Handles MIME encodings from FTP now too

[frenchie/icalparse.git] / icalparse.py

diff --git a/icalparse.py b/icalparse.py
index 193a00a..293295c 100755 (executable)
--- a/icalparse.py
+++ b/icalparse.py
@@ -30,6 +30,16 @@ class IncompleteICS(InvalidICS): pass
 
 def lineJoiner(oldcal, encoding='utf-8'):
        '''Takes a string containing a calendar and returns an array of its lines'''
+       try:
+               oldcal = unicode(oldcal, encoding)
+               oldcal = oldcal.replace('\r\n ', '').replace('\r\n\t','')
+       except UnicodeDecodeError:
+               # This is probably a file with badly folded lines
+               oldcal = oldcal.replace('\r\n ', '').replace('\r\n\t','')
+               try: oldcal = unicode(oldcal, encoding)
+               except UnicodeDecodeError:
+                       sys.stderr.write('Malformed File')
+                       raise
 
        if not oldcal[0:15] == 'BEGIN:VCALENDAR':
                raise InvalidICS, "Does not appear to be a valid ICS file"
@@ -40,8 +50,7 @@ def lineJoiner(oldcal, encoding='utf-8'):
        if list(oldcal) == oldcal:
                oldcal = '\r\n'.join(oldcal)
 
-       oldcal = oldcal.replace('\r\n ', '').replace('\r\n\t','')
-       return [unicode(x, encoding) for x in oldcal.strip().split('\r\n')]
+       return oldcal.split('\r\n')
 
 
 def lineFolder(oldcal, length=75):
@@ -81,12 +90,14 @@ 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]
+       ical = []
 
        # 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)
+       for line in [tuple(x.split(':',1)) for x in cal]:
+               if not len(line) == 2 and line[0]:
+                       raise InvalidICS, 'Unusual content line: %s'%line
+               elif line[0]:
+                       ical.append(line)
 
        return ical
 
@@ -114,22 +125,21 @@ def getContent(url='',stdin=False):
                content = sys.stdin.read()
                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, 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()
+               if 'content-type' in res.info(): ct = res.info()['content-type']
+               else: ct = ''
                res.close()
        except (urllib2.URLError, OSError), e:
                sys.stderr.write('%s\n'%e)
                sys.exit(1)
+
+       encoding = 'charset' in ct and ct.split(';')[-1].lower().split('=')[-1].strip() or ''
        return (content, encoding)
 
 
@@ -142,38 +152,39 @@ def getHTTPContent(url='',cache='.httplib2-cache'):
        except ImportError:
                import urllib2
 
-       if not url: return ''
-
-       encoding = '' # If we don't populate this, the script will assume UTF-8
+       if not url: return ('','')
+       if not 'http' in parsedURL[0]: return ('','')
 
        if 'httplib2' in sys.modules:
                try: h = httplib2.Http('.httplib2-cache')
                except OSError: h = httplib2.Http()
        else: h = False
 
-       try:
-               if h:
+       if h:
+               try:
                        req = h.request(url)
-                       content = req[1]
-                       if 'content-type' in req[0]:
-                               for ct in req[0]['content-type'].split(';'):
-                                       ct = ct.lower()
-                                       print ct
-                                       if 'charset' in ct:
-                                               encoding = ct.split('=')[1]
-               return (content, encoding)
-       except ValueError, e:
-               sys.stderr.write('%s\n'%e)
-               sys.exit(1)
+               except ValueError, e:
+                       sys.stderr.write('%s\n'%e)
+                       sys.exit(1)
 
-       try:
-               content = urllib2.urlopen(url).read()
-               return (content, encoding)
-       except (urllib2.URLError, OSError), e:
-               sys.stderr.write('%s\n'%e)
-               sys.exit(1)
+               content = req[1]
+               if 'content-type' in req[0]: ct = req[0]['content-type']
+
+       else:
+               try:
+                       req = urllib2.urlopen(url)
+               except urllib2.URLError, e:
+                       sys.stderr.write('%s\n'%e)
+                       sys.exit(1)
+
+               content = req.read()
+               info = req.info()
 
-       return ('', '')
+               ct = info['content-type']
+
+       encoding = 'charset' in ct and ct.split(';')[-1].lower().split('=')[-1].strip() or ''
+
+       return (content, encoding)
 
 
 def generateRules():