This is an archive of the discontinued Mercurial Phabricator instance.

def addressencode(ui, address, charsets=None, display=False):
'''Turns address into RFC-2047 compliant header.'''
if display or not address:
    return address or ''
name, addr = email.utils.parseaddr(address)

+ name, addr = email.utils.parseaddr(pycompat.strurl(address))

strurl() is too strict to convert a user input. encoding.strfromlocal()
can be used instead.

result = []
for name, addr in email.utils.getaddresses(addrs):

+ for name, addr in email.utils.getaddresses(
+ [pycompat.strurl(a) for a in addrs]):

Same here.

durin42 updated this revision to Diff 10190.Aug 9 2018, 4:43 PM

durin42 updated this revision to Diff 10194.Aug 9 2018, 4:48 PM

Closed by commit rHGeabdf3c25b8b: mail: cope with Py3 unicode antics on email addresses (authored by durin42). · Explain WhyAug 9 2018, 11:00 PM

This revision was automatically updated to reflect the committed changes.

def _addressencode(ui, name, addr, charsets=None):
name = headencode(ui, name, charsets)
try:
acc, dom = addr.split('@')

+ acc, dom = addr.split(r'@')

I think addr here should be a byte string since dom is decoded to unicode,
and addr is embedded in abort message.

acc = acc.encode('ascii')
dom = dom.decode(encoding.encoding).encode('idna')
addr = '%s@%s' % (acc, dom)
@@ -311,27 +311,31 @@
    addr = addr.encode('ascii')
except UnicodeDecodeError:
    raise error.Abort(_('invalid local address: %s') % addr)
return email.utils.formataddr((name, addr))

+ return pycompat.bytesurl(
+ email.utils.formataddr((name, encoding.strfromlocal(addr))))

result = []
for name, addr in email.utils.getaddresses(addrs):

+ for name, addr in email.utils.getaddresses(
+ [encoding.strfromlocal(a) for a in addrs]):
if name or addr:
    result.append(_addressencode(ui, name, addr, charsets))
return result

+ return [pycompat.bytesurl(r) for r in result]

_addressencode() retunrs bytes.

Revision Contents
Changeset List

			Path	Packages
M			mercurial/mail.py (14 lines)

Diff	ID	Description	Created	Lint	Unit
Base		Base
Diff 1	9610		Jul 16 2018, 7:16 PM	★	★
Diff 2	10190		Aug 9 2018, 4:43 PM	★	★
Diff 3	10194		Aug 9 2018, 4:48 PM	★	★
Diff 4	10246	rHGeabdf3c25b8b0ddaa3f13813fae4ed7a0c2d5c62	Jul 16 2018, 5:49 PM	★	★

Status	Author	Revision
Closed	durin42	D4234 tests: allow for a bonus newline in base64'd email payload
Closed	durin42	D4233 mail: be more cautious about bytes vs str for py3 compat
Closed	durin42	D4232 patchbomb: use native strings when determining attachment disposition
Closed	durin42	D4231 mail: always fall back to iso-8859-1 if us-ascii won't work (BC)
Closed	durin42	D4230 tests: put some Python 3 polish on inline Python invocations
Closed	durin42	D4229 patchbomb: don't unintentionally duplicate headers
Closed	durin42	D4228 patchbomb: use sysstrs when describing content-type
Closed	durin42	D4223 patchbomb: use native strs for email header keys and values
Closed	durin42	D4222 patchbomb: use email.encoders instead of email.Encoders
Closed	durin42	D4221 mail: properly handle email addresses typically being unicodes
Closed	durin42	D4220 tests: fix Python3 issues in Python one-liners in test-patchbomb.t
Closed	durin42	D4219 dummysmtpd: accept additional kwargs from stdlib smtpd
Closed	durin42	D4218 mail: fix debug print, which appears to have been broken for a long time
Closed	durin42	D3957 patchbomb: python 3 really wants those email addresses in unicode
Closed	durin42	D3956 mail: stop using the smtplib.SSLFakeFile and use socket.socket.makefile
Closed	durin42	D3955 mail: modernize check for Python-with-TLS
Closed	durin42	D3954 mail: cope with Py3 unicode antics on email addresses
Closed	durin42	D3953 mail: fix _encode to be more correct on Python 3
Closed	durin42	D3951 patchbomb: work around email module really wanting to write unicode data

Diff 10246

mercurial/mail.py

	# split into words?			# split into words?
	s, cs = _encode(ui, s, charsets)			s, cs = _encode(ui, s, charsets)
	return str(email.header.Header(s, cs))			return str(email.header.Header(s, cs))
	return s			return s

	def _addressencode(ui, name, addr, charsets=None):			def _addressencode(ui, name, addr, charsets=None):
	name = headencode(ui, name, charsets)			name = headencode(ui, name, charsets)
	try:			try:
	acc, dom = addr.split('@')			acc, dom = addr.split(r'@')
	acc = acc.encode('ascii')			acc = acc.encode('ascii')
	dom = dom.decode(encoding.encoding).encode('idna')			dom = dom.decode(encoding.encoding).encode('idna')
	addr = '%s@%s' % (acc, dom)			addr = '%s@%s' % (acc, dom)
	except UnicodeDecodeError:			except UnicodeDecodeError:
	raise error.Abort(_('invalid email address: %s') % addr)			raise error.Abort(_('invalid email address: %s') % addr)
	except ValueError:			except ValueError:
	try:			try:
	# too strict?			# too strict?
	addr = addr.encode('ascii')			addr = addr.encode('ascii')
	except UnicodeDecodeError:			except UnicodeDecodeError:
	raise error.Abort(_('invalid local address: %s') % addr)			raise error.Abort(_('invalid local address: %s') % addr)
	return email.utils.formataddr((name, addr))			return pycompat.bytesurl(
				email.utils.formataddr((name, encoding.strfromlocal(addr))))

	def addressencode(ui, address, charsets=None, display=False):			def addressencode(ui, address, charsets=None, display=False):
	'''Turns address into RFC-2047 compliant header.'''			'''Turns address into RFC-2047 compliant header.'''
	if display or not address:			if display or not address:
	return address or ''			return address or ''
	name, addr = email.utils.parseaddr(address)			name, addr = email.utils.parseaddr(encoding.strfromlocal(address))
	return _addressencode(ui, name, addr, charsets)			return _addressencode(ui, name, addr, charsets)

	def addrlistencode(ui, addrs, charsets=None, display=False):			def addrlistencode(ui, addrs, charsets=None, display=False):
	'''Turns a list of addresses into a list of RFC-2047 compliant headers.			'''Turns a list of addresses into a list of RFC-2047 compliant headers.
	A single element of input list may contain multiple addresses, but output			A single element of input list may contain multiple addresses, but output
	always has one address per item'''			always has one address per item'''
				for a in addrs:
				assert isinstance(a, bytes), (r'%r unexpectedly not a bytestr' % a)
	if display:			if display:
	return [a.strip() for a in addrs if a.strip()]			return [a.strip() for a in addrs if a.strip()]

	result = []			result = []
	for name, addr in email.utils.getaddresses(addrs):			for name, addr in email.utils.getaddresses(
				[encoding.strfromlocal(a) for a in addrs]):
	if name or addr:			if name or addr:
	result.append(_addressencode(ui, name, addr, charsets))			result.append(_addressencode(ui, name, addr, charsets))
	return result			return [pycompat.bytesurl(r) for r in result]

	def mimeencode(ui, s, charsets=None, display=False):			def mimeencode(ui, s, charsets=None, display=False):
	'''creates mime text object, encodes it if needed, and sets			'''creates mime text object, encodes it if needed, and sets
	charset and transfer-encoding accordingly.'''			charset and transfer-encoding accordingly.'''
	cs = 'us-ascii'			cs = 'us-ascii'
	if not display:			if not display:
	s, cs = _encode(ui, s, charsets)			s, cs = _encode(ui, s, charsets)
	return mimetextqp(s, 'plain', cs)			return mimetextqp(s, 'plain', cs)

This is an archive of the discontinued Mercurial Phabricator instance.

mail: cope with Py3 unicode antics on email addressesClosedPublic

Details

Diff Detail

Event Timeline

Revision ContentsChangeset List

Diff 10246

mercurial/mail.py

mail: cope with Py3 unicode antics on email addresses
ClosedPublic

Revision Contents
Changeset List