-# Copyright (C) 2002-2007 Python Software Foundation\r
-# Contact: email-sig@python.org\r
-\r
-"""Email address parsing code.\r
-\r
-Lifted directly from rfc822.py. This should eventually be rewritten.\r
-"""\r
-\r
-__all__ = [\r
- 'mktime_tz',\r
- 'parsedate',\r
- 'parsedate_tz',\r
- 'quote',\r
- ]\r
-\r
-import time\r
-\r
-SPACE = ' '\r
-EMPTYSTRING = ''\r
-COMMASPACE = ', '\r
-\r
-# Parse a date field\r
-_monthnames = ['jan', 'feb', 'mar', 'apr', 'may', 'jun', 'jul',\r
- 'aug', 'sep', 'oct', 'nov', 'dec',\r
- 'january', 'february', 'march', 'april', 'may', 'june', 'july',\r
- 'august', 'september', 'october', 'november', 'december']\r
-\r
-_daynames = ['mon', 'tue', 'wed', 'thu', 'fri', 'sat', 'sun']\r
-\r
-# The timezone table does not include the military time zones defined\r
-# in RFC822, other than Z. According to RFC1123, the description in\r
-# RFC822 gets the signs wrong, so we can't rely on any such time\r
-# zones. RFC1123 recommends that numeric timezone indicators be used\r
-# instead of timezone names.\r
-\r
-_timezones = {'UT':0, 'UTC':0, 'GMT':0, 'Z':0,\r
- 'AST': -400, 'ADT': -300, # Atlantic (used in Canada)\r
- 'EST': -500, 'EDT': -400, # Eastern\r
- 'CST': -600, 'CDT': -500, # Central\r
- 'MST': -700, 'MDT': -600, # Mountain\r
- 'PST': -800, 'PDT': -700 # Pacific\r
- }\r
-\r
-\r
-def parsedate_tz(data):\r
- """Convert a date string to a time tuple.\r
-\r
- Accounts for military timezones.\r
- """\r
- data = data.split()\r
- # The FWS after the comma after the day-of-week is optional, so search and\r
- # adjust for this.\r
- if data[0].endswith(',') or data[0].lower() in _daynames:\r
- # There's a dayname here. Skip it\r
- del data[0]\r
- else:\r
- i = data[0].rfind(',')\r
- if i >= 0:\r
- data[0] = data[0][i+1:]\r
- if len(data) == 3: # RFC 850 date, deprecated\r
- stuff = data[0].split('-')\r
- if len(stuff) == 3:\r
- data = stuff + data[1:]\r
- if len(data) == 4:\r
- s = data[3]\r
- i = s.find('+')\r
- if i > 0:\r
- data[3:] = [s[:i], s[i+1:]]\r
- else:\r
- data.append('') # Dummy tz\r
- if len(data) < 5:\r
- return None\r
- data = data[:5]\r
- [dd, mm, yy, tm, tz] = data\r
- mm = mm.lower()\r
- if mm not in _monthnames:\r
- dd, mm = mm, dd.lower()\r
- if mm not in _monthnames:\r
- return None\r
- mm = _monthnames.index(mm) + 1\r
- if mm > 12:\r
- mm -= 12\r
- if dd[-1] == ',':\r
- dd = dd[:-1]\r
- i = yy.find(':')\r
- if i > 0:\r
- yy, tm = tm, yy\r
- if yy[-1] == ',':\r
- yy = yy[:-1]\r
- if not yy[0].isdigit():\r
- yy, tz = tz, yy\r
- if tm[-1] == ',':\r
- tm = tm[:-1]\r
- tm = tm.split(':')\r
- if len(tm) == 2:\r
- [thh, tmm] = tm\r
- tss = '0'\r
- elif len(tm) == 3:\r
- [thh, tmm, tss] = tm\r
- else:\r
- return None\r
- try:\r
- yy = int(yy)\r
- dd = int(dd)\r
- thh = int(thh)\r
- tmm = int(tmm)\r
- tss = int(tss)\r
- except ValueError:\r
- return None\r
- # Check for a yy specified in two-digit format, then convert it to the\r
- # appropriate four-digit format, according to the POSIX standard. RFC 822\r
- # calls for a two-digit yy, but RFC 2822 (which obsoletes RFC 822)\r
- # mandates a 4-digit yy. For more information, see the documentation for\r
- # the time module.\r
- if yy < 100:\r
- # The year is between 1969 and 1999 (inclusive).\r
- if yy > 68:\r
- yy += 1900\r
- # The year is between 2000 and 2068 (inclusive).\r
- else:\r
- yy += 2000\r
- tzoffset = None\r
- tz = tz.upper()\r
- if tz in _timezones:\r
- tzoffset = _timezones[tz]\r
- else:\r
- try:\r
- tzoffset = int(tz)\r
- except ValueError:\r
- pass\r
- # Convert a timezone offset into seconds ; -0500 -> -18000\r
- if tzoffset:\r
- if tzoffset < 0:\r
- tzsign = -1\r
- tzoffset = -tzoffset\r
- else:\r
- tzsign = 1\r
- tzoffset = tzsign * ( (tzoffset//100)*3600 + (tzoffset % 100)*60)\r
- # Daylight Saving Time flag is set to -1, since DST is unknown.\r
- return yy, mm, dd, thh, tmm, tss, 0, 1, -1, tzoffset\r
-\r
-\r
-def parsedate(data):\r
- """Convert a time string to a time tuple."""\r
- t = parsedate_tz(data)\r
- if isinstance(t, tuple):\r
- return t[:9]\r
- else:\r
- return t\r
-\r
-\r
-def mktime_tz(data):\r
- """Turn a 10-tuple as returned by parsedate_tz() into a UTC timestamp."""\r
- if data[9] is None:\r
- # No zone info, so localtime is better assumption than GMT\r
- return time.mktime(data[:8] + (-1,))\r
- else:\r
- t = time.mktime(data[:8] + (0,))\r
- return t - data[9] - time.timezone\r
-\r
-\r
-def quote(str):\r
- """Prepare string to be used in a quoted string.\r
-\r
- Turns backslash and double quote characters into quoted pairs. These\r
- are the only characters that need to be quoted inside a quoted string.\r
- Does not add the surrounding double quotes.\r
- """\r
- return str.replace('\\', '\\\\').replace('"', '\\"')\r
-\r
-\r
-class AddrlistClass:\r
- """Address parser class by Ben Escoto.\r
-\r
- To understand what this class does, it helps to have a copy of RFC 2822 in\r
- front of you.\r
-\r
- Note: this class interface is deprecated and may be removed in the future.\r
- Use rfc822.AddressList instead.\r
- """\r
-\r
- def __init__(self, field):\r
- """Initialize a new instance.\r
-\r
- `field' is an unparsed address header field, containing\r
- one or more addresses.\r
- """\r
- self.specials = '()<>@,:;.\"[]'\r
- self.pos = 0\r
- self.LWS = ' \t'\r
- self.CR = '\r\n'\r
- self.FWS = self.LWS + self.CR\r
- self.atomends = self.specials + self.LWS + self.CR\r
- # Note that RFC 2822 now specifies `.' as obs-phrase, meaning that it\r
- # is obsolete syntax. RFC 2822 requires that we recognize obsolete\r
- # syntax, so allow dots in phrases.\r
- self.phraseends = self.atomends.replace('.', '')\r
- self.field = field\r
- self.commentlist = []\r
-\r
- def gotonext(self):\r
- """Parse up to the start of the next address."""\r
- while self.pos < len(self.field):\r
- if self.field[self.pos] in self.LWS + '\n\r':\r
- self.pos += 1\r
- elif self.field[self.pos] == '(':\r
- self.commentlist.append(self.getcomment())\r
- else:\r
- break\r
-\r
- def getaddrlist(self):\r
- """Parse all addresses.\r
-\r
- Returns a list containing all of the addresses.\r
- """\r
- result = []\r
- while self.pos < len(self.field):\r
- ad = self.getaddress()\r
- if ad:\r
- result += ad\r
- else:\r
- result.append(('', ''))\r
- return result\r
-\r
- def getaddress(self):\r
- """Parse the next address."""\r
- self.commentlist = []\r
- self.gotonext()\r
-\r
- oldpos = self.pos\r
- oldcl = self.commentlist\r
- plist = self.getphraselist()\r
-\r
- self.gotonext()\r
- returnlist = []\r
-\r
- if self.pos >= len(self.field):\r
- # Bad email address technically, no domain.\r
- if plist:\r
- returnlist = [(SPACE.join(self.commentlist), plist[0])]\r
-\r
- elif self.field[self.pos] in '.@':\r
- # email address is just an addrspec\r
- # this isn't very efficient since we start over\r
- self.pos = oldpos\r
- self.commentlist = oldcl\r
- addrspec = self.getaddrspec()\r
- returnlist = [(SPACE.join(self.commentlist), addrspec)]\r
-\r
- elif self.field[self.pos] == ':':\r
- # address is a group\r
- returnlist = []\r
-\r
- fieldlen = len(self.field)\r
- self.pos += 1\r
- while self.pos < len(self.field):\r
- self.gotonext()\r
- if self.pos < fieldlen and self.field[self.pos] == ';':\r
- self.pos += 1\r
- break\r
- returnlist = returnlist + self.getaddress()\r
-\r
- elif self.field[self.pos] == '<':\r
- # Address is a phrase then a route addr\r
- routeaddr = self.getrouteaddr()\r
-\r
- if self.commentlist:\r
- returnlist = [(SPACE.join(plist) + ' (' +\r
- ' '.join(self.commentlist) + ')', routeaddr)]\r
- else:\r
- returnlist = [(SPACE.join(plist), routeaddr)]\r
-\r
- else:\r
- if plist:\r
- returnlist = [(SPACE.join(self.commentlist), plist[0])]\r
- elif self.field[self.pos] in self.specials:\r
- self.pos += 1\r
-\r
- self.gotonext()\r
- if self.pos < len(self.field) and self.field[self.pos] == ',':\r
- self.pos += 1\r
- return returnlist\r
-\r
- def getrouteaddr(self):\r
- """Parse a route address (Return-path value).\r
-\r
- This method just skips all the route stuff and returns the addrspec.\r
- """\r
- if self.field[self.pos] != '<':\r
- return\r
-\r
- expectroute = False\r
- self.pos += 1\r
- self.gotonext()\r
- adlist = ''\r
- while self.pos < len(self.field):\r
- if expectroute:\r
- self.getdomain()\r
- expectroute = False\r
- elif self.field[self.pos] == '>':\r
- self.pos += 1\r
- break\r
- elif self.field[self.pos] == '@':\r
- self.pos += 1\r
- expectroute = True\r
- elif self.field[self.pos] == ':':\r
- self.pos += 1\r
- else:\r
- adlist = self.getaddrspec()\r
- self.pos += 1\r
- break\r
- self.gotonext()\r
-\r
- return adlist\r
-\r
- def getaddrspec(self):\r
- """Parse an RFC 2822 addr-spec."""\r
- aslist = []\r
-\r
- self.gotonext()\r
- while self.pos < len(self.field):\r
- if self.field[self.pos] == '.':\r
- aslist.append('.')\r
- self.pos += 1\r
- elif self.field[self.pos] == '"':\r
- aslist.append('"%s"' % quote(self.getquote()))\r
- elif self.field[self.pos] in self.atomends:\r
- break\r
- else:\r
- aslist.append(self.getatom())\r
- self.gotonext()\r
-\r
- if self.pos >= len(self.field) or self.field[self.pos] != '@':\r
- return EMPTYSTRING.join(aslist)\r
-\r
- aslist.append('@')\r
- self.pos += 1\r
- self.gotonext()\r
- return EMPTYSTRING.join(aslist) + self.getdomain()\r
-\r
- def getdomain(self):\r
- """Get the complete domain name from an address."""\r
- sdlist = []\r
- while self.pos < len(self.field):\r
- if self.field[self.pos] in self.LWS:\r
- self.pos += 1\r
- elif self.field[self.pos] == '(':\r
- self.commentlist.append(self.getcomment())\r
- elif self.field[self.pos] == '[':\r
- sdlist.append(self.getdomainliteral())\r
- elif self.field[self.pos] == '.':\r
- self.pos += 1\r
- sdlist.append('.')\r
- elif self.field[self.pos] in self.atomends:\r
- break\r
- else:\r
- sdlist.append(self.getatom())\r
- return EMPTYSTRING.join(sdlist)\r
-\r
- def getdelimited(self, beginchar, endchars, allowcomments=True):\r
- """Parse a header fragment delimited by special characters.\r
-\r
- `beginchar' is the start character for the fragment.\r
- If self is not looking at an instance of `beginchar' then\r
- getdelimited returns the empty string.\r
-\r
- `endchars' is a sequence of allowable end-delimiting characters.\r
- Parsing stops when one of these is encountered.\r
-\r
- If `allowcomments' is non-zero, embedded RFC 2822 comments are allowed\r
- within the parsed fragment.\r
- """\r
- if self.field[self.pos] != beginchar:\r
- return ''\r
-\r
- slist = ['']\r
- quote = False\r
- self.pos += 1\r
- while self.pos < len(self.field):\r
- if quote:\r
- slist.append(self.field[self.pos])\r
- quote = False\r
- elif self.field[self.pos] in endchars:\r
- self.pos += 1\r
- break\r
- elif allowcomments and self.field[self.pos] == '(':\r
- slist.append(self.getcomment())\r
- continue # have already advanced pos from getcomment\r
- elif self.field[self.pos] == '\\':\r
- quote = True\r
- else:\r
- slist.append(self.field[self.pos])\r
- self.pos += 1\r
-\r
- return EMPTYSTRING.join(slist)\r
-\r
- def getquote(self):\r
- """Get a quote-delimited fragment from self's field."""\r
- return self.getdelimited('"', '"\r', False)\r
-\r
- def getcomment(self):\r
- """Get a parenthesis-delimited fragment from self's field."""\r
- return self.getdelimited('(', ')\r', True)\r
-\r
- def getdomainliteral(self):\r
- """Parse an RFC 2822 domain-literal."""\r
- return '[%s]' % self.getdelimited('[', ']\r', False)\r
-\r
- def getatom(self, atomends=None):\r
- """Parse an RFC 2822 atom.\r
-\r
- Optional atomends specifies a different set of end token delimiters\r
- (the default is to use self.atomends). This is used e.g. in\r
- getphraselist() since phrase endings must not include the `.' (which\r
- is legal in phrases)."""\r
- atomlist = ['']\r
- if atomends is None:\r
- atomends = self.atomends\r
-\r
- while self.pos < len(self.field):\r
- if self.field[self.pos] in atomends:\r
- break\r
- else:\r
- atomlist.append(self.field[self.pos])\r
- self.pos += 1\r
-\r
- return EMPTYSTRING.join(atomlist)\r
-\r
- def getphraselist(self):\r
- """Parse a sequence of RFC 2822 phrases.\r
-\r
- A phrase is a sequence of words, which are in turn either RFC 2822\r
- atoms or quoted-strings. Phrases are canonicalized by squeezing all\r
- runs of continuous whitespace into one space.\r
- """\r
- plist = []\r
-\r
- while self.pos < len(self.field):\r
- if self.field[self.pos] in self.FWS:\r
- self.pos += 1\r
- elif self.field[self.pos] == '"':\r
- plist.append(self.getquote())\r
- elif self.field[self.pos] == '(':\r
- self.commentlist.append(self.getcomment())\r
- elif self.field[self.pos] in self.phraseends:\r
- break\r
- else:\r
- plist.append(self.getatom(self.phraseends))\r
-\r
- return plist\r
-\r
-class AddressList(AddrlistClass):\r
- """An AddressList encapsulates a list of parsed RFC 2822 addresses."""\r
- def __init__(self, field):\r
- AddrlistClass.__init__(self, field)\r
- if field:\r
- self.addresslist = self.getaddrlist()\r
- else:\r
- self.addresslist = []\r
-\r
- def __len__(self):\r
- return len(self.addresslist)\r
-\r
- def __add__(self, other):\r
- # Set union\r
- newaddr = AddressList(None)\r
- newaddr.addresslist = self.addresslist[:]\r
- for x in other.addresslist:\r
- if not x in self.addresslist:\r
- newaddr.addresslist.append(x)\r
- return newaddr\r
-\r
- def __iadd__(self, other):\r
- # Set union, in-place\r
- for x in other.addresslist:\r
- if not x in self.addresslist:\r
- self.addresslist.append(x)\r
- return self\r
-\r
- def __sub__(self, other):\r
- # Set difference\r
- newaddr = AddressList(None)\r
- for x in self.addresslist:\r
- if not x in other.addresslist:\r
- newaddr.addresslist.append(x)\r
- return newaddr\r
-\r
- def __isub__(self, other):\r
- # Set difference, in-place\r
- for x in other.addresslist:\r
- if x in self.addresslist:\r
- self.addresslist.remove(x)\r
- return self\r
-\r
- def __getitem__(self, index):\r
- # Make indexing, slices, and 'in' work\r
- return self.addresslist[index]\r
projects / mirror_edk2.git / blobdiff