Source Code for Module gluon.decoder

 1  import codecs 
 2   
 3   
 4  """Caller will hand this library a buffer and ask it to either convert 
 5  it or auto-detect the type. 
 6   
 7  Based on http://code.activestate.com/recipes/52257/ 
 8   
 9  Licensed under the PSF License 
10  """ 
11   
12  # None represents a potentially variable byte. "##" in the XML spec... 
13  autodetect_dict = {  # bytepattern     : ("name", 
14                                           (0x00, 0x00, 0xFE, 0xFF): ("ucs4_be"), 
15      (0xFF, 0xFE, 0x00, 0x00): ("ucs4_le"), 
16      (0xFE, 0xFF, None, None): ("utf_16_be"), 
17      (0xFF, 0xFE, None, None): ("utf_16_le"), 
18      (0x00, 0x3C, 0x00, 0x3F): ("utf_16_be"), 
19      (0x3C, 0x00, 0x3F, 0x00): ("utf_16_le"), 
20      (0x3C, 0x3F, 0x78, 0x6D): ("utf_8"), 
21      (0x4C, 0x6F, 0xA7, 0x94): ("EBCDIC") 
22  } 
23   
24   
25 -def autoDetectXMLEncoding(buffer): 
26      """ buffer -> encoding_name 
27      The buffer should be at least 4 bytes long. 
28          Returns None if encoding cannot be detected. 
29          Note that encoding_name might not have an installed 
30          decoder (e.g. EBCDIC) 
31      """ 
32      # a more efficient implementation would not decode the whole 
33      # buffer at once but otherwise we'd have to decode a character at 
34      # a time looking for the quote character...that's a pain 
35   
36      encoding = "utf_8"  # according to the XML spec, this is the default 
37                            # this code successively tries to refine the default 
38                            # whenever it fails to refine, it falls back to 
39                            # the last place encoding was set. 
40      if len(buffer) >= 4: 
41          bytes = (byte1, byte2, byte3, byte4) = tuple(map(ord, buffer[0:4])) 
42          enc_info = autodetect_dict.get(bytes, None) 
43          if not enc_info:  # try autodetection again removing potentially 
44              # variable bytes 
45              bytes = (byte1, byte2, None, None) 
46              enc_info = autodetect_dict.get(bytes) 
47      else: 
48          enc_info = None 
49   
50      if enc_info: 
51          encoding = enc_info  # we've got a guess... these are 
52                              #the new defaults 
53   
54          # try to find a more precise encoding using xml declaration 
55          secret_decoder_ring = codecs.lookup(encoding)[1] 
56          (decoded, length) = secret_decoder_ring(buffer) 
57          first_line = decoded.split("\n")[0] 
58          if first_line and first_line.startswith(u"<?xml"): 
59              encoding_pos = first_line.find(u"encoding") 
60              if encoding_pos != -1: 
61                  # look for double quote 
62                  quote_pos = first_line.find('"', encoding_pos) 
63   
64                  if quote_pos == -1:                 # look for single quote 
65                      quote_pos = first_line.find("'", encoding_pos) 
66   
67                  if quote_pos > -1: 
68                      quote_char, rest = (first_line[quote_pos], 
69                                          first_line[quote_pos + 1:]) 
70                      encoding = rest[:rest.find(quote_char)] 
71   
72      return encoding 
73   
74   
75 -def decoder(buffer): 
76      encoding = autoDetectXMLEncoding(buffer) 
77      return buffer.decode(encoding).encode('utf8') 
78