X-Git-Url: https://git.ucc.asn.au/?a=blobdiff_plain;f=icalparse.py;h=edfca4c6ee7a70a4ae6c16e8124a6b5b4cd2270c;hb=c973d20ef57023dc0994d7ffeb0a6e1fd6f9eb0f;hp=110a19ae703d822afa6fd918672bcf69c158543e;hpb=aace9ea8b2dbcdbdc1b3ed30a672a52b0a619e3f;p=frenchie%2Ficalparse.git diff --git a/icalparse.py b/icalparse.py index 110a19a..edfca4c 100755 --- a/icalparse.py +++ b/icalparse.py @@ -24,8 +24,9 @@ import sys import urlparse import os + class InvalidICS(Exception): pass -class notJoined(Exception): pass +class IncompleteICS(InvalidICS): pass def lineJoiner(oldcal): '''Takes a string containing a calendar and returns an array of its lines''' @@ -33,11 +34,14 @@ def lineJoiner(oldcal): 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.replace('\r\n ', '') - return oldcal.split('\r\n') + 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): @@ -50,19 +54,43 @@ def lineFolder(oldcal, length=75): sl = length - 1 for line in oldcal: + line = line.encode('utf-8') # Line fits inside length, do nothing if len(line.rstrip()) <= length: cal.append(line) else: - brokenline = [line[0:length] + '\r\n'] + brokenline = [line[0:length]] ll = length - while ll < len(line.rstrip('\r\n')) + 1: - brokenline.append(' ' + line[ll:sl+ll].rstrip('\r\n') + '\r\n') + while ll < len(line) + 1: + brokenline.append(line[ll:sl+ll]) ll += sl - cal += brokenline + brokenline = '\r\n '.join(brokenline) + cal.append(brokenline) 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] + + 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). @@ -90,7 +118,7 @@ def getContent(url='',stdin=False): res = urllib2.urlopen(url) content = res.read() res.close() - except (urllib2.URLError, ValueError), e: + except (urllib2.URLError, OSError), e: sys.stderr.write('%s\n'%e) sys.exit(1) return content @@ -98,9 +126,7 @@ def getContent(url='',stdin=False): 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. It will (by default) - slurp the first 20 bytes of the file and check that we are indeed looking - at an ICS file before going for broke.''' + services. It's what you should use in a CGI script.''' try: import httplib2 @@ -124,12 +150,81 @@ def getHTTPContent(url='',cache='.httplib2-cache'): try: content = urllib2.urlopen(url).read() return content - except urllib2.URLError, e: + except (urllib2.URLError, OSError), e: sys.stderr.write('%s\n'%e) sys.exit(1) return '' + +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(ical, 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 + + +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) + + if cal[-1]: cal.append('') + + out.write('\r\n'.join(cal)) + + if not out == sys.stdout: + out.close() + + if __name__ == '__main__': from optparse import OptionParser # If the user passed us a 'stdin' argument, we'll go with that, @@ -138,6 +233,8 @@ 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)') @@ -153,4 +250,6 @@ if __name__ == '__main__': content = getContent(url, options.stdin) cal = lineJoiner(content) - print cal + ical = applyRules(splitFields(cal), generateRules(), options.verbose) + output = lineFolder(joinFields(ical)) + writeOutput(output, options.outfile)