110.html

国外python经典教材,python爱好者的首选

 and <I><tt cLass="monofont">decode</tT></i>
 arguments define the encoding and decoding functions that are returned or accepted by the <tt CLASs="monofont">read()</tt> and <tT CLAss="monofont">write()</tt> methods, respectively; that is, data returned by <TT CLass="monofont">read()</tT>  is encoded according to <I><TT class="monofont">encode</tt></i>
 and data given to <tt class="monofont">write()</tt>  is decoded according to <i><tt ClaSs="monofont">decode</tt></I>
. <i><tt cLass="monofont">reader</TT></I>
 and <I><tt clASS="monofont">writer</Tt></i>
 are the <tt CLASs="monofont">StreamReader</tt> and <tT CLAss="monofont">StreamWriter</tt>  classes used to read and write the actual contents of the data stream. A <tt class="monofont">StreamRecoder</tt>  object provides the combined interface of <tt class="monofont">StreamReader</Tt>  and <tT claSs="monofont">StreamWriter</tt>.</p>

<P><tt cLASS="monofont">codecs</tt>  also defines the following byte-order marker constants that can be used to help interpret platform-specific files:</p>

<p><TABLe borDER="1" CellsPACIng="0" cellpadding="1" width="100%">
<coLgrOup sPan="2">
<tr>
<Th vaLIGN="top">
<font SIZE="2">
<p><b>Constant</b></p>
</FONT></th>
<th VALIgn="top">
<font size="2">
<p><b>Description</b></p>
</font></th>
</Tr>
<tR>
<td vAlign="top">
<Font SIZE="2">
<p><tt cLASS="monofont">BOM</tt></p>
</fONT></Td>
<td vALIGn="top">
<font size="2">
<p>Native byte-order marker for the machine</p>
</font></td>
</tr>
<tR>
<td ValiGn="top">
<fonT sizE="2">
<P><TT clasS="monofont">BOM_BE</TT></P>
</font></TD>
<TD valiGN="top">
<FOnt size="2">
<p>Big-endian byte-order marker</p>
</font></td>
</tr>
<tr>
<td ValIgn="top">
<fOnt siZe="2">
<p><tT CLAss="monofont">BOM_LE</tt></P>
</FONt></td>
<tD VALign="top">
<fONT Size="2">
<p>Little-endian byte-order marker</p>
</font></td>
</tr>
<tr>
<td valIgn="top">
<Font Size="2">
<p><Tt clASS="monofont">BOM32_BE</Tt></p>
</foNT></TD>
<td vaLIGN="top">
<font SIZE="2">
<p>32-bit big-endian marker</p>
</font></td>
</tr>
<tr>
<td valign="top">
<FonT sizE="2">
<p><tt cLass="monofont">BOM32_LE</TT></P>
</Font></tD>
<TD ValigN="top">
<FONt sizE="2">
<P>32-bit little-endian marker</P>
</Font></td>
</tr>
<tr>
<td valign="top">
<fonT siZe="2">
<p><tT clasS="monofont">BOM64_BE</tt></p>
</FONT></td>
<td VALIgn="top">
<foNT SIze="2">
<p>64-bit big-endian marker</p>
</FONT></td>
</tr>
<tr>
<td valign="top">
<font siZe="2">
<p><Tt clAss="monofont">BOM64_LE</tt></P>
</fonT></TD>
<Td valIGN="top">
<Font sIZE="2">
<P>64-bit little-endian marker</p>
</fonT></TD>
</Tr>
</colgroup>
</table></p>


<h5>Example</h5>
<p>The following example illustrates the implementation of a new encoding using simple exclusive-or (XOR) based encryption. This only works for 8-bit strings, but it could be extended to support Unicode:</p>

<pRe>

# xor.py: Simple encryption using XOR 
import codecs 

# Encoding/decoding function (works both ways) 
def xor_encode(input, errors = 'strict', key=0xff): 
    output = "".join([chr(ord(c) ^ key) for c in input]) 
    return (output,len(input)) 

# XOR Codec class 
class Codec(codecs.Codec): 
    key = 0xff 
    def encode(self,input, errors='strict'): 
        return xor_encode(input,errors,self.key) 
    def decode(self,input, errors='strict'): 
        return xor_encode(input,errors,self.key) 

# StreamWriter and StreamReader classes 
class StreamWriter(Codec,codecs.StreamWriter): 
    pass 

class StreamReader(Codec,codecs.StreamReader): 
    pass 

# Factory functions for creating StreamWriter and 
# StreamReader objects with a given key value. 

def xor_writer_factory(stream,errors,key=0xff): 
    s = StreamWriter(stream,errors) 
    s.key = key 
    return s; 

def xor_reader_factory(stream,errors,key=0xff): 
    r = StreamReader(stream,errors) 
    r.key = key 
    return r 

# Function registered with the codecs module.  Recognizes any 
# encoding of the form 'xor-hh' where hh is a hexadecimal number. 

def lookup(s): 
    if (s[:4] == 'xor-'): 
        key = int(s[4:],16) 
    # Create some functions with key set to desired value 
    e = lambda x,err='strict',key=key:xor_encode(x,err,key) 
    r = lambda x,err='strict',key=key:xor_reader_factory(x,err,key) 
    w = lambda x,err='strict',key=key:xor_writer_factory(x,err,key) 
    return (e,e,r,w) 

# Register with the codec module 
codecs.register(lookup) </pRe>

<p>Now, here抯 a short program that uses the encoding:</p>

<Pre>

import xor, codecs 
f = codecs.open("foo","w","xor-37") 
f.write("Hello World\n")        # Writes an "encrypted" version 
f.close() 

(enc,dec,r,w) = codecs.lookup("xor-ae") 
a = enc("Hello World") 
# a = ('\346\313\302\302\301\216\371\301\334\302\312', 11) </prE>


<h5>Notes</h5>
<uL>
<LI>
<P>Further use of the <tt clASS="monofont">codecs</Tt>  module is described in <a hrEF="89.html">Chapter 9</A>.</P>
</li>
<li>
<P>Most of the built-in encodings are provided to support Unicode string encoding. In this case, the encoding functions produce 8-bit strings and the decoding functions produce Unicode strings.</P>
</LI>
</ul>
<p>? <b>See Also</b> <a href="89.html">Chapter 9</a>.</p>

<a name="8"></a>
<h4><tT clAss="monofont">re</tT></h4>
<p>The <tt ClasS="monofont">re</TT>  module is used to perform regular-expression pattern matching and replacement in strings. Both ordinary and Unicode strings are supported. Regular-expression patterns are specified as strings containing a mix of text and special-character sequences. Since patterns often make extensive use of special characters and the backslash, they抮e usually written as 搑aw