameters
----------
url : str
change url
Returns
-------
pd.DataFrame:
symbol date type
SH600000 2019-11-11 add
SH600000 2020-11-10 remove
dtypes:
symbol: str
date: pd.Timestamp
type: str, value from ["add", "remove"]
"""
resp = retry_request(url)
_text = resp.text
date_list = re.findall(r"(\d{4}).*?年.*?(\d+).*?月.*?(\d+).*?日", _text)
if len(date_list) >= 2:
add_date = pd.Timestamp("-".join(date_list[0]))
else:
_date = pd.Timestamp("-".join(re.findall(r"(\d{4}).*?年.*?(\d+).*?月", _text)[0]))
add_date = get_trading_date_by_shift(self.calendar_list, _date, shift=0)
remove_date = get_trading_date_by_shift(self.calendar_list, add_date, shift=-1)
logger.info(f"get {add_date} changes")
try:
excel_url = re.findall('.*href="(.*?xls.*?)".*', _text)[0]
content = retry_request(f"http://www.csindex.com.cn{excel_url}", exclude_status=[404]).content
_io = BytesIO(content)
df_map = pd.read_excel(_io, sheet_name=None)
with self.cache_dir.joinpath(
f"{self.index_name.lower()}_changes_{add_date.strftime('%Y%m%d')}.{excel_url.split('.')[-1]}"
).open("wb") as fp:
fp.write(content)
tmp = []
for _s_name, _type, _date in [("调入", self.ADD, add_date), ("调出", self.REMOVE, remove_date)]:
_df = df_map[_s_name]
_df = _df.loc[_df["指数代码"] == self.index_code, ["证券代码"]]
_df = _df.applymap(self.normalize_symbol)
_df.columns = [self.SYMBOL_FIELD_NAME]
_df["type"] = _type
_df[self.DATE_FIELD_NAME] = _date
tmp.append(_df)
df = pd.concat(tmp)
except Exception as e:
df = None
_tmp_count = 0
for _df in pd.read_html(resp.content):
if _df.shape[-1] != 4:
continue
_tmp_count += 1
if self.html_table_index + 1 > _tmp_count:
continue
tmp = []
for _s, _type, _date in [
(_df.iloc[2:, 0], self.REMOVE, remove_date),
(_df.iloc[2:, 2], self.ADD, add_date),
]:
_tmp_df = pd.DataFrame()
_tmp_df[self.SYMBOL_FIELD_NAME] = _s.map(self.normalize_symbol)
_tmp_df["type"] = _type
_tmp_df[self.DATE_FIELD_NAME] = _date
tmp.append(_tmp_df)
df = pd.concat(tmp)
df.to_csv(
str(
self.cache_dir.joinpath(
f"{self.index_name.lower()}_changes_{add_date.strftime('%Y%m%d')}.csv"
).resolve()
)
)
break
return df
def state_list():
fiddy_states = pd.read_html(
'https://simple.wikipedia.org/wiki/List_of_U.S._states')
return fiddy_states[0][0][1:]
# In[1]:
import os
import glob
import pandas as pd
import numpy as np
import time
from datetime import datetime
# In[2]:
dep_df, = pd.read_html("https://www.dublinairport.com/flight-information/live-departures", header=0)
dep_df.tail(1)
# In[3]:
# Initial Cleaning
dep1_df = dep_df.dropna()
dep1_df = dep1_df.drop('Status', axis=1)
dep1_df.columns = ['Terminal', 'Destination', 'Airline', 'Flight No.', 'Scheduled DateTime', 'Actual Departure']
# Month Column
new2 = dep1_df["Scheduled DateTime"].str.split(" ", n = 2, expand = True)
dep1_df["Month"]= new2[1]
def scrape_eng_pages(filename, sheet, check):
print("SHEETNAME********:", sheet)
book = load_workbook(filename)
old_sheet = sheet
sheet = book[sheet]
dictionary = {}
msg = ""
# filling up dictionary with city and corresponding website values
first = False
for row in sheet.rows:
dictionary[row[2].value] = row[7].value
keywords = ['Design', 'Professional ', 'Consult', 'Civil', 'Transportation', 'Bike', 'Pedestrian', 'Sidewalk', 'Street'
'Road','Boulevard', 'Blvd', 'Way', 'On-call']
keywords_track = {}
pattern = re.compile(
"(Jan(uary)?|Feb(ruary)?|Mar(ch)?|Apr(il)?|May|Jun(e)?|"
"Jul(y)?|Aug(ust)?|Sep(tember)?|Oct(ober)?|Nov(ember)?|"
"Dec(ember)?)\s+\d{1,2},\s+\d{4}")
regexp = re.compile(pattern)
dates = []
#list of websites that either have no current RFPS, or are broken - Can't check the checked/not working field because not all are updated so program will break
user_agent = 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.7) Gecko/2009021910 Firefox/3.0.7'
for each in dictionary.keys():
if each == "City":
continue
if each is not None and dictionary[each] is not None and each not in check:
user_agent = 'Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.9.0.7) Gecko/2009021910 Firefox/3.0.7'
url = dictionary[each]
headers = {'User-Agent': user_agent, }
request = urllib.request.Request(url, None, headers) # The assembled request
response = urllib.request.urlopen(request)
html = response.read()
soup = BeautifulSoup(html)
tables = soup.find_all('table')
final_dates = []
for table in tables:
# do your stuff for every table
try:
df = pd.read_html(str(table))
if len(df) == 0:
continue
else:
#convert table from website into string paragraphs
a = tabulate(df[0], headers='keys', tablefmt='psql')
# run through keywords
for key in keywords:
if key in a:
#print("EACH IS IN KEY: ", each, key)
if each not in keywords_track:
keywords_track[each] = [key]
else:
num_occ = a.count(key)
if not len(keywords_track[each]) == num_occ:
for i in range(num_occ - 1):
keywords_track[each].append(key)
if regexp.search(a):
print("FOUND DATE!")
dates.append((each, re.findall(r"[\d]{1,2}[/.-][\d]{1,2}[/.-][\d]{4}", a), dictionary[each], a))
except:
continue
print("KEY WORD DICT AFTER FILLING: ", keywords_track)
array = build_dates(dates)
print("Array", array)
email_msg = build_email_msg(array, msg, keywords_track)
return email_msg
#Visit the Mars Weather twitter account here and scrape the latest Mars weather tweet from the page.
#Save the tweet text for the weather report as a variable called mars_weather.
url = 'https://twitter.com/marswxreport?lang=en'
browser.visit(url)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
mars_weather_element = soup.find('div', class_="content")
mars_weather = print(mars_weather_element.p.text)
#Visit the Mars Facts webpage here and use Pandas to scrape the table containing facts about the planet including Diameter, Mass, etc.
#Use Pandas to convert the data to a HTML table string.
url = 'https://space-facts.com/mars/'
tables = pd.read_html(url)
tables
#Mars - Earth Comparison
mars_earth_comparison = tables[0]
mars_earth_comparison
#Mars Planet Profile
mars_planet_profile = tables[1].rename(columns={0:"Mars Category", 1:"Value"})
mars_planet_profile
#Visit the USGS Astrogeology site here to obtain high resolution images for each of Mar's hemispheres.
url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(url)
html = browser.html
soup = BeautifulSoup(html, 'html.parser')
AIRTABLE_API_KEY = 'keybBQGNdYeJkRwcs'
base_key = 'appmCQ7CzGefKPdmu'
url = 'http://www.espn.com/golf/leaderboard?tournamentId=' + str(tournament_id)
table_name = 'PGA_Field'
entries_table = 'PGA2019Entries'
pd.options.mode.chained_assignment = None
page = urlopen(url)
soup = BeautifulSoup(page, "html.parser")
html = soup.find(
"table",
attrs={
"class": "Table2__table-scroller Table2__right-aligned Table2__table"
})
table = pd.read_html(html.prettify())
df = table[0]
df['PLAYER'] = df['PLAYER'].str.replace("'", "")
df = df.set_index("PLAYER")
df.to_csv('espnfield.csv')
airtable = Airtable(base_key, table_name, AIRTABLE_API_KEY)
field = airtable.get_all()
field = pd.DataFrame.from_dict(field)
field_data = [0] * len(field)
for (i, entry) in enumerate(field_data):
entry = field.loc[i]['fields']
field_data[i] = entry
field_data = pd.DataFrame.from_dict(field_data)
def scrape():
browser = init_browser()
client = pymongo.MongoClient("mongodb://localhost:27017")
db = client.mars_db
# Retrieving news title and teaser
browser.visit("https://mars.nasa.gov/news/")
time.sleep(2)
soup = bs(browser.html, "html.parser")
items = soup.find("ul", class_="item_list")
slides = items.find_all("li", class_="slide")
news_titles = []
news_paragraphs = []
for slide in slides:
news_title = slide.find("div", class_="content_title").text
news_p = slide.find("div", class_="article_teaser_body").text
news_titles.append(news_title)
news_paragraphs.append(news_p)
# Retrieving featured image url
browser.visit(
"https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars")
browser.find_by_id("full_image").click()
time.sleep(2)
soup = bs(browser.html, "html.parser")
image_src = soup.find("img", class_="fancybox-image")["src"]
featured_image_url = f"https://jpl.nasa.gov{image_src}"
# Retriving mars facts table
browser.visit("https://space-facts.com/mars/")
df = pd.read_html(browser.html)[1]
mars_facts_table_html = df.to_html(index=False, justify="center")
mars_facts_table_html = mars_facts_table_html.replace("\n", "")
browser.visit(
"https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars"
)
time.sleep(1)
soup = bs(browser.html, "html.parser")
# Retrieving hemishere page's urls
hemisphere_urls = []
hemispheres = soup.find_all("div", class_="description")
for hemisphere in hemispheres:
url = hemisphere.find("a")["href"]
url = f"https://astrogeology.usgs.gov{url}"
hemisphere_urls.append(url)
# Retrieving titles and image links of different hemispheres
hemisphere_list = []
for hemisphere_url in hemisphere_urls:
browser.visit(hemisphere_url)
time.sleep(2)
soup = bs(browser.html, "html.parser")
title = soup.find("h2", class_="title").text
title = re.sub(" Enhanced", "", title)
image_url = soup.find_all("li")[0].find("a")["href"]
hemisphere_list.append({"title": title, "image_url": image_url})
return_dict = {}
return_dict["news_titles"] = news_titles
return_dict["news_paragraphs"] = news_paragraphs
return_dict["featured_image_url"] = featured_image_url
return_dict["mars_facts_table_html"] = mars_facts_table_html
return_dict["hemisphere_list"] = hemisphere_list
return_dict["date"] = datetime.datetime.utcnow()
db.mission_to_mars.update({}, return_dict, upsert=True)
browser.quit()
return return_dict
def scrape():
""" Scrapes all websites for Mars data """
# Create a python dictionary to store all data
scrape_mars_dict = {}
# Use requests and BeautifulSoup to scrape Nasa News for latest news
url = 'https://mars.nasa.gov/news/?page=0&per_page=40&order=publish_date+desc%2Ccreated_at+desc&search=&category=19%2C165%2C184%2C204&blank_scope=Latest'
response = requests.get(url)
soup = bs(response.text, 'lxml')
results = soup.find('div', class_='features')
news_title = results.find('div', class_='content_title').text
newsp = results.find('div', class_='rollover_description').text
# Store scraped data into dictionary
scrape_mars_dict['news_title'] = news_title
scrape_mars_dict['newsp'] = newsp
# Scrape Mars Weather twitter for latest weather report
twitter_url = 'https://twitter.com/marswxreport?lang=en'
twitter_response = requests.get(twitter_url)
twitter_soup = bs(twitter_response.text, 'lxml')
twitter_result = twitter_soup.find('div', class_='js-tweet-text-container')
mars_weather = twitter_result.find('p', class_='js-tweet-text').text
# Store scraped data into dictionary
scrape_mars_dict['mars_weather'] = mars_weather
# Scrape facts about Mars from space-facts.com using Pandas read_html function
mars_facts_url = 'https://space-facts.com/mars/'
tables = pd.read_html(mars_facts_url)
df = tables[0]
# Cleanup the Index
df.rename({0:"Mars - Earth Comparison", 1:"Mars", 2: "Earth"}, axis=1, inplace=True)
df.set_index("Mars - Earth Comparison", inplace=True)
# Export scraped table into an html script
mars_facts = df.to_html()
mars_facts.replace("\n","")
df.to_html('mars_facts.html')
# Store html file to dictionary
scrape_mars_dict['mars_facts'] = mars_facts
# Call on chromedriver function to use for splinter
browser = init_browser()
# Scrape Nasa for url of latest featured image of Mars
nasa_url = 'https://www.jpl.nasa.gov/spaceimages/?search=&category=Mars'
browser.visit(nasa_url)
nasa_html = browser.html
nasa_soup = bs(nasa_html, "lxml")
featured_image = nasa_soup.find('div', class_='default floating_text_area ms-layer').find('footer')
featured_image_url = 'https://www.jpl.nasa.gov'+ featured_image.find('a')['data-fancybox-href']
# Store url to dictionary
scrape_mars_dict['featured_image_url'] = featured_image_url
# Scrape astrogeology.usgs.gov for urls of hemisphere images of Mars
hemisphere_url = 'https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars'
browser.visit(hemisphere_url)
hemisphere_html = browser.html
hemisphere_soup = bs(hemisphere_html, 'lxml')
base_url ="https://astrogeology.usgs.gov"
image_list = hemisphere_soup.find_all('div', class_='item')
# Create a list to store dictionary of urls and image titles
hemisphere_image_urls = []
# Loop through list of hemispheres and click on each one to find large resolution image
for image in image_list:
# Create a dicitonary to store urls and titles
hemisphere_dict = {}
# Find link to large image
href = image.find('a', class_='itemLink product-item')
link = base_url + href['href']
# Visit the link
browser.visit(link)
# Wait 1 second
time.sleep(1)
# Parse the html of the new page
hemisphere_html2 = browser.html
hemisphere_soup2 = bs(hemisphere_html2, 'lxml')
# Find the title
img_title = hemisphere_soup2.find('div', class_='content').find('h2', class_='title').text
# Append to dict
hemisphere_dict['title'] = img_title
# Find image url
img_url = hemisphere_soup2.find('div', class_='downloads').find('a')['href']
# Append to dict
hemisphere_dict['url_img'] = img_url
# Append dict to list
hemisphere_image_urls.append(hemisphere_dict)
# Store hemisphere image urls to dictionary
scrape_mars_dict['hemisphere_image_urls'] = hemisphere_image_urls
return scrape_mars_dict
# https://splinter.readthedocs.io/en/latest/drivers/chrome.html
# BeautifulSoup, Pandas, and Requests/Splinter
from bs4 import BeautifulSoup
from selenium import webdriver
import pandas as pd
# In[2]:
# EIA browser to grab table
url = 'https://www.eia.gov/electricity/state/archive/2014/'
# In[3]:
# Grab table with pandas
table_list = pd.read_html(url)
table = table_list[0]
table.head()
# In[4]:
# Remove total row from list
remove_list = ['U.S. Total']
states_table = table[~table.Name.isin(remove_list)]
# remove non-price columns
ecost_df = states_table.iloc[:, [0, 1]]
# push to csv
ecost_df.to_csv('csv/eia_2014_scrape.csv')
def scrape():
browser = init_browser()
listings = {}
url = "https://mars.nasa.gov/news/"
browser.visit(url)
time.sleep(1)
html = browser.html
soup = BeautifulSoup(html, "html.parser")
news_title = soup.find('div', class_='content_title').text
news_p = soup.find('div', class_='article_teaser_body').text
url_base = "https://www.jpl.nasa.gov"
url_add = '/spaceimages/?search=&category=Mars'
browser.visit(url_base + url_add)
time.sleep(1)
html = browser.html
soup = BeautifulSoup(html, "html.parser")
bttn_image_url = soup.find('article', class_='carousel_item').get('style')
start = bttn_image_url.find("url('")
end = bttn_image_url.find("');")
featured_image_url = url_base + bttn_image_url[start + 3 + len("('"):end]
url_base = "https://twitter.com/marswxreport?lang=en"
browser.visit(url_base)
time.sleep(1)
html = browser.html
soup = BeautifulSoup(html, "html.parser")
mars_weather = soup.find('p', class_='TweetTextSize').text
url_base = "https://space-facts.com/mars/"
browser.visit(url_base)
time.sleep(1)
html = browser.html
soup = BeautifulSoup(html, "html.parser")
table = pd.read_html(url_base)
htmltable = table[0].to_html()
link2 = []
link3 = []
link4 = []
url_base = "https://astrogeology.usgs.gov/search/results?q=hemisphere+enhanced&k1=target&v1=Mars"
browser.visit(url_base)
time.sleep(2)
links = browser.find_link_by_partial_text('Hemisphere')
[link2.append(link['href']) for link in links]
for link in link2:
browser.visit(link)
time.sleep(2)
url_link = browser.find_link_by_partial_text('Sample')
title_text = browser.find_by_css('.title')
link3.append(url_link['href'])
link4.append(title_text.html)
hemisphere_image_urls = []
for i in range(len(link3)):
hemisphere_image_urls.append({"title": link4[i], "img_url": link3[i]})
listings["news_p"] = news_p
listings["news_title"] = news_title
listings["featured_image_url"] = featured_image_url
listings["mars_weather"] = mars_weather
listings["html_table"] = htmltable
listings["hemisphere_img_dict"] = hemisphere_image_urls
return listings
except:
## proxies that do not work are removed from the list
print(f"{pick} did not work")
proxies_list.remove(pick)
print(f"{pick} removed")
print(len(proxies_list))
print(Exception)
else: ## if proxies_list is empty, we get our proxies without configuring urllib for using proxies
req = urllib.request.Request(url, headers={'User-Agent': "Magic Browser"})
sauce = urllib.request.urlopen(req).read()
soup = bs.BeautifulSoup(sauce, 'lxml')
print(soup)
## use pandas to get tables and choose columns
df = pd.read_html(sauce) ## using pandas read_html method to parse through url
print(df, len(df))
df = df[0] ## df is a list of tables. We want first table.
df.to_csv("proxiesraw.csv", index=True) ## saving df to csv
## print(df[0].columns) ## choosing dataframe with the index 0 and checking all the columns. You may need to check columns if name of column have weird spacing and etc.
## df = df[0] ## setting df as the df[0] the data frame with the ip, port, and etc.
df = pd.read_csv("proxiesraw.csv")
df = df[['IP Address', "Port",
"Https"]] ## making df only show the columns we want to see.
df = df.dropna(axis=0, how='any', thresh=None, subset=None,
inplace=False) ## dropping all rows with missing values
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from scipy.optimize import minimize
import math
url = 'http://www.pangoo.it/coronavirus/?t=region&r=Lombardia&data=y#table'
data = pd.read_html(url)[0]
# data = pd.read_csv('cov19_data.csv')
count = data['Totale casi'][:-1]
n = len(count)
temp = np.array([])
for i in range(n):
temp = np.append(temp, int(count[i]))
y = np.diff(temp)
time = np.linspace(1, n-1, n-1)
plt.figure(figsize=[10, 5])
plt.plot(time, y)
plt.xlabel('Index')
plt.ylabel('y')
plt.title('New Covid19 cases in Italy')
plt.show()
obs = len(y)-5
x = y[:obs]
a, p, k, v = [np.ones(n) for _ in range(4)]
a[0], p[0], v[0] = x[1], 10000, 0
def getWorksheetsListFromExcelURL(downloadURL, isXLSX):
print('downloading file : ' + downloadURL)
# handles well formatted .xlsx files with dat reader bitch
if isXLSX:
urllib.request.urlretrieve(downloadURL, "temp.xlsx")
xlsxFile = pd.ExcelFile("temp.xlsx")
worksheets = []
print('file downloaded, tranforming into sheets')
for xlsxSheet in xlsxFile.sheet_names:
worksheets.append(pd.read_excel(xlsxFile, xlsxSheet))
print('file transformed, cleaning up')
os.remove("temp.xlsx")
# handles .xls schema which is stored string like - parses that bitch
else:
urllib.request.urlretrieve(downloadURL, "temp.xls")
file1 = open('temp.xls', 'r')
lines = file1.readlines()
worksheets = []
worksheet = []
isWorksheet = False
isFirstWorkSheet = True
count = 0
print('file downloaded, tranforming into sheets')
for line in lines:
if '<html' in line:
isWorksheet = True
if '</html' in line:
isWorksheet = False
if isWorksheet:
worksheet.append(line)
else:
if len(worksheet) > 0:
worksheet.append(line)
if not isFirstWorkSheet:
temp = '\n'.join(worksheet)
temp = temp.replace('3D1', '1')
temp = temp.replace('3D2', '2')
temp = temp.replace('3D3', '3')
temp = temp.replace('3D4', '4')
temp = temp.replace('3D5', '5')
temp = temp.replace('3D6', '6')
temp = temp.replace('3D7', '7')
temp = temp.replace('3D8', '8')
temp = temp.replace('3D9', '9')
temp = temp.replace('3D10', '10')
temp = pd.read_html(temp)
temp = formatDF(temp, count)
count += 1
worksheets.append(temp)
worksheet = []
else:
worksheet = []
isFirstWorkSheet = False
else:
worksheet = []
print('file transformed, cleaning up')
os.remove("temp.xls")
return worksheets
'ultrasound nerve segmentation',
'pneumothorax',
'2021 prostate',
'2021 pneumonia',
'2021 intracrancial, INVERTED METRIC', #NOTE: performance metric is inverted
'2021 covid19',
'2021 chest xray'
]
# Load the data
data = dict()
interesting_columns = ['Team Name', 'Score', 'Entries']
for i, name in enumerate(names):
public = pd.read_html('kaggle/' + name +
'_public.html')[0][interesting_columns]
private = pd.read_html('kaggle/' + name +
'_private.html')[0][interesting_columns]
# Select teams who did two or more submissions (to avoid people who
# didn't really participate
public = public.query('Entries >= 2')
private = private.query('Entries >= 2')
print(public.head())
# Merge the two
public = public.drop(columns='Entries').rename(columns=dict(
Score='public'))
private = private.drop(columns='Entries').rename(columns=dict(
Score='private'))
scores = pd.merge(public, private)