bs4_demo.py

""" Obtain a quote of the Dow Jones Industrial Average from Bloomberg.

"""

import json

from bs4 import BeautifulSoup
import requests

url = 'http://www.bloomberg.com/quote/INDU:IND'

# Get the entire Bloomberg web page from the web.
resp = requests.get(url)
data = resp.text

# Save a copy of the HTML (possibly to cache subsequent requests).
with open('INDU_IND.html', 'a') as f:
    f.write(data.encode('utf-8'))

soup = BeautifulSoup(data)

# Find quote information by CSS class attribute.
quotes = soup.find_all('div', class_='schema-org-financial-quote')
quote = quotes[0]

# Extract the meta tags from inside the 'schema-org-financial-quote' tag. These
# tags hold the actual financial info; each meta tag has attributes
# 'itemprop' and 'content', describing the category (e.g. 'tickerSymbol') and
# the corresponding value (e.g. 'INDU').
quote_info = {
    meta['itemprop']: meta['content'] for meta in quote.find_all('meta')
}

# Print out a report of the data obtained so far.
print 'Quote Summary'
print '============='
for attribute, value in quote_info.items():
    print '{:<20}: {}'.format(attribute, value)

# Encode the quote info to JSON and save for further processing.
with open('INDU_IND.json', 'a') as f:
    json.dump(quote_info, f)

csv_demo.py

""" Process historical stock records in CSV.

Source: https://uk.finance.yahoo.com/q/hp?s=AAPL

"""

# This script assumes that a set of stock records has been downloaded to the
# current folder as table.csv .

import csv

with open('table.csv', 'rb') as csvfile:
    csvreader = csv.reader(csvfile)
    header = csvreader.next()
    rows = list(csvreader)

with open('table.csv', 'rb') as csvfile:
    csvreader = csv.DictReader(csvfile)
    stockdata = list(csvreader)

# Loop over the data, and add a data member indicating how much the stock
# fluctuated over one day.

processed_data = []
for quote in stockdata:
    quote_hi = float(quote['High'])
    quote_lo = float(quote['Low'])
    quote['Rel Difference'] = (quote_hi - quote_lo) / quote_lo

# Write processed data back to file.
fieldnames = header + ['Rel Difference']
with open('table-processed.csv', 'wb') as csvfile:
    csvwriter = csv.DictWriter(csvfile, fieldnames)
    csvwriter.writeheader()
    csvwriter.writerows(stockdata)

h5_demo.py

import tables


class StockRecord(tables.IsDescription):
    date = tables.StringCol(16)
    high = tables.Float64Col()
    low = tables.Float64Col()


h5file = tables.open_file("stocks.h5", mode="w", title="Stock data")
group = h5file.create_group("/", 'stocks', 'Stock info')

for name in ['AAPL', 'GOOG', 'FB']:
    table = h5file.create_table(group, name, StockRecord, "Simple example")

    stockrecord = table.row
    for i in range(1, 31):
        stockrecord['date'] = '2015-1-{}'.format(i)
        stockrecord['high'] = 100.0 + i
        stockrecord['low'] = 99.0 + i
        stockrecord.append()

    table.flush()

h5file.close()

xml_parsing.py

import xml.etree.cElementTree as etree  # Just ElementTree in Python 3

# curl -O http://www.ibiblio.org/xml/examples/shakespeare/hamlet.xml

tree = etree.parse('hamlet.xml')
root = tree.getroot()
print root

# Print all characters in the play.
for persona_tag in root.iter('PERSONA'):
    print persona_tag.text

for persona_tag in root.iter('PERSONA'):
    if persona_tag.text.startswith('CLAUDIUS'):
        persona_tag.text = 'CLAUDIUS (he did it)'
        persona_tag.set('bad', 'yes')

tree.write('hamlet_modified.xml')