gistfile1.py

import requests
import re
from bs4 import BeautifulSoup

contents = requests.get("http://www.nhl.com/scores/htmlreports/20022003/PL020413.HTM").text

def NotEmpty(str):
	return not re.match("^\s*$",str)

def SplitOnColumns(unsplit,positions):
	result = []
	for column in positions:
		result.append(unsplit[column[0]:column[1]+1].strip())
	return result

soup = BeautifulSoup(contents)

table = soup.find("pre").text

# Get all nonempty rows
rows = filter(NotEmpty,table.split("\r\n"))

# For tracking col positions
columns = []

# Find the dashes row
for row in rows:
	# If it's all spaces/dashes with at least one dash
	if re.match("^[\s-]+$",row) and not re.match("^\s+$",row):
		# Loop through characters, find the start and end position of each column
		for i,char in enumerate(row):
			if char == "-":
				if i == 0 or row[i-1] != "-":
					start = i
				if i == len(row)-1 or row[i+1] != "-":
					end = i
					columns.append([start,end])
		break

# Get an array of headers
headers = SplitOnColumns(rows[0],columns)

as_json = []

# For all rows besides the first two
for row in rows[2:]:

	# Empty object
	objectified = {}

	# For each column, set objectified[header] to that value
	for i,value in enumerate(SplitOnColumns(row,columns)):
		objectified[headers[i]] = value

	as_json.append(objectified)

print as_json