X-Git-Url: https://git.ucc.asn.au/?a=blobdiff_plain;f=icalparse.py;h=f00b400d889dd17b50fb2216e8738d59310f6462;hb=d6a7b9600696f577ea2c6f49e99dc6f01b4f76d8;hp=5010d1d417ae54833860c8737d6b4a70f810f83c;hpb=ab19510e73ff3048acb58aa648b86c58c4ea9882;p=frenchie%2Ficalparse.git diff --git a/icalparse.py b/icalparse.py index 5010d1d..f00b400 100755 --- a/icalparse.py +++ b/icalparse.py @@ -1,6 +1,6 @@ #!/usr/bin/python # -# Copyright (c) 2010 James French +# Copyright (c) 2011 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 @@ -21,51 +21,140 @@ # THE SOFTWARE. import sys -import re import urlparse import os +import vobject +from cgi import parse_header -class InvalidICS(Exception): pass -class notJoined(Exception): pass -icalEntry = re.compile('^[A-Z\-]+:.*') +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). + ''' + + 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 = urlparse.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 + import urllib2 + try: + res = urllib2.urlopen(url) + content = res.read() + ct = res.info().getplist() + res.close() + except (urllib2.URLError, OSError), e: + sys.stderr.write('%s\n'%e) + sys.exit(1) -def lineJoiner(oldcal): - '''Takes a string containing a calendar and returns an array of its lines''' + for param in ct: + if 'charset' in param: + encoding = param.split('=')[1] + break - if list(oldcal) == oldcal: - oldcal = '\r\n'.join(oldcal) + return (content, encoding) - oldcal.replace('\r\n ', '') - return oldcal.split('\r\n') +def getHTTPContent(url='',cache='.httplib2-cache'): + '''This function attempts to play nice when retrieving content from HTTP + services. It's what you should use in a CGI script.''' -def lineFolder(oldcal, length=75): - '''Folds content lines to a specified length, returns a list''' + try: + import httplib2 + except ImportError: + import urllib2 - if length > 75: - sys.stderr.write('WARN: lines > 75 octets are not RFC compliant\n') + if not url: return ('','') - cal = [] - sl = length - 1 + if not 'http' in urlparse.urlparse(url)[0]: return ('','') - for line in oldcal: - # Line fits inside length, do nothing - if len(line.rstrip()) <= length: - cal.append(line) + if 'httplib2' in sys.modules: + try: h = httplib2.Http('.httplib2-cache') + except OSError: h = httplib2.Http() + else: h = False + + if h: + try: + req = h.request(url) + except ValueError, 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 = urllib2.urlopen(url) + except urllib2.URLError, 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(): + '''Attempts to load a series of rules into a list''' + try: + import parserrules + except ImportError: + return [] + + 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 -def getContent(url='',stdin=False): - pass +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), e: + sys.stderr.write('%s\n'%e) + sys.exit(1) + + cal.serialize(out) + if not out == sys.stdout: + out.close() if __name__ == '__main__': from optparse import OptionParser @@ -75,8 +164,13 @@ if __name__ == '__main__': 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)') (options, args) = parser.parse_args() @@ -88,43 +182,10 @@ if __name__ == '__main__': else: url = '' - # Work out what url parsers we're going to need based on what the user - # gave us on the command line - we do like files after all - parsedURL = urlparse.urlparse(url) - http = 'http' in parsedURL[0] + (content, encoding) = getContent(url, options.stdin) + encoding = encoding or options.encoding or 'utf-8' - if not parsedURL[0]: u = False - else: u = True + cal = vobject.readOne(unicode(content, encoding)) + cal = applyRules(cal, generateRules(), options.verbose) - if not options.stdin and http: - try: - import httplib2 - except ImportError: - import urllib2 - - # Try and play nice with HTTP servers unless something goes wrong. We don't - # really care about this cache (A lot of ics files seem to be generated with - # php which hates caching with a passion). - h = False - if 'httplib2' in sys.modules: - try: h = httplib2.Http('.httplib2-cache') - except OSError: h = httplib2.Http() - - # Load urllib2 if this is not a stdin - if not options.stdin and (not http or not 'httplib2' in sys.modules): - import urllib2 - - try: - content = u and (h and h.request(url)[1] or urllib2.urlopen(url).read()) - except (ValueError, urllib2.URLError), e: - sys.stderr.write('%s\n'%e) - sys.exit(1) - - if not u and not options.stdin: - try: content = open(os.path.abspath(url),'r').read() - except (IOError, OSError), e: - sys.stderr.write('%s\n'%e) - sys.exit(1) - - if options.stdin: - content = sys.stdin.read() + writeOutput(cal, options.outfile)