1# Copyright (C) 2001-2006 Python Software Foundation 2# Author: Barry Warsaw, Thomas Wouters, Anthony Baxter 3# Contact: email-sig@python.org 4 5"""A parser of RFC 2822 and MIME email messages.""" 6 7__all__ = ['Parser', 'HeaderParser'] 8 9import warnings 10from cStringIO import StringIO 11 12from email.feedparser import FeedParser 13from email.message import Message 14 15 16 17class Parser: 18 def __init__(self, *args, **kws): 19 """Parser of RFC 2822 and MIME email messages. 20 21 Creates an in-memory object tree representing the email message, which 22 can then be manipulated and turned over to a Generator to return the 23 textual representation of the message. 24 25 The string must be formatted as a block of RFC 2822 headers and header 26 continuation lines, optionally preceeded by a `Unix-from' header. The 27 header block is terminated either by the end of the string or by a 28 blank line. 29 30 _class is the class to instantiate for new message objects when they 31 must be created. This class must have a constructor that can take 32 zero arguments. Default is Message.Message. 33 """ 34 if len(args) >= 1: 35 if '_class' in kws: 36 raise TypeError("Multiple values for keyword arg '_class'") 37 kws['_class'] = args[0] 38 if len(args) == 2: 39 if 'strict' in kws: 40 raise TypeError("Multiple values for keyword arg 'strict'") 41 kws['strict'] = args[1] 42 if len(args) > 2: 43 raise TypeError('Too many arguments') 44 if '_class' in kws: 45 self._class = kws['_class'] 46 del kws['_class'] 47 else: 48 self._class = Message 49 if 'strict' in kws: 50 warnings.warn("'strict' argument is deprecated (and ignored)", 51 DeprecationWarning, 2) 52 del kws['strict'] 53 if kws: 54 raise TypeError('Unexpected keyword arguments') 55 56 def parse(self, fp, headersonly=False): 57 """Create a message structure from the data in a file. 58 59 Reads all the data from the file and returns the root of the message 60 structure. Optional headersonly is a flag specifying whether to stop 61 parsing after reading the headers or not. The default is False, 62 meaning it parses the entire contents of the file. 63 """ 64 feedparser = FeedParser(self._class) 65 if headersonly: 66 feedparser._set_headersonly() 67 while True: 68 data = fp.read(8192) 69 if not data: 70 break 71 feedparser.feed(data) 72 return feedparser.close() 73 74 def parsestr(self, text, headersonly=False): 75 """Create a message structure from a string. 76 77 Returns the root of the message structure. Optional headersonly is a 78 flag specifying whether to stop parsing after reading the headers or 79 not. The default is False, meaning it parses the entire contents of 80 the file. 81 """ 82 return self.parse(StringIO(text), headersonly=headersonly) 83 84 85 86class HeaderParser(Parser): 87 def parse(self, fp, headersonly=True): 88 return Parser.parse(self, fp, True) 89 90 def parsestr(self, text, headersonly=True): 91 return Parser.parsestr(self, text, True) 92