X-Git-Url: https://git.ucc.asn.au/?a=blobdiff_plain;f=icalparse.py;h=59e409b4325e65046437eab6cd771868c742ae92;hb=ee8a5360ac1f00ccbeef97c53d8615f6615108bf;hp=6cb7c39f55bead4fd73f7579f169b6d7cd5c5232;hpb=cec1435fe4a612d02087b715617f24fd21904bdd;p=frenchie%2Ficalparse.git diff --git a/icalparse.py b/icalparse.py index 6cb7c39..59e409b 100755 --- a/icalparse.py +++ b/icalparse.py @@ -1,6 +1,6 @@ -#!/usr/bin/python +#!/usr/bin/python3 # -# Copyright (c) 2010 James French +# Copyright (c) 2013 James French # # Permission is hereby granted, free of charge, to any person obtaining a copy # of this software and associated documentation files (the "Software"), to deal @@ -20,86 +20,247 @@ # OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN # THE SOFTWARE. -import sys -import re -from optparse import OptionParser - -class InvalidICS(Exception): pass -class notJoined(Exception): pass - -# RFC5545 and RFC5546 iCalendar registries contain upper case letters -# and dashes only and are separated from the value by a colon (:) -icalEntry = re.compile('^[A-Z\-]+:.*') - -def lineJoiner(oldcal): - '''Unfolds a calendar so that items can be parsed''' - - cal = [] - - # Strip newlines ( - for line in oldcal: - line = line.rstrip('\r\n') - - # Reassemble broken Lines - if not line: - if not cal: continue - else: cal[-1] += '\\n' - elif line[0] == ' ': - if not cal: raise InvalidICS, 'First line of ICS must be element' - line = line[1:len(line)] - cal[-1] += line - elif not icalEntry.match(line): - if not cal: raise InvalidICS, 'First line of ICS must be element' - cal[-1] += '\\n' + line - else: - if cal: cal[-1] += '\r\n' - cal.append(line) +import sys, os +from urllib import (request, error, parse) +import vobject +from cgi import parse_header - cal[-1] += '\r\n' +def getContent(url='',stdin=False): + '''Generic content retriever, DO NOT use this function in a CGI script as + it can read from the local disk (which you probably don't want it to). + ''' - return cal + encoding = '' # If we don't populate this, the script will assume UTF-8 + + # Special case, if this is a HTTP url, return the data from it using + # the HTTP functions which attempt to play a bit nicer. + parsedURL = parse.urlparse(url) + if 'http' in parsedURL[0]: return getHTTPContent(url) + + if stdin: + content = sys.stdin.read() + return (content, encoding) + + if not parsedURL[0]: url = 'file://' + os.path.abspath(url) + + # If we've survived, use python's generic URL opening library to handle it + try: + res = request.urlopen(url) + content = res.read() + ct = res.info().getplist() + res.close() + except (error.URLError, OSError) as e: + sys.stderr.write('%s\n'%e) + sys.exit(1) + + for param in ct: + if 'charset' in param: + encoding = param.split('=')[1] + break + + return (content, encoding) -def lineSplitter(oldcal, length=75): - '''Folds content lines to a specified length, returns a list''' - cal = [] - sl = length - 1 +def getHTTPContent(url=''): + '''This function attempts to play nice when retrieving content from HTTP + services. It's what you should use in a CGI script.''' - for line in oldcal: - # Line & line ending line ending fit inside length, do nothing - if len(line.rstrip()) <= length: - cal.append(line) + try: + import httplib2 + except ImportError: + pass + + if not url: return ('','') + + if not 'http' in parse.urlparse(url)[0]: return ('','') + + if 'httplib2' in sys.modules: + try: h = httplib2.Http() + except OSError: h = httplib2.Http() + else: h = False + + if h: + try: + req = h.request(url) + except ValueError as e: + sys.stderr.write('%s\n'%e) + sys.exit(1) + + resp, content = req + if 'content-type' in resp: + ct = 'Content-Type: %s'%req[0]['content-type'] + ct = parse_header(ct) + if 'charset' in ct[1]: encoding = ct[1]['charset'] + else: encoding = '' else: - brokenline = [line[0:length] + '\r\n'] - ll = length - while ll < len(line.rstrip('\r\n')) + 1: - brokenline.append(' ' + line[ll:sl+ll].rstrip('\r\n') + '\r\n') - ll += sl - cal += brokenline + ct = '' + encoding = '' + + else: + try: + req = request.urlopen(url) + except error.URLError as e: + sys.stderr.write('%s\n'%e) + sys.exit(1) + + content = req.read() + ct = req.info().getplist() + for param in ct: + if 'charset' in param: + encoding = param.split('=')[1] + break + + return (content, encoding) + + +def generateRules(ruleConfig): + '''Attempts to load a series of rules into a list''' + try: + import parserrules + except ImportError: + return [] + + for conf in ruleConfig: + parserrules.ruleConfig[conf] = ruleConfig[conf] + + rules = [getattr(parserrules, rule) for rule in dir(parserrules) if callable(getattr(parserrules, rule))] + + return rules + + +def applyRules(cal, rules=[], verbose=False): + 'Runs a series of rules on the lines in ical and mangles its output' + + for rule in rules: + cal = rule(cal) return cal -if __name__ == '__main__': - # If the user passed us a 'stdin' argument, we'll go with that, - # otherwise we'll try for a url opener - parser = OptionParser() +def writeOutput(cal, outfile=''): + '''Takes a list of lines and outputs to the specified file''' + + if not cal: + sys.stderr.write('Refusing to write out an empty file') + sys.exit(0) + + if not outfile: + out = sys.stdout + else: + try: + out = open(outfile, 'w') + except (IOError, OSError) as e: + sys.stderr.write('%s\n'%e) + sys.exit(1) + + cal.serialize(out) + + if not out == sys.stdout: + out.close() + + +def runLocal(): + '''Main run function if this script is called locally''' + + from optparse import OptionParser + + parser = OptionParser('usage: %prog [options] url') parser.add_option('-s', '--stdin', action='store_true', dest='stdin', default=False, help='Take a calendar from standard input') + parser.add_option('-v', '--verbose', action='store_true', dest='verbose', + default=False, help='Be verbose when rules are being applied') parser.add_option('-o', '--output', dest='outfile', default='', help='Specify output file (defaults to standard output)') + parser.add_option('-m','--encoding', dest='encoding', default='', + help='Specify a different character encoding' + '(ignored if the remote server also specifies one)') + parser.add_option('-t','--timezone', dest='timezone', default='', + help='Specify a timezone to use if the remote calendar doesn\'t set it properly') (options, args) = parser.parse_args() + ruleConfig["defaultTZ"] = options.timezone or ruleConfig["defaultTZ"] + + # If the user passed us a 'stdin' argument, we'll go with that, + # otherwise we'll try for a url opener + if not args and not options.stdin: + parser.print_usage() + sys.exit(0) + elif not options.stdin: + url = args[0] + else: + url = '' + + (content, encoding) = getContent(url, options.stdin) + encoding = encoding or options.encoding or 'utf-8' + + cal = vobject.readOne(str(content, encoding)) + cal = applyRules(cal, generateRules(ruleConfig), options.verbose) + + writeOutput(cal, options.outfile) - if not options.stdin: + +def exitQuiet(exitstate=0): + '''When called as a CGI script, exit quietly if theres any errors''' + print('Content-Type: text/html\n') + sys.exit(exitstate) + + +def runCGI(): + '''Main run function if this script is called as a CGI script + to process facebook ical files''' + import cgi + import re + import cgitb; cgitb.enable() + + ruleConfig["facebook"] = True + + form = cgi.FieldStorage() + if "uid" not in form or "key" not in form: + print('Content-Type: text/calendar\n') + sys.exit(0) + try: + # UID should be numeric, if it's not we have someone playing games + uid = int(form['uid'].value) + except: + exitQuiet() + + # The user's key will be a 16 character string + key = form['key'].value + re.search('[&?]+', key) and exitQuiet() + len(key) == 16 or exitQuiet() + + # Historically facebook has been notoriously bad at setting timzeones + # in their stuff so this should be a user setting. If it is set in + # their calendar it'll be used otherwise if the user feeds crap or + # nothing just assume they want Australia/Perth + tz = "" + if "tz" in form: + from pytz import timezone try: - import httplib2 - urllib = False - except ImportError: - try: - import urllib - urllib = True - except ImportError: - sys.stderr.write('Failed to find a suitable http downloader\n') - raise + timezone(form['tz'].value) + tz = form['tz'].value + except: pass + + ruleConfig["defaultTZ"] = tz or ruleConfig["defaultTZ"] + + # Okay, we're happy that the input is sane, lets serve up some data + url = 'https://www.facebook.com/events/ical/upcoming/?uid=%s&key=%s'%(uid,key) + (content, encoding) = getHTTPContent(url) + + cal = vobject.readOne(str(content, encoding)) + cal = applyRules(cal, generateRules(ruleConfig), False) + + print(('Content-Type: text/calendar; charset=%s\n'%encoding)) + writeOutput(cal) + + +if __name__ == '__main__': + # Ensure the rules process using the desired timezone + ruleConfig = {} + ruleConfig["defaultTZ"] = 'Australia/Perth' + # Detect if this script has been called by CGI and proceed accordingly + if 'REQUEST_METHOD' in os.environ: + runCGI() + else: + runLocal()