• Home
  • Line#
  • Scopes#
  • Navigate#
  • Raw
  • Download
1# Copyright (C) 2001-2007 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', 'BytesParser', 'BytesHeaderParser',
8           'FeedParser', 'BytesFeedParser']
9
10from io import StringIO, TextIOWrapper
11
12from email.feedparser import FeedParser, BytesFeedParser
13from email._policybase import compat32
14
15
16class Parser:
17    def __init__(self, _class=None, *, policy=compat32):
18        """Parser of RFC 2822 and MIME email messages.
19
20        Creates an in-memory object tree representing the email message, which
21        can then be manipulated and turned over to a Generator to return the
22        textual representation of the message.
23
24        The string must be formatted as a block of RFC 2822 headers and header
25        continuation lines, optionally preceded by a `Unix-from' header.  The
26        header block is terminated either by the end of the string or by a
27        blank line.
28
29        _class is the class to instantiate for new message objects when they
30        must be created.  This class must have a constructor that can take
31        zero arguments.  Default is Message.Message.
32
33        The policy keyword specifies a policy object that controls a number of
34        aspects of the parser's operation.  The default policy maintains
35        backward compatibility.
36
37        """
38        self._class = _class
39        self.policy = policy
40
41    def parse(self, fp, headersonly=False):
42        """Create a message structure from the data in a file.
43
44        Reads all the data from the file and returns the root of the message
45        structure.  Optional headersonly is a flag specifying whether to stop
46        parsing after reading the headers or not.  The default is False,
47        meaning it parses the entire contents of the file.
48        """
49        feedparser = FeedParser(self._class, policy=self.policy)
50        if headersonly:
51            feedparser._set_headersonly()
52        while data := fp.read(8192):
53            feedparser.feed(data)
54        return feedparser.close()
55
56    def parsestr(self, text, headersonly=False):
57        """Create a message structure from a string.
58
59        Returns the root of the message structure.  Optional headersonly is a
60        flag specifying whether to stop parsing after reading the headers or
61        not.  The default is False, meaning it parses the entire contents of
62        the file.
63        """
64        return self.parse(StringIO(text), headersonly=headersonly)
65
66
67class HeaderParser(Parser):
68    def parse(self, fp, headersonly=True):
69        return Parser.parse(self, fp, True)
70
71    def parsestr(self, text, headersonly=True):
72        return Parser.parsestr(self, text, True)
73
74
75class BytesParser:
76
77    def __init__(self, *args, **kw):
78        """Parser of binary RFC 2822 and MIME email messages.
79
80        Creates an in-memory object tree representing the email message, which
81        can then be manipulated and turned over to a Generator to return the
82        textual representation of the message.
83
84        The input must be formatted as a block of RFC 2822 headers and header
85        continuation lines, optionally preceded by a `Unix-from' header.  The
86        header block is terminated either by the end of the input or by a
87        blank line.
88
89        _class is the class to instantiate for new message objects when they
90        must be created.  This class must have a constructor that can take
91        zero arguments.  Default is Message.Message.
92        """
93        self.parser = Parser(*args, **kw)
94
95    def parse(self, fp, headersonly=False):
96        """Create a message structure from the data in a binary file.
97
98        Reads all the data from the file and returns the root of the message
99        structure.  Optional headersonly is a flag specifying whether to stop
100        parsing after reading the headers or not.  The default is False,
101        meaning it parses the entire contents of the file.
102        """
103        fp = TextIOWrapper(fp, encoding='ascii', errors='surrogateescape')
104        try:
105            return self.parser.parse(fp, headersonly)
106        finally:
107            fp.detach()
108
109
110    def parsebytes(self, text, headersonly=False):
111        """Create a message structure from a byte string.
112
113        Returns the root of the message structure.  Optional headersonly is a
114        flag specifying whether to stop parsing after reading the headers or
115        not.  The default is False, meaning it parses the entire contents of
116        the file.
117        """
118        text = text.decode('ASCII', errors='surrogateescape')
119        return self.parser.parsestr(text, headersonly)
120
121
122class BytesHeaderParser(BytesParser):
123    def parse(self, fp, headersonly=True):
124        return BytesParser.parse(self, fp, headersonly=True)
125
126    def parsebytes(self, text, headersonly=True):
127        return BytesParser.parsebytes(self, text, headersonly=True)
128