-
-
Save ChristopherDaigle/605c305ca00eeed8f6028d64f337aefa to your computer and use it in GitHub Desktop.
Python 3 code to extract stock market data from yahoo finance
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
from lxml import html | |
import requests | |
from time import sleep | |
import json | |
import argparse | |
from collections import OrderedDict | |
from time import sleep | |
def parse(ticker): | |
url = "http://finance.yahoo.com/quote/%s?p=%s"%(ticker,ticker) | |
response = requests.get(url, verify=False) | |
print ("Parsing %s"%(url)) | |
sleep(4) | |
parser = html.fromstring(response.text) | |
summary_table = parser.xpath('//div[contains(@data-test,"summary-table")]//tr') | |
summary_data = OrderedDict() | |
other_details_json_link = "https://query2.finance.yahoo.com/v10/finance/quoteSummary/{0}?formatted=true&lang=en-US®ion=US&modules=summaryProfile%2CfinancialData%2CrecommendationTrend%2CupgradeDowngradeHistory%2Cearnings%2CdefaultKeyStatistics%2CcalendarEvents&corsDomain=finance.yahoo.com".format(ticker) | |
summary_json_response = requests.get(other_details_json_link) | |
try: | |
json_loaded_summary = json.loads(summary_json_response.text) | |
y_Target_Est = json_loaded_summary["quoteSummary"]["result"][0]["financialData"]["targetMeanPrice"]['raw'] | |
earnings_list = json_loaded_summary["quoteSummary"]["result"][0]["calendarEvents"]['earnings'] | |
eps = json_loaded_summary["quoteSummary"]["result"][0]["defaultKeyStatistics"]["trailingEps"]['raw'] | |
datelist = [] | |
for i in earnings_list['earningsDate']: | |
datelist.append(i['fmt']) | |
earnings_date = ' to '.join(datelist) | |
for table_data in summary_table: | |
raw_table_key = table_data.xpath('.//td[contains(@class,"C(black)")]//text()') | |
raw_table_value = table_data.xpath('.//td[contains(@class,"Ta(end)")]//text()') | |
table_key = ''.join(raw_table_key).strip() | |
table_value = ''.join(raw_table_value).strip() | |
summary_data.update({table_key:table_value}) | |
summary_data.update({'1y Target Est':y_Target_Est,'EPS (TTM)':eps,'Earnings Date':earnings_date,'ticker':ticker,'url':url}) | |
return summary_data | |
except: | |
print ("Failed to parse json response") | |
return {"error":"Failed to parse json response"} | |
if __name__=="__main__": | |
argparser = argparse.ArgumentParser() | |
argparser.add_argument('ticker',help = '') | |
args = argparser.parse_args() | |
ticker = args.ticker | |
print ("Fetching data for %s"%(ticker)) | |
scraped_data = parse(ticker) | |
print ("Writing data to output file") | |
with open('%s-summary.json'%(ticker),'w') as fp: | |
json.dump(scraped_data,fp,indent = 4) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment