X-Git-Url: https://git.ucc.asn.au/?p=frenchie%2Ficalparse.git;a=blobdiff_plain;f=icalparse.py;h=3907da75547cd90c2ad19a2f4f0865423bcf408d;hp=395283fd427107a521ffa77135e8143aa81477be;hb=6913e5571252b2ebdaec7f21cf139f17cde24119;hpb=c567bf8333d6cb3ab29aabf5ccba0e712c3b2005 diff --git a/icalparse.py b/icalparse.py index 395283f..3907da7 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 @@ -23,89 +23,8 @@ import sys import urlparse import os - - -class InvalidICS(Exception): pass -class IncompleteICS(InvalidICS): pass - -def lineJoiner(oldcal): - '''Takes a string containing a calendar and returns an array of its lines''' - - if not oldcal[0:15] == 'BEGIN:VCALENDAR': - raise InvalidICS, "Does not appear to be a valid ICS file" - - if not 'END:VCALENDAR' in oldcal[-15:-1]: - raise IncompleteICS, "File appears to be incomplete" - - if list(oldcal) == oldcal: - oldcal = '\r\n'.join(oldcal) - - oldcal = oldcal.replace('\r\n ', '').replace('\r\n\t','') - return [unicode(x, 'utf-8') for x in oldcal.strip().split('\r\n')] - - -def lineFolder(oldcal, length=75): - '''Folds content lines to a specified length, returns a list''' - - if length > 75: - sys.stderr.write('WARN: lines > 75 octets are not RFC compliant\n') - - cal = [] - sl = length - 1 - - for uline in oldcal: - line = uline.encode('utf-8') - - # Line fits inside length, do nothing - if len(line) <= length: - cal.append(line) - - else: - ll = length - foldedline = [] - while uline: - # This algorithm prevents slicing multi-byte chars in half - - # Convert up to length octets to unicode, dropping any - # half characters - ufold = unicode(line[0:ll], 'utf-8', 'ignore') - fold = ufold.encode('utf-8') - - # Remove what we've converted from the line - uline = uline.replace(ufold,u'',1) - line = uline.encode('utf-8') - - # And add the fold to the list - foldedline.append(fold) - - # Subsequent lines are shorter as they include a space - ll = length - 1 - - # Finally, add the fold 'marker' to the line - cal.append('\r\n '.join(foldedline)) - - return cal - - -def splitFields(cal): - '''Takes a list of lines in a calendar file and returns a list of tuples - as (key, value) pairs''' - - ical = [tuple(x.split(':',1)) for x in cal] - - # Check that we got 2 items on every line - for line in ical: - if not len(line) == 2: - raise InvalidICS, "Didn't find a content key on: %s"%(line) - - return ical - - -def joinFields(ical): - '''Takes a list of tuples that make up a calendar file and returns it to a - list of lines''' - - return [':'.join(x) for x in ical] +import vobject +from cgi import parse_header def getContent(url='',stdin=False): @@ -113,6 +32,8 @@ def getContent(url='',stdin=False): 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) @@ -120,25 +41,27 @@ def getContent(url='',stdin=False): if stdin: content = sys.stdin.read() - return content + return (content, encoding) - if not parsedURL[0]: - try: content = open(os.path.abspath(url),'r').read() - except (IOError, OSError), e: - sys.stderr.write('%s\n'%e) - sys.exit(1) - return content + 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) - return content + + for param in ct: + if 'charset' in param: + encoding = param.split('=')[1] + break + + return (content, encoding) def getHTTPContent(url='',cache='.httplib2-cache'): @@ -150,73 +73,70 @@ def getHTTPContent(url='',cache='.httplib2-cache'): except ImportError: import urllib2 - if not url: return '' + if not url: return ('','') + + if not 'http' in urlparse.urlparse(url)[0]: return ('','') if 'httplib2' in sys.modules: try: h = httplib2.Http('.httplib2-cache') except OSError: h = httplib2.Http() else: h = False - try: - if h: content = h.request(url)[1] - return content - except ValueError, e: - sys.stderr.write('%s\n'%e) - sys.exit(1) + if h: + try: + req = h.request(url) + except ValueError, e: + sys.stderr.write('%s\n'%e) + sys.exit(1) - try: - content = urllib2.urlopen(url).read() - return content - except (urllib2.URLError, OSError), 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: + ct = '' + encoding = '' + + else: + try: + req = urllib2.urlopen(url) + except urllib2.URLError, e: + sys.stderr.write('%s\n'%e) + sys.exit(1) - return '' + 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(): + +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(ical, rules=[], verbose=False): +def applyRules(cal, rules=[], verbose=False): 'Runs a series of rules on the lines in ical and mangles its output' for rule in rules: - output = [] - if rule.__doc__ and verbose: - print(rule.__doc__) - for line in ical: - try: - out = rule(line[0],line[1]) - except TypeError, e: - output.append(line) - print(e) - continue - - # Drop lines that are boolean False - if not out and not out == None: continue - - # If the rule did something and is a tuple or a list we'll accept it - # otherwise, pay no attention to the man behind the curtain - try: - if tuple(out) == out or list(out) == out and len(out) == 2: - output.append(tuple(out)) - else: - output.append(line) - except TypeError, e: - output.append(line) - - ical = output - - return ical + cal = rule(cal) + return cal def writeOutput(cal, outfile=''): '''Takes a list of lines and outputs to the specified file''' @@ -234,18 +154,15 @@ def writeOutput(cal, outfile=''): sys.stderr.write('%s\n'%e) sys.exit(1) - if cal[-1]: cal.append('') - - out.write('\r\n'.join(cal)) + cal.serialize(out) if not out == sys.stdout: out.close() - if __name__ == '__main__': + # Only load options parsing if this script was called directly, skip it + # if it's being called as a module. from optparse import OptionParser - # If the user passed us a 'stdin' argument, we'll go with that, - # otherwise we'll try for a url opener parser = OptionParser('usage: %prog [options] url') parser.add_option('-s', '--stdin', action='store_true', dest='stdin', @@ -254,9 +171,20 @@ if __name__ == '__main__': 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='Australia/Perth', + help='Specify a timezone to use if the remote calendar doesn\'t set it properly') (options, args) = parser.parse_args() + # Ensure the rules process using the desired timezone + ruleConfig = {} + ruleConfig["defaultTZ"] = options.timezone + + # 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) @@ -265,8 +193,10 @@ if __name__ == '__main__': else: url = '' - content = getContent(url, options.stdin) - cal = lineJoiner(content) - ical = applyRules(splitFields(cal), generateRules(), options.verbose) - output = lineFolder(joinFields(ical)) - writeOutput(output, options.outfile) + (content, encoding) = getContent(url, options.stdin) + encoding = encoding or options.encoding or 'utf-8' + + cal = vobject.readOne(unicode(content, encoding)) + cal = applyRules(cal, generateRules(ruleConfig), options.verbose) + + writeOutput(cal, options.outfile)