urllib.r2py

def urllib_quote(string, safe="/"):
  """
  <Purpose>
    Encode a string such that it can be used safely in a URL or XML
    document.

  <Arguments>
    string:
           The string to urlencode.

    safe (optional):
           Specifies additional characters that should not be quoted --
           defaults to "/".

  <Exceptions>
    TypeError if the safe parameter isn't an enumerable.

  <Side Effects>
    None.

  <Returns>
    Urlencoded version of the passed string.
  """

  resultstr = ""

  # We go through each character in the string; if it's not in [0-9a-zA-Z]
  # we wrap it.

  safeset = set(safe)

  for char in string:
    asciicode = ord(char)
    if (asciicode >= ord("0") and asciicode <= ord("9")) or \
        (asciicode >= ord("A") and asciicode <= ord("Z")) or \
        (asciicode >= ord("a") and asciicode <= ord("z")) or \
        asciicode == ord("_") or asciicode == ord(".") or \
        asciicode == ord("-") or char in safeset:
      resultstr += char
    else:
      resultstr += "%%%02X" % asciicode

  return resultstr




def urllib_quote_plus(string, safe=""):
  """
  <Purpose>
    Encode a string to go in the query fragment of a URL.

  <Arguments>
    string:
           The string to urlencode.

    safe (optional):
           Specifies additional characters that should not be quoted --
           defaults to the empty string.

  <Exceptions>
    TypeError if the safe parameter isn't a string.

  <Side Effects>
    None.

  <Returns>
    Urlencoded version of the passed string.
  """

  return urllib_quote(string, safe + " ").replace(" ", "+")




def urllib_unquote(string):
  """
  <Purpose>
    Unquote a urlencoded string.

  <Arguments>
    string:
           The string to unquote.

  <Exceptions>
    ValueError thrown if the last wrapped octet isn't a valid wrapped octet
    (i.e. if the string ends in "%" or "%x" rather than "%xx". Also throws
    ValueError if the nibbles aren't valid hex digits.

  <Side Effects>
    None.

  <Returns>
    The decoded string.
  """

  resultstr = ""

  # We go through the string from end to beginning, looking for wrapped
  # octets. When one is found we add it (unwrapped) and the following
  # string to the resultant string, and shorten the original string.

  while True:
    lastpercentlocation = string.rfind("%")
    if lastpercentlocation < 0:
      break

    wrappedoctetstr = string[lastpercentlocation+1:lastpercentlocation+3]
    if len(wrappedoctetstr) != 2:
      raise ValueError("Quoted string is poorly formed")

    resultstr = \
        chr(int(wrappedoctetstr, 16)) + \
        string[lastpercentlocation+3:] + \
        resultstr
    string = string[:lastpercentlocation]

  resultstr = string + resultstr
  return resultstr




def urllib_unquote_plus(string):
  """
  <Purpose>
    Unquote the urlencoded query fragment of a URL.

  <Arguments>
    string:
           The string to unquote.

  <Exceptions>
    ValueError thrown if the last wrapped octet isn't a valid wrapped octet
    (i.e. if the string ends in "%" or "%x" rather than "%xx". Also throws
    ValueError if the nibbles aren't valid hex digits.

  <Side Effects>
    None.

  <Returns>
    The decoded string.
  """

  return urllib_unquote(string.replace("+", " "))




def urllib_quote_parameters(dictionary):
  """
  <Purpose>
    Encode a dictionary of (key, value) pairs into an HTTP query string or
    POST body (same form).

  <Arguments>
    dictionary:
           The dictionary to quote.

  <Exceptions>
    None.

  <Side Effects>
    None.

  <Returns>
    The quoted dictionary.
  """

  quoted_keyvals = []
  for key, val in dictionary.items():
    quoted_keyvals.append("%s=%s" % (urllib_quote(key), urllib_quote(val)))

  return "&".join(quoted_keyvals)




def urllib_unquote_parameters(string):
  """
  <Purpose>
    Decode a urlencoded query string or POST body.

  <Arguments>
    string:
           The string to decode.

  <Exceptions>
    ValueError if the string is poorly formed.

  <Side Effects>
    None.

  <Returns>
    A dictionary mapping keys to values.
  """

  keyvalpairs = string.split("&")
  res = {}

  for quotedkeyval in keyvalpairs:
    # Throw ValueError if there is more or less than one '='.
    quotedkey, quotedval = quotedkeyval.split("=")
    key = urllib_unquote_plus(quotedkey)
    val = urllib_unquote_plus(quotedval)
    res[key] = val

  return res